Software Engineering Glossary

.: special directory in Unix-based operating systems that represents the current directory
..: special directory in Unix-based operating systems that represents the parent directory
.htaccess: configuration file for Apache

.NET: open source managed software framework developed by Microsoft based on C#

2D: space with two dimensions

3D: space with three dimensions

3D printing: additive manufacturing process of making 3D objects from a digital blueprint

7z: file format for compression, also a program implementing it and command for it

A record: DNS record that maps a domain name to an IPv4 address

AAAA record: DNS record that maps a domain name to an IPv6 address

AAI: abbreviation for authentication, authorization and identity

ABC, abstract base class: abstract base classes module in Python which decorates methods as abstract and then registers concrete classes as implementations of the abstract base

abstract class: class containing at least one abstract method
abstract data type, ADT: mathematical model for data types defined by its behaviour

abstract method: method that has a declaration but does not have an implementation

abstract syntax tree, AST: tree representation of the syntax of a code

acceptance criteria: set of conditions that a software must satisfy to be accepted by a stakeholder or a customer

access control: security technique that restricts access to a resource based on authentication and authorization

accessibility: property of a software that makes it usable by people with disabilities or special needs

accuracy: metric for machine learning and information retrieval

ACID: set of properties of database transactions; see atomicity, consistency, isolation, durability

ack: command for searching in text files with regular expressions

activation (function): function that transforms the output of a neuron into a value that is then passed to the next layer; e.g. ReLU, sigmoid, tanh
Ada: statically typed object-oriented high-level programming language developed by the US Department of Defense for embedded systems

Adam optimizer: optimizer that uses weight decay and momentum from SGD optimizer
AdamW optimizer: optimizer that uses weight decay and momentum from Adam optimizer

Adobe: company that develops software for creative professionals

Advent of Code, AOC, AoC: yearly programming challenge
affine function: composition of a linear function followed by a translation

aggregation: function which combines several items (e.g. rows) into one value; e.g. average, count, min, max, median, sum, range etc.

agile: flexible and iterative approach to project management from 90s that prioritizes adaptability
Aho-Corasick algorithm: algorithm for finding multiple search patterns in a text
AI alignment: subfield of AI that studies how to align artificial intelligence with human values

AI, artificial intelligence: field of computer science that studies how to make computers do things that people are (or have been) better at
Airbyte: software for connecting a source of data with a consumer of the data
Airflow: open source software developed by Apache written in Python for workflow management

AJAX, asynchronous JavaScript and XML: technique used in web development to create interactive and dynamic web applications based on JavaScript and XML

Alexa: virtual assistant developed by Amazon
Alexa: voice assistant developed by Amazon
AlexNet: convolutional network model architecture that won the ImageNet competition in 2012

ALGOL: high-level programming language developed in 1958

algorithm: finite sequence of instructions for solving a problem or doing a computation

alpha channel: layer of transparency in a bitmap image
AlphaZero: AI algorithm that learned to play chess, shogi and go by playing against itself

ALSA, advanced Linux sound architecture: software framework and part of the Linux kernel for providing an API for sound card device drivers

Alt key: modifier key on a keyboard that modifies the function of other keys

Amazon: company that owns AWS and Alexa and Kindle and IMDb and Twitch and Goodreads
Anaconda: distribution of the Python and R programming languages for scientific computing

anchor: HTML element in a hypertext document that links to another document or a specific part of it

Android: mobile operating system developed by Google based on Linux
Android Studio: IDE for developing Android apps developed by Google based on IntelliJ
Angular: TypeScript web framework developed by Google
AngularJS: discontinued (as of 2023) JavaScript web framework developed by Google

anonymous function: function that does not have a name
Ansible: open source CLI automation software developed by Red Hat written in Python; it can configure systems, deploy software, and orchestrate advanced workflows to support application deployment, system updates, and more using ssh execution from controlled to managed nodes

Apache software foundation, ASF: software foundation that develops open source software

Apache (HTTP server): HTTP server developed by Red Hat

API, application programming interface: set of functions and procedures that allow the creation of applications which access the features or data of an operating system, application, or other service
APK, Android application package: file format for Android apps

APL, a programming language: programming language developed in the 60s for mathematical notation

append: operation which adds items to the end of a list or file

Apple: technology company that specializes in consumer electronics, proprietary software, and online services; e.g. macOS, Xcode. Swift, App Store
AppleScript: scripting language developed by Apple for macOS
applet: small application

application, app: see software

AppStore: software distribution platform for iOS apps

Arch Linux: Linux distribution that follows the KISS principle

architecture (machine learning: functional form or type of a model; sometimes incorrectly used as a synonym of the model itself

archive: file containing one or more files and/or directories
Ardour: open source DAW
Arduino: microcontroller board
argmax: operation that finds the argument that gives the maximum value from a target function

argument: value passed to a function when calling it; it can be passed by value or by reference

arithmetic mean: sum of a collection of numbers divided by the count of numbers in the collection

ARM, Advanced RISC Machine: processor architecture using a RISC instruction set

array: data structure for storing a sequence of simpler (usually homogenous) data types

array programming language: programming language operating mainly on arrays; e.g. APL, J, K, NumPy
Arrow: framework for tabular in-memory analytics developed by Apache
Asana: software for project management

ASCII, American Standard Code for Information Interchange: encoding standard for plain text

ASGI, asynchronous server gateway interface: protocol for web servers written in Python

ASP.NET: web framework developed by Microsoft for C# and .NET
assembler: software for converting assembly into machine code

assembly: low-level programming language that is a human-readable representation of machine code

assertion: statement in a programming language that is assumed to be true used for debugging

asymmetric cryptography: cryptography system that uses different keys for encryption and decryption

asynchronous programming: programming paradigm that allows the program to concurrently execute multiple tasks without waiting for the completion of the previous task

Atlassian: company that develops software for software development teams

atomicity: property of a transaction that guarantees that either all or none of the operations are performed

attention: mechanism in neural networks that allows the network to focus on specific parts of the input
Attention is all you need: paper that introduced the transformer architecture

attribute: property of an object
audio embedding: embedding of an audio signal

authentication: process of verifying the identity of a user or a system
authoritative name server: server that stores the DNS records for a domain

authorization: process of verifying that a user or a system has access to a resource
autocomplete: feature that suggests possible completions for a word or phrase as the user is typing

automaton: mathematical model of a machine that can perform computations
autoregressive model: model which uses past predictions for predicting the value

AVI, audio video interleave: file format for video and audio

AVL tree, Adelson-Velskii Landis tree: self-balancing binary search tree with height difference of at most 1 between left and right subtrees
Awesome WM: tiling window manager

AWK, awk, Aho–Weinberger–Kernighan: programming language for processing files in text format

AWS, Amazon Web Services: cloud computing platform from Amazon

axiom: a formal statement that is assumed to be true
Azure: IaaS from Microsoft

B2B, business to business: providing services, products to other businesses

B2C, business to costumer: business model where a company sells products or services directly to consumers
Babel: transpilation library for JavaScript
back button: GUI button in a web browser that allows the user to go back to the previously visited page
back propagation: updating gradients for all elements of the functions of the model
backdoor: attack TODO

backend, be, back-end: part of a software that is not directly accessed by the user and is usually running on a server
backport: TODO
Backus-Naur form: notation for describing the formal grammar of a programming language

backward compatibility: property of a software that allows it to work with older versions of itself
bar chart: chart that uses bars to show the frequency of categorical data
base64: binary-to-text encoding that represents binary data in an ASCII string format

Bash, Bourne Again Shell: shell for Linux-based operating systems developed by GNU

BASIC: programming language from 60s designed for students to learn programming

batch: random portion of training data used for forward pass (every time different/random) to save compute time; too small batches may introduce noise in loss values over time
batch normalization: technique for normalization of neural networks that normalizes the output of a layer to have a mean of 0 and a standard deviation of 1
batch size: hyperparameter of neural networks that determines the number of examples in a batch
Bazaar: version control system developed by Canonical
Bazel: open source build system developed by Google; similar to make
Beam: framework for distributed batch and streaming data processing developed by Apache

BEAM, Bogdan Erlang Abstract Machine: virtual machine for Erlang programming language used also by Elixir
BeautifulSoup: Python library for web scraping and HTML parsing

behavioral pattern: design pattern that describes communication between objects

benchmark: running an experiment on system, component, algorithm to asses its performance, efficiency or other metric
bfloat16: float format suitable for NVIDIA cards

bias: hyperparameter of neural networks that determines the value added to the dot product of weights and inputs before passing it to activation function

bidi, bidirectional text: text containing both left-to-right and right-to-left scripts
big data: term for a large amount of data that is too large to be processed by traditional methods

big O notation, O(): mathematical notation that describes the time complexity of a function (or algorithm) and puts an upper bound on time, memory or other resources required for the function
BigQuery: data warehouse from Google

binary: compiled executable file

binary representation: representation of a value (number) in base 2; i.e. using only 0s and 1s

binary search: search algorithm that finds the position of a target value within a sorted array in logarithmic time by a recursive process of splitting the searched part of the array in half

binary search tree: binary tree where the left child is smaller than the parent and the right child is larger than the parent

binary tree: tree data structure where each node has at most two children
Bison: parser generator from GNU

bit: smallest data representable by a computer, 0 or 1, true or false, on or off
Bitbucket: version control system developed by Atlassian
bitcoin: cryptocurrency and blockchain protocol

bitmap: data structure that represents a bit array

bitmap image: digital image represented by a bitmap
bitrate: data rate of a digital signal measured in bits per second
bitwise operation: operation that is performed on bit level
Blackberry: smartphone brand
blob, binary large object: data type that can store binary data in a database

blockchain: distributed database that is used to maintain a continuously growing list of records called blocks

blog: website that contains posts usually in reverse chronological order
blogosphere: all blogs on the internet and their interconnections

bluetooth: wireless technology for exchanging data over short distances
boilerplate: code that is repeated in many places; e.g. a menu on a website

bookmark: URL stored and easily accessible in a browser
bookmarklet: short JavaScript code executed by visiting a URL starting with javascript: which operates on the current web page and can be saved as a bookmark in browsers

boolean: data type that can have only two values: true or false
Boolean algebra: branch of mathematics that deals with boolean values and logic operations

Boolean formula: formula composed of boolean variables and logic operations
Bootstrap: CSS framework for responsive web design developed by Twitter

bot: software that performs automated tasks; short for robot
Bottle.py: Python web framework distributed in a single file

bottleneck (machine learning): layer in a neural network that has a smaller number of parameters than the previous layer
BPF, Berkeley Packet Filter: technology used for running sandboxed programs in a virtual machine-like construct in the Linux kernel without changing kernel source code or adding additional modules

branch: version control system term for a development copy of a repository that can be merged back to the original repository

breadth-first search, BFS: graph algorithm that starts at the root node and explores all the neighboring nodes before moving to the next level; as opposed to DFS

broadband: network connection with a high data transfer rate

broadcast: network communication method where a message is sent to all nodes in the network

browser: software for viewing and navigating web; usually a GUI application, sometimes TUI application for a terminal

brute force algorithm: algorithm that tries all possible solutions to a problem

BSD, Berkley Software Distribution: Unix operating system developed at the University of California, Berkley

btrfs: file system for Linux developed by Oracle

bubble sort: algorithm for sorting

buffer: data structure for storing data in a memory to be processed later

bug: error in a source code causing an error in compilation or execution

build: process of converting source code into an executable program
build automation: see build system

build system: software for automating the build process

bundle: web app distribution in a single file

bus: computer component for transferring data between components; e.g. PCI bus
bus factor: minimum number of team members needed for running a business
business intelligence, BI: software for analyzing business data

busy waiting: process of waiting for a condition to become true by repeatedly checking it; see also polling

byte: 8 bits

byte pair encoding: encoding used in GPT models, shorter tokens are used for encoding longer words (tokens)

bytecode: code which is compiled for a runtime; e.g. Java and Python bytecodes

C#: object oriented programming language from Microsoft

C++, cpp: object oriented programming language

C, clang: low level programming language

cache: intermediate memory for storing data that is frequently accessed

callback: function passed as an argument to another function to be called later

camelCase: naming convention for identifiers where words are joined together without spaces and each word starts with a capital letter; e.g. getMoreData

Canonical: company behind Ubuntu
Canva: software for creating graphics

canvas: HTML element for drawing graphics
CAP theorem: theorem that states that a distributed system cannot be Consistent, Available and Partition tolerant at the same time

captcha, Completely Automated Public Turing test to tell Computers and Humans Apart: task for distinguishing humans from bots; e.g. a web page with a distorted image of text
captive portal: web page that a user is required to visit and interact with before accessing a public Wi-Fi network
car: Lisp function for returning the first element of a list
cardinality: number of elements in a set
case (flow control): branch of code executed when a condition is met

Cassandra: distributed NoSQL database from Apache

casting: process of converting a data type to another

cat: Unix command for printing the contents of files
catastrophic cancellation: when the difference of two approximate numbers (e.g. lengths) is significantly different from the difference of the real values

categorical data: such data which can have only a discrete set of values; e.g. type of education, gender etc.; as opposed to continuous data

category (mathematics): collection of objects and morphisms between them

CB, continuous build: process of building software after every change to the code base

cd: command for changing the current directory

CD ROM: read only CD

CD RW: rewritable CD

CD, compact disk: medium for storing data

CD, continuous delivery: automatic process of releasing software in short cycles

cdr, contents of decrement part of register: Lisp function for returning the second element of a list

ceil: mathematical function for rounding a number up

Celery: Python library for asynchronous tasks

CentOS: Linux distribution based on Red Hat Enterprise Linux
CEO, chief executive officer: executive position in a company responsible for overall management of a company

certificate: digital document that certifies the identity of a person or an organization

CESNET, Czech Education and Scientific NETwork: NREN for the Czech Republic

CGI, common gateway interface: standard for running programs on a web server

cgroup: Linux kernel feature for limiting and isolating resource usage of a process
chain of responsibility pattern: behavioral pattern

changelog: list of changes in a software project

char: data type for storing a single character

character: representation of a symbol; e.g. a letter, a digit, a punctuation mark etc.

chatbot: bot for chatting with humans

checkbox: GUI element for selecting one or more options

checksum: value calculated from data for detecting errors in the data when the data is transmitted or stored

chip (hardware): integrated circuit

chipset: set of chips on a motherboard
Chomsky hierarchy: classification of formal grammars into four types: regular, context-free, context-sensitive and recursively enumerable

Chrome: web browser from Google
ChromeOS: operating system from Google based on Linux
Chromium: open source web browser from Google

CI, continuous integration: frequent merging of small changes to the main branch

cipher: encryption algorithm
CircleCI: Saas for continuous integration and continuous delivery
circular buffer: see ring buffer

class: blueprint for creating objects with properties (or attributes) and methods (functions)

class inheritance: mechanism of object oriented programming where a class inherits properties of another class
class method: method of a class that is called on the class itself, not on an instance of the class

classification model: model which predicts a class or a category e.g. given a picture: what animal is in the picture or given an audio file: what genre is the song in the file; see also regression

CLI, command line interface, command line: TUI for interaction with the OS and programs using commands
ClickUp: Saas for project management
Clippy: MS Office assistant in the shape of a paperclip
Clojure: functional programming language on JVM runtime based on Lisp
closure: function that has access to variables from the scope where it was defined

cloud computing: model of computing where resources are provided as a service over a network

cluster: group of computers working together

clustering: machine learning technique for grouping data into groups
cmake: build system for C++

Cmd key, command key: modifier key on a keyboard on Apple computers

CMS, content management system: software for managing content on a website; e.g. WordPress

CMYK, cyan magenta yellow black: color model used for printing
CNAME record, canonical name record: DNS record for mapping a domain to another domain

COBOL, common business oriented language: imperative procedural high-level programming language for business applications from the 1950s still in use today (as of 2023)

code: text representation of a software

code completion: IDE feature for completing code based on the context

code control: process of controlling changes to a codebase

code review: process of reviewing code changes resulting either in an approval (for merge) or asking for changes

code smell: characteristic in source code that might indicate a deeper problem

codebase: collection of source code

codec: software for encoding and decoding data, usually audio or video

Colemak: keyboard layout optimized for in English

collaborative filtering: machine learning task for filtering out items that a user might like on the basis of reactions by users with similar preferences; by searching a large group of people and finding a smaller set of users with tastes similar to a particular user

collation: set of rules that determine how data is sorted and compared in a database
collectd: daemon which collects system and application performance metrics
collection (data type): data type that holds multiple values; e.g. array, set
collection (Mongo): group of documents similar to a table in a RDBMS

color model: model for representing colors as tuples of numbers; e.g. RGB, CMYK
combinator: higher-order function
combobox: GUI element that combines a textbox with a dropdown

command, cmd: written instruction causing a computer to do something; consists of a program name, arguments

commit: version control operation that saves changes to a repository
Common Lisp: Lisp dialect
CommonCrawl: web crawler making the data available for research
communicating sequential processes, CSP: concurrency model where processes communicate by sending messages; also a programming language implementing the model

compilation: process of translating source code into machine code

compiler: program that translates source code into machine code
compose key: modifier key that allows to enter special characters by pressing the compose key and then a sequence of other keys; e.g. compose + " + a = ä

compression: process of reducing the size of data
computable function: function that can be calculated by an algorithm

computer: machine that can be programmed to perform computation

computer science: scientific field studying computation, information and automation

computer security, cybersecurity: branch of computer science concerned with the protection of computer systems and networks

computer vision: subfield of computer science
concatenation: operation which joins two data structures into one; e.g. joining two strings or using cat command on two files

concurrency: property of a program which allows multiple tasks to be executed at the same time
Conda: package manager for Python

configuration management: process of managing changes to a system in a way that it maintains integrity over time
conflict: when there are two changes to the same part of a code base which can’t be resolved automatically
confusion matrix: 2D matrix showing the number of correct and incorrect predictions made by a classification model

consistency: property of a database which ensures that all data is valid according to defined rules
console: TUI for interaction with the OS and programs using commands; see also terminal
const: keyword in some programming languages which declares a variable as constant

constant: variable which is not changed during execution of a program
constructor: special method of a class which is called when an instance of the class is created

container: virtualization technology for running applications in isolation
context switching: process of changing focus when doing multiple tasks

context-free grammar, CFG: formal grammar where each production rule is of the form A -> α, where A is a non-terminal and α is a string of terminals and/or nonterminals
context-free language, CFL: formal language that can be generated by a context-free grammar
context-sensitive grammar, CSG: formal grammar where each production rule is of the form αAβ -> αγβ, where A is a non-terminal and α, β, γ are strings of terminals and/or nonterminals

continuous data: data which contain a number that represents a quantity, e.g. weight; as opposed to categorical data

convolution: operation on two functions which produces a third function expressing how the shape of one is modified by the other

convolutional neural network, CNN: neural network using convolution; particularly good for computer vision tasks
cookies: small piece of data sent from a server and stored in the user’s browser used to identify the user and store user-specific data
cooperative multitasking: tasks must yield when they do not need resources
coordinate: number(s) representing a position in space
copy on write: optimization technique where a copy of an object is created only when it is modified
copyleft: the legal technique of granting certain freedoms over copies of copyrighted works with the requirement that the same rights be preserved in derivative works

copyright: legal right to control the use and distribution of a creative work
core: CPU unit that reads and executes instructions; see also multicore

coredump: file containing the memory image of a process when it crashed
coreutils: set of basic Unix CLI tools; e.g. [cat](#cat)

corollary: proposition that follows from and is often appended to one already proved

coroutine: function that can pause execution and return control to the caller
corpus lexicography: field of lexicography that studies corpora for the purpose of creating dictionaries
corpus linguistics: field of linguistics that studies language phenomena using corpora

corpus, text corpus: collection of documents in a readable format

CORS, cross origin resource sharing: mechanism that allows restricted resources on a web page to be requested from another domain
Cortex: complex system for monitoring large systems
CouchDB: NoSQL database from Apache

CPU, central processing unit: the most important processor in a computer; its “control center”
creational patterns: design pattern that deals with object creation mechanisms

Creative Commons: non-profit organization that provides licenses for creative works

credential: piece of document (or information) of an individual issued by a third party

CRM, customer relationship management: business strategy for managing a company’s interactions with customers and also software that provides tools for this

cron: command line utility for scheduling and running tasks for Unix-like OS

cronjob: job or task scheduled and run by cron

crontab: configuration file that specifies shell commands to run on a given schedule (on reboot, every 10 minutes, hourly etc.)

cross compilation: process of compiling code for one platform on a different platform

cross entropy: measure of the difference between two probability distributions

cross validation: technique for assessing how the results of a statistical analysis will generalize to an independent data set

CRUD, create, read, update, delete: set of basic operations for databases

cryptocurrency: digital currency that uses cryptography for security

cryptography: field of computer science concerned with encryption and decryption in order to securely transmit data

CSIRT, computer security incident response team: team that handles computer security incidents

CSS, cascading style sheets: language for styling HTML documents

CSV, comma separated values: text format for storing tabular data
CTO, chief technology officer: executive position in a company responsible for technology used and developed in a company

Ctrl key, control key: modifier key on a keyboard
CUDA, compute unified device architecture: parallel computing platform and programming model developed by Nvidia for GPUs
curl: command line tool for transferring data over network protocols
curryfication: process of transforming a function that takes multiple arguments into a sequence of functions that take one argument each
curse of dimensionality: phenomenon in which algorithms suffer from poor performance on data with high dimensionality

cursor (database): pointer to a row in a database table

cursor (GUI): graphical representation of mouse position

cursor (TUI): movable indicator in TUI showing the current position for new text

CVS, concurrent versions system: software for tracking changes in a software projects

cyclomatic complexity: the number of possible execution paths inside a given piece of code
Cygwin: collection of GNU software that allows running Unix software on Windows
Cygwin (library): dynamic library that provides POSIX compatibility layer for Windows
D3.js, D3: JavaScript library for producing dynamic, interactive data visualizations

daemon: non-interactive background process in Unix whose parent process is init process

DAG, directed acyclic graph: graph with no cycles

Dagster: data orchestration framework

Dash (library): Python framework for building web pplications with interactive visualizations

Dasher: software for writing by eye gaze

data abstraction: TODO

data analysis: process of inspecting, cleansing, transforming and modeling data with the goal of discovering useful information from it

data augmentation: technique for increasing the amount of training data by adding slightly modified copies of already existing data

data contamination: when test examples are included in a model’s training data

data flow diagram: diagram that shows how data is processed by a system
data hazard: TODO
data lake: scalable storage repository that holds data of various types and formats in its raw form for later data analysis
data leakage: see information leakage

data model: model that describes the structure of data and the operations that can be performed on it

data orchestration: process of automating data pipelines

data pipeline: set of data processing elements connected in series where the output of one element is the input of the next one usually in the form of a DAG

data science: field of computer science that uses scientific methods to extract knowledge from data
data splitting: train, development, validation, test parts

data structure: way of storing various data in computers, e.g. graphs, images, list of words, … usually linked to an algorithm which works with the structure effectively to solve a problem

data type: classification of data which tells the compiler or interpreter how the programmer intends to use the data
data versioning: process of tracking changes to data

data visualization: graphical representation of data which helps to understand the data and find patterns in it

data warehouse: centralized repository that stores cleaned and transformed structured data designed for query and analysis
database session: period of time during which a database connection is open

database, database system: collection of data organized in a way that allows access, management and updates
dataframe: data structure for storing data in a tabular format used in Pandas

dataset: collection of data used for machine learning or data science

DAW, digital audio workstation: software for recording, editing and producing music in the form of audio files
dbt, data build tool: open source CLI tool
deadlock: situation in which two or more competing actions are each waiting for the other to finish
Debian: Linux distribution

debugger: software for debugging software; e.g. GDB

debugging: process of finding and fixing bugs in a software

decentralized system: system in which components are distributed across a network

decision tree: machine learning architecture that uses a tree-like graph to model decisions

declaration: statement that introduces a name into a program

declarative programming language: programming language that expresses the logic of a computation without describing its control flow; e.g. SQL
decoding: process of converting encoded data into its original form
decorator: function that takes another function and returns a modified function with additional functionality

decryption: process of converting encrypted data into its original form

deep learning: technique to extract and transform data by using multiple layers of neural networks
defragmentation: process of reducing fragmentation of files in a file system
Delphi: IDE for Pascal

dense matrix: matrix with many non-zero elements

dense vector: vector with many non-zero elements

dependency: software or library or framework that is required for another software, library of framework to work
dependency injection: technique for passing dependencies to a function or object from the outside

deployment: process of making a software available for use by others

depth-first search, DFS: graph algorithm that explores as far as possible along each branch before backtracking
deque: data structure that allows adding and removing elements from both ends
derivative: measure of how a function changes when its input changes
deserialization: process of converting serialized data into its original form
design by contract, DBC: software design approach that uses formal specification of software components
design for testability patterns: design pattern that deals with testing

design pattern: general, reusable solution to a commonly occurring problem in software design; e.g. MVC
desktop: computer designed for regular use at a single location

desktop environment: graphical user interface for an OS

deterministic finite automata: type of finite automata where each state has exactly one transition for each possible input
development set, dev set: see validation set

DevOps, development and operations: set of practices that combines software development and IT operations; e.g. CI/CD

dialog (UX): UI element that allows users to interact with a computer
dictionary: data structure that maps keys to values
differentiability: property of a function that it can be approximated by a linear function
dig: Unix CLI tool for querying DNS name servers

digit: symbol used to represent numbers; e.g. 2

digital data: discrete representation of data
Digital Millennium Copyright act section 1201(A): U.S. law that makes it illegal to circumvent digital locks over copyrighted works, restricting user’s ability to access, use, or discuss materials that they own

dimension: the minimum number of coordinates needed to specify any point within a given space

dimensionality: number of features in a dataset

directory: file system data structure which contains references to children files and directories in a tree-like structure
Discord: proprietary software for chatting and video conferencing
discretization: process of transforming continuous data into discrete data

disk, disc: storage device for computers

dispatcher: function that receives a request and sends it to the appropriate handler

display: physical or virtual screen of a computer

distributed computing: field of computer science that studies distributed systems

distributed system: system whose components are located on different computers in a network

distribution: see Linux distribution
Django: Python web framework
DNS packet: packet that contains a DNS query or a DNS response

DNS record: data record in the DNS that contains information about a domain
DNS resolver: server that receives DNS queries and returns the corresponding DNS records

DNS, domain name system: hierarchical and decentralized naming system for computers, services, or other resources connected to the internet or a private network
DocBook: markup language for technical documentation

Docker: containerization platform and software for building, running and shipping containers
docker compose: Docker CLI tool for defining and running multi-container Docker applications

Docker image: file that contains a container and all its dependencies
Dockerfile: file that contains instructions for building a Docker image

documentation: description of the behavior of a software or API, intended for its users

Docutils: Python library for parsing reStructuredText into HTML, LaTeX, XML or plain text
Dojo: JavaScript toolkit for building web applications

DOM, document object model: cross-platform and language-independent API for representing and interacting with XML documents

domain: name that identifies one or more IP addresses with a name that is easier to remember and use in URLs

domain specific language, DSL: high level abstration programming language that uses concepts and rules from a specific domain
DoS, denial of service: cyber attack that prevents legitimate users from accessing a service by overwhelming it with requests

dot product: a scalar operation that takes two equal-length sequences of numbers and returns a single number
dotfiles: files that start with a dot (.) and contain configuration for software and operating systems
double: data type for representing floating-point numbers with double precision in C and C++
download: process of copying data from a remote server to a local computer

driver: software that controls a device
driver code: short code demonstrating or testing a specific program, function, code
DRM, digital rights management: technologies that restrict the use of digital content and devices after purchase

dropdown: UI element that allows users to select one value from a list
Drupal: content management system written in PHP

DRY, don’t repeat yourself: principle of software development that states that every piece of knowledge must have a single, unambiguous, authoritative representation within a system

durability: property that ensures database transactions are permanent
Dvorak: keyboard layout designed for typing speed and comfort

dynamic library, dll: library that is loaded at run time

dynamically typed language: programming language that does not require variable types to be defined explicitly
e-ink, eink: display technology that mimics the appearance of ink on paper

E2E test, end to end test: test that tests a complete software system from beginning to end

early binding: process of linking a function call to its definition at compile time

ebook: book in digital form
Eclipse: IDE for Java

Ecma international: international standards organization for information and communication systems
ECMAScript: JavaScript standard by Ecma International
ecosystem: libraries, community, online resources etc. related to a given programming language

edge: link between two nodes in a graph

edge computing: computing that takes place near the source of data

editor: software for editing files
egg: logical structure for releasing a specific version of a Python project (code, resources, metadata) which is importable and distributable

Elasticsearch: search engine based on Lucene
Electron: framework for building desktop applications with web technologies

element: component (node) of a XML document; e.g. <head>

Elixir: functional programming language running on BEAM runtime supporting extreme concurrency and fault tolerance

Emacs: text based editor with a built-in Lisp interpreter

email: electronic mail sent over a network
embedded system: computer system with a dedicated function within a larger mechanical or electrical system

embedding: mapping of a discrete feature to a vector
Ember.js: JavaScript framework for building web applications
emdash: dash that is as wide as the letter m
encapsulation: OOP principle of bundling data and methods that operate on that data

encoding: process of converting data from one form to another

encryption: process of encoding a message or information in such a way that only authorized parties can access it
endash: dash that is as wide as the letter n
endofunctor: functor mapping a category to itself
enum: data type consisting of a set of named values
environment variable: variable whose value is set outside the program

epoch: one complete pass through the training data
epub: ebook format based on XML
ER diagram, entity relationship diagram: diagram that shows the relationships between entities in a database

ereader, e-reader: device for reading ebooks in digital format

Erlang: functional programming language using BEAM runtime supporting extreme concurrency and fault tolerance

ERP, enterprise resource planning: software for managing business processes
error: mistake
error handling: process of responding to error conditions
error output, stderr: output stream for error messages in Unix

error rate: the proportion of validation set which was incorrectly predicted or labelled or classified by a model
Esc, escape: key on a keyboard used to generate the escape character (used to exit a mode)
escape sequence: sequence of characters that represents a special character; e.g. \n represents a newline character in Python
esoteric programming language, esolang: programming language designed to test the boundaries of computer programming language design
esports: competitive video game playing
estimator: model that learns from data
ETH, ether: cryptocurrency and blockchain platform
Ethereum: blockchain platform for smart contracts
ethernet: standard set of protocols for LAN, MAN and WAN networks

evaluation (machine learning): process of assessing a model’s performance on validation set
event-driven programming: programming paradigm in which the flow of the program is determined by events such as user actions, sensor outputs, or messages from other programs

exabyte: 10^18 bytes or 1,000 petabytes

Excel: spreadsheet application by Microsoft; part of Microsoft Office

exception: situation when a program cannot continue its execution due to an error or unexpected situation; the flow of execution is interrupted

executable, exe: binary file that can be run as a program
execution: process of running a program, command or script

EXIF, exchangeagle image file format: standard that specifies formats for images to store special tags (attributes) used by digital cameras

exponent: number that indicates how many times a base number is multiplied by itself
expression: combination of values, variables, operators and functions that is evaluated to a single value
Ext js: JavaScript framework for building desktop applications
ext3: file system used by Linux
ext4: journaling file system used by Linux

extension: the suffix of filename indicating its file format
F score, F1 score, F measure: harmonic mean of precision and recall
F#, F sharp: functional programming language running on .NET runtime

Facebook: SaaS for social networking, also the name of the company that created it
factorial: unary function with the definition f(n) = n! = n * (n-1) * (n-2) * … * 2

factory: design pattern that defines an interface for creating an object but lets subclasses decide which class to instantiate
FAISS, Facebook AI Similarity Search: library for efficient similarity search and clustering of dense vectors

false negative: item which was incorrectly classified as negative (not belonging to a class)
fan in: number of inputs to a component
fan out: number of outputs from a component

fast.ai: library which simplifies building and training models built upon Pytorch
fat finger: situation when a user accidentally presses the wrong key on a touch screen

FAT, file allocation table: file system for hard drives by Microsoft
FAT32: 32-bit version of the FAT file system

feature: measurable property of a phenomenon being modeled in machine learning; e.g. a pixel value in an image or a word in a sentence
feature creep: situation when a software product has more features than originally planned

Fedora: Linux distribution sponsored by Red Hat

fetch (git): operation that downloads commits from a remote repository to a local repository
few shot learning: machine learning technique that learns from a small number of examples
ffmpeg: command line tool for manipulating video and audio

FIFO, first in first out: data structure where the first element added is the first element removed
Figma: collaborative web app for design

file: representation of data stored on a disk; not necessarily contiguous
file buffer: buffer used to store data read from or written to a file

file descriptor: number that uniquely identifies an open file in an operating system

file format: standard that specifies the structure and encoding of a file

file system: method of storing and organizing files on a disk

fine tuning: transfer learning technique where the parameters of a pretrained model are updated by training for additional epochs; many-shot learning

finite automata: automata with a finite number of states used to model computation and regular expressions
Firebase: app development platform by Google
Firefox: web browser by Mozilla
firewall: security system that monitors and controls incoming and outgoing network traffic
first-order logic: see predicate logic

fit: the process of updating the parameters of a model such that the predictions of the model using the training data match the target labels

fixed point (programming language): representation of real numbers within a set range by maintaining a fixed number of digits after the decimal point; cf. floating point representation
Flash: multimedia platform by Adobe

flash memory: see flash storage

flash storage: non-volatile storage device that can be electrically erased and reprogrammed
Flask: Python web framework
flat file: file that contains records with no structured relationships
flexbox: CSS layout mode that arranges elements in a single direction

float: data type for representing floating point numbers

floating point number: binary representation of real numbers using a significand and an exponent
floor: mathematical function that rounds a real number down to the nearest integer
Flutter: UI toolkit for building mobile applications developed by Google
fold (functional programming): see reduce
folder: see directory

font: set of glyphs used to render text
foo bar baz: placeholder names used in programming
for loop: loop that repeats a block of code for a specified number of times
foreign key: key that references a primary key in another table

fork: copy of a repository in a version control system

formal grammar: set of rules for constructing valid sequences of word or tokens in a language

formal language: set of strings of symbols that may be constrained by rules
formal verification: process of proving that a program satisfies a specification
FORTH: programming language based on stack data structure
Fortran: programming language designed for scientific computing

forward pass: applying the model to the data based on the current parameter values

fragmentation: phenomenon where a file is stored in non-contiguous blocks on a disk
frame: single image in a sequence of images that make up a video
FrameNet: lexical database of English representing the meaning of words in terms of frames and semantic roles
framerate, fps, frames per second: number of frames displayed or processed per second

framework: collection of tools, components, guidelines and libraries to simplify the development of software

frontend, FE, front-end: software component that interacts directly with the user
FTP, file transfer protocol: protocol for transferring files between computers on a network
FTR: abbreviation of “for the record”

full duplex communication: communication where both parties can send and receive data simultaneously

full text search: information retrieval technique for finding documents that contain a given set of words from a query

function: subroutine that can be called from other parts of the program

functional paradigm: programming paradigm based on the concept of functions and immutability

functional programming language: programming language based on functional paradigm; e.g. Haskell, Erlang

functor: morphism function between categories
FUSE, file system in userspace: API for implementing file systems in userspace
g++: C++ compiler based on GCC

garbage collector, GC: part of a programming language runtime system that manages the memory used by a program; its goal is to identify and reclaim memory which is no longer in use (referenced by the program), prevent memory leaks and improve memory efficiency
garden path sentence: ambiguous sentence that leads to an incorrect (partial) interpretation or syntactic analysis; e.g. time flies like an arrow; fruit flies like a banana
gate (machine learning): TODO

GB, gigabyte: 10^9 bytes or 1,000 megabytes
GCC, GNU compiler collection: collection of compilers for programming languages developed by the GNU project
gdb, GNU debugger: debugger for C and C++ developed by the GNU project
generics: programming language feature allowing to define functions and types that can be used with different types of data
gensim: Python library for topic modeling

geolocation: the process of determining the location of a device using its IP address

geometry: branch of mathematics that studies the properties of shapes
getaddrinfo: system call that resolves a domain name to an IP address
getter (OOP): method that returns the value of a property
gevent: coroutine-based Python library for asynchronous programming based on greenlets

gibibyte: 2^30 bytes or 1,024 mebibytes

GIL, global interpreter lock: mutex allowing only one thread to hold the control of the Python interpreter
GIMP, GNU image manipulation program: raster graphics editor by the GNU project

git: distributed version control system
GitHub: SaaS git
GitLab: open source SaaS git with self-hosting option
global scope: scope that is visible in the whole program
global variable: variable with a global scope
glue tag: TODO

glyph: graphical representation of a character
Gmail: SaaS email service from Google
Gnome: desktop environment

GNU, GNU is not Unix: project to create a free operating system
GNU/Linux: Unix-like operating system based on the Linux kernel and the GNU userland; not to be confused with Linux

Go, golang: open source programming language designed to be suitable for systems programming

Google: company that provides SaaS services and open source software
Google Analytics: SaaS for tracking website traffic by Google

Google Cloud Platform, GCP: Iaas by Google
Google Maps: SaaS for maps by Google
Google Play: SaaS for distributing Android apps
Google Plus: discontinued (as of 2019) social network by Google
Google Search: SaaS search engine
Google Wave: discontinued (as of 2012) SaaS for real-time communication and collaboration developed by Google
goroutine: thread in Go
GOTO: statement that causes the program to jump to another part of the code; included e.g. in BASIC and C
GPS, global positioning system: satellite navigation system

GPT, generative pre-trained transformer: family of large language models from OpenAI
GPT4: large language model from OpenAI based on GPT
GPTQ, generative pre-trained transformer quantization: type of post-training quantization for LLM compressions (from 32 bits down to 3–4 bits)

GPU, graphics processing unit: special type of processor suitable for massive parallel computing

gradient: how each element is contributing to the final value (objective function) of the model; it is the slope of the function or the partial derivation; all functions and their elements must be differentiable
gradient boosting: machine learning technique for regression and classification problems, which produces a prediction model in the form of an ensemble of weak prediction models, typically decision trees

gradient descent: method for finding local/global maxima/minima in a function
Grafana: dashboard for time series data

graph (discrete mathematics): set of vertices and edges connecting them

graph (mathematical analysis): visual representation of a function

graph database: NoSQL database for storing and easy querying graph data

graph theory: branch of mathematics studying graphs
graphene: Python library for GraphQL
graphics card: hardware component that generates and outputs images to a display it also allows parallel computing

GraphQL: query language for API developed by Facebook

greenlet: Python library for cooperative multitasking
grep, global regular expression print: CLI tool for searching and filtering text files
grid (CSS): grid is a set of intersecting horizontal and vertical lines defining the columns and rows
grid search cross validation: trying all combinations; suffers from exponential growth of combinations

gRPC, google remote procedure call: implementation of RPC using protobufs sent over HTTP 2.0 protocol in full-duplex mode

GRU, gated recurrent unit: recurrent neural network that uses gates to control the flow of information
GTK, GIMP toolkit: widget toolkit for creating GUI applications

GUI, graphical user interface: user interface that allows users to interact with computer programs through graphical icons and visual indicators; a mouse is typically used to manipulate the elements of the user interface
gulp: JavaScript build system
gunicorn: HTTP server
gzip: CLI tool for compressing data
H.264: video compression standard
H.265: video compression standard

Hadoop: framework for distributed computing developed by Apache
hallucination: nonsensical output of a generative model; coined by Andrej Karpathy

hardware: physical part of a computer
hardware abstraction layer, HAL: layer of abstraction that allow the operating system to interact with the hardwareo at a general level rather than at a detailed hardware level

hardware accelerator: processor designed to accelerate machine learning tasks

harmonic mean: reciprocal of the arithmetic mean of the reciprocals of the values

hash: result of a hash function
hash collision, collision: the situation when two different inputs produce the same output in a hash function

hash function: function that maps data of arbitrary size to data of fixed size

hash table, hash map: data structure that maps keys to values using a hash function
Haskell: functional programming language from 1990s with a strong type system

HCI, human-computer interaction: discipline concerned with the design, evaluation and implementation of interactive computing systems for human use and with the study of major phenomena surrounding them

HDD, hard disk drive: electro-mechanical non-volatile data storage device
head (Unix command): CLI tool for printing the first few lines of a file
head (model): the part of a pretrained which is removed and replaced with a randomly initialized layers suitable for fine-tuning on the target dataset
headless browser: web browser without a GUI usually used for automated testing and web scraping

heap: tree based data structure that satisfies the invariant property that the parent node is greater than its children nodes (in the case of a max heap)
heap sort: algorithm for sorting using heap in O(n log n)
Helm: package manager for Kubernetes

Heroku: PaaS that enables developers to build, run, and operate applications entirely in the cloud
Heroku: platform as a service for web applications

hexadecimal number, hex: number in base 16
hibernation: power-saving state of a computer that uses the least amount of power when the memory is saved to the hard drive

hidden Markov model, HMM: statistical model that assumes the Markov property and that the state is not directly observable

high level programming language: programming language with strong abstraction from the details of the computer

higher-order function: function that takes another function as an argument, returns a function as a result, or both

host: computer connected to a network
hot swapping: replacing a module without stopping the whole system
hover: user interface event when the mouse is moved over an element without pressing any buttons
HP, Hewlett-Packard: software and hardware company
HPC, high-performance computing: field of computer science that deals with supercomputers and parallel computing

HTML, hypertext markup language: markup language for creating hypertext documents
HTML5: version of HTML with new elements, attributes, and behaviors e.g. <video>, <audio>, <canvas>, <svg>
htmx: JavaScript library for AJAX with HTML attributes
htop: CLI tool for monitoring system resources
HTTP DELETE: HTTP method for deleting data
HTTP GET: HTTP method for retrieving data

HTTP POST: HTTP method for creating data

HTTP server: server that serves HTTP requests
HTTP verb: GET, DELETE, PUT, POST

HTTP, hypertext transfer protocol: text based protocol for client-server communication over the Internet

HTTP/1: version of HTTP
HTTPS: secured HTTP
HuggingFace: company that develops NLP models and framework of the same name
Hungarian notation: naming convention for variables where the name of the variable indicates its type; e.g. strName or lNames

hyperlink: reference to another document or part of it that a reader can follow by clicking

hyperparameter: parameter of the model; e.g. learning rate

hyperparameter tuning: process of finding optimal hyperparameters of a model

hypertext: text that contains hyperlinks

hypervisor: software that creates and runs virtual machines and manages them

hyphen: punctuation mark -

IaaS, infrastructure as a service: cloud computing service that provides virtualized computing resources over the internet
iCloud: cloud storage service by Apple
ICQ: instant messaging application; the name derives from “I seek you”
id: unique identifier, usually a number

IDE, integrated development environment: software containing tools to facilitate software development

identifier: name that identifies a variable, function, class, or other object
IDL, interface definition language: programming language for defining interfaces
iframe: HTML element that allows to embed another HTML document inside the current one

IM, instant messaging: form of communication that offers real-time text transmission over the Internet
image classification: task in machine learning
ImageNet: image-based dataset for machine learning with 14 million images in more than 20,000 categories

IMDb: website with information about movies and TV shows

immutability: property of an object that cannot be changed

imperative programming language: programming language that uses statements to change a program’s state

implementation: realization of a specification in a programming language

implicit type conversion: type conversion that is done automatically by the compiler

IMSI, international mobile subscriber identity: key used to identify a subscriber on mobile telephony devices

indexing: process of creating an index for a database to improve the speed of data retrieval

inference (logic): process of deriving logical conclusions from premises

inference (ML): process of using a model to make predictions

infimum: greatest lower bound

infix: notation where the operator is placed between its operands
information hiding: TODO

information leakage: situation when a model is able to learn validation set from the training data

information retrieval: discipline of computer science
ingest: process of importing multiple data into a single data store
ingress: Kubernetes resource that manages external access to the cluster by exposing HTTP and HTTPS routes
inheritance: see class inheritance
INI: text format used for configuration

init process: process with PID 1 in Linux that starts and manages other processes

init system: system in Linux that starts and manages services
initialization (of weights): it’s advisable not to initialize weights to zero even though the initial loss is close to optimal in this case but biases can be zero; some neurons might be initialized to dead neurons unable of learning (tanh in flat regions, ReLU in the negative numbers, …)
inorder: tree traversal where the root is visited between the left and right subtrees

input device: hardware that sends data to a computer; e.g. keyboard, mouse
INSERT: SQL statement to insert data into a table

insert: operation in database systems to add one item into a table or collection
insert sort, insertion sort: algorithm for sorting

instruction set: set of instructions that a processor can execute; e.g. x86, ARM, MIPS, RISC-V, …

integer: basic data type to represent whole numbers
integral: mathematical object that can be interpreted as an area or a volume

integrated circuit: electronic circuit made of semiconductor material
integration test: test that checks if the components of a system work together
integration tests: tests that check if the whole system works as expected

Intel: company that designs and manufactures microprocessors
IntelliJ: IDE for Java and other languages
intercom, intercommunication system: communication system within a building

interface (OOP): contract that specifies what a class can do without specifying how it does it

internet: global network of computers; connected via internet protocol
interpreted language: programming language which doesn’t need to be compiled into machine code and is interpreted by an interpreter; e.g. Python

interpreter: software that interprets source code of a programming language and executes it

interprocess communication, IPC: mechanism that allows processes to communicate with each other
interruption: situation when a process is stopped by an interrupt request
intersection: set of elements that are common to two or more sets
invariant: property which stays the same after a set of operations is executed
IO, I/O, input/output: type of operation that reads from or writes to a storage

iOS: operating system for Apple devices

IoT, internet of things: network of devices connected to the internet

IP address: unique identifier of a device connected to the internet used in the internet protocol

IP, internet protocol: set of rules that govern how data is sent and received over a network
iPhone: smartphone developed by Apple
iptables: program for configuring firewall in Linux

IPv4: version of IP using 4 bytes for addresses

IPv6: version of IP using 6 bytes for addresses

IRC, internet relay chat: text based instant messaging system

IRQ, interrupt request: signal sent to CPU that temporarily stops a running process

ISP, internet service provider: company that provides access to the internet

J: array programming language

Java: object-oriented programming language developed by Sun Microsystems

Java Virtual Machine, JVM: virtual machine that executes Java bytecode
JavaDoc: documentation generator for Java

JavaScript: programming language used mainly for web development
Jenkins: open source automation server for continuous build and continuous deployment

Jinja: templating language for Python

join: operation that combines two or more database tables into one
Joomla: content management system

journaling file system: file system that keeps track of changes not yet commited to the file system’s main part by recording the goal of such changes

JPEG, JPG, joint photographic experts group: method of lossy compression for digital images; files with this compression usually have .jpg or .jpeg extension

jQuery: JavaScript library for DOM manipulation
jQuery UI: jQuery library for creating user interfaces
JSF, JavaServer Faces: Java framework for web development

JSON, JavaScript Object Notation: text format for representing structured data

JSONL, JSON Lines: text format where each line is a JSON object
Julia: LLVM based, dynamically typed programming language for high performance
Julia: LLVM based, dynamically typed programming language for high performance
Jupyter notebook: web application for creating and sharing computational documents
Jupyter notebook: web application for creating and sharing computational documents

JWT, JSON web token: standard for creating access tokens that assert some number of claims

K: proprietary array programming language
k8s: see Kubernetes
Kafka: framework for distributed streaming data processing developed by Apache
Kaggle: data science competition platform and online community
Kaimin normalization: torch.nn.init.kaiming_normal_()
kanban: visual system supporting decision making in project management; commonly in combination with scrum
kata: short practise, e.g. implementing a known algorithm

KDE, K Desktop Environment: desktop environment for Linux and other Unix systems
KDY, Know Your Domain: process of understanding the domain of the software being developed
keepassx: password manager for Linux
Keras: open source neural network library written in Python

kernel (machine learning): function that takes two inputs and returns a similarity score

kernel (operating system): program at the core of an operating system controlling everything in the system; it manages processes, memory, I/O operations etc
kernel based virtual machine: open source virtualization module in the Linux kernel; developed by Red Hat

key (cryptography): piece of information that controls the operation of a cipher

key value store: database that stores data as a collection of key-value pairs

keyboard: peripheral device for input of text
keynav: software for controlling the mouse cursor with the keyboard

kibibyte: 2^10 bytes or 1,024 bytes

kilobyte: 1,000 bytes

Kindle: e-reader developed by Amazon

KISS, keep it simple stupid: approach to software development based on the idea that most systems work best if they are kept simple rather than made complicated

knowledge base: collection of knowledge in digital form
Kotlin: programming language for Java virtual machine and Android
KPI, key performance indicator: measurable value that demonstrates how effectively a company is achieving key business objectives

Kubernetes: open source system for automating deployment, scaling and managing containerized apps

KVM, kernel-based virtual machine: virtualization technology built into Linux kernel since version 2.6.20 (2006)
KYC, Know Your Customer: process for identifying and verifying the client’s identity to protect financial institutions against fraud, corruption, money laundering etc.

L1 regularization: regularization technique that adds a penalty equal to the sum of the absolute value of the coefficients

L2 regularization: regularization technique that adds a penalty equal to the sum of the squared value of the coefficients

label: correct answer of the ground truth associated with the input data
lambda calculus: mathematical model of computation based on function abstraction and application using variable binding and substitution; introduced by Alonzo Church in 1930s
lambda function: anonymous function
lamma.cpp: library for running pretrained models; it has a wrapper for Python

LAN, local area network: computer network that interconnects computers within a limited area such as a school, university campus or office building
LanceDB: open source vector database for machine learning applications
laptop: portable computer
Laravel: PHP web framework
laser printer, laserjet: printer that uses a laser beam to produce an image on a drum that is rolled through a reservoir of toner and then onto a sheet of paper
late binding: technique of deferring the resolution of a symbol until runtime

LaTeX: Tex macro package for typesetting documents

lattice: partially ordered set in which every two elements have a unique supremum and a unique infimum

layer: neural network component that transforms input data into output data represented as a matrix
layout engine: software component that combines content with a presentation to generate an output that can be displayed on a screen
lazy evaluation: evaluation strategy that delays the evaluation of an expression until its value is needed

LDA, latent Dirichlet allocation: generative statistical model that allows sets of observations to be explained by unobserved groups that explain why some parts of the data are similar

LDAP, lightweight directory access protocol: protocol for accessing and maintaining services over IP; a common use is to provide a central place to store usernames and passwords

leaf node: node that has no children

lean: set of practices that emphasize efficiency, minimizing waste, and delivering value to customers; originated in Toyota manufacturing

learner: combination of a model and data; used e.g. in fastai library

learning rate: how much to change the weights each pass; too large: the learning is unstable; too small: the learning takes ages
learning rate decay: learning rate can be dynamic (with regards to the step number within optimization); can be determined by tracking losses while altering learning rates
least upper bound: element that is greater than or equal to other elements; see also supremum
Leet Code: website for practicing design and implementation of algorithms
left join: join that returns all rows from the left table and the matched rows from the right table

lemma (linguistics): the canonical form of a lexeme

lemma (mathematics): proven proposition which is a corollary to one or more of the axioms of a system

lexeme (linguisitics): unit of lexical meaning that underlies a set of words that are related through inflection; e.g. run, runs, ran and running are forms of the same lexeme run
lexer: program that performs lexical analysis
lexical analysis: process of converting a string into a sequence of lexemes

LGTM, looks good to me: commonly used in code review as an approval formula

library: collection of code (usually organized into modules) that can be used by other programs

license: legal document that grants permission to use a software under certain conditions

LIFO, last in first out: data structure where the last element added is the first one to be removed
linear data structure: data structure where the elements are arranged in a linear order
linear probing: hashing technique that resolves collisions by trying the next slot in the hash table
linear regression: not to be confused with regression

linked list: data structure for storing a sequence of items
linkrot: process by which hyperlinks on individual websites in general point to web pages, servers or other resources that have become permanently unavailable
linter: program that analyzes source code to flag programming errors, bugs, stylistic errors, and suspicious constructs

Linux: Unix-like operating system assembled under the model of free and open-source software development and distribution

Linux distribution: Linux kernel bundled with additional software; e.g. Ubuntu, Fedora or Arch Linux

Liskov substitution principle: principle in object-oriented programming stating that if S is a subtype of T, then objects of type T may be replaced with objects of type S without altering any of the desirable properties of the program

Lisp: functional programming language with fully parenthesized syntax from the 1950s
list: see array
list comprehension: syntax for creating a list based on existing lists; e.g. [x**2 for x in range(10)]
literate programming: programming paradigm introduced by Donald Knuth that combines documentation and source code into a single source file
little endian: byte order of a binary number in which the least significant byte is stored first
livelock: situation when two or more processes are in a loop waiting for each other to finish and thus none of them can finish

LL grammar, left linear grammar: grammar in which all productions are of the form A → wB or A → w where A and B are nonterminals and w is a string of terminals
LL1 grammar, left linear grammar with one lookahead: LL grammar in which the first terminal of B is unique

LLM, large language model: language model with a large number of parameters usually based on transformers

LLVM, low level virtual machine: compiler and toolchain between any instruction set architecture and any [programming language](#programming language) written in C++

LM, language model: complicated function with many parameters which is very flexible
LMGTFY, let me google that for you: sarcastic response to a question that could have been easily answered by a search engine

load balancing: technique used to distribute workloads across multiple resources
load factor: ratio of the current number of elements in a hash table to the size of the hash table
LOC, line of code: number of lines of code used as a metric for measuring the size of a software program

lock: synchronization mechanism for enforcing limits on access to a resource in an environment where there are many threads of execution

log: special output of a program used for debugging and troubleshooting, monitoring and performance analysis and security and auditing
log linear time complexity: time complexity of an algorithm that is proportional to the logarithm of the input size
log loss: see cross entropy loss function
logging: process of recording events, actions or messages that occur during a program execution and which produces logs
logic gate: electronic circuit that implements a logic operation

logic operation: function that takes one or more boolean values and produces a boolean value

logistic function: function which startsoff increasing slowly, then grows more rapidly and eventually levels off; the f(x) values are usually between 0 and 1
logit: function that maps probabilities to the real numbers which are easier to work with in logistic regression
long: representation of integers in some programming languages
lookahead (parser): a number of tokens that are read ahead to decide which production to use in parsing
lookahead (regular expression): a number of characters that are read ahead in a string to decide whether a pattern matches

loop: sequence of instructions that is repeated until a certain condition is reached; a part of flow control in programming languages

loss: function which takes a model’s predictions and the desired output and computes a number expressing how good is the model’s prediction; used for driving the stochastic gradient descend; the measure of model’s performance

lossless compression: compression where no information is lost when the compressed data are uncompressed; cf. lossy compression

lossy compression: compression where some information is lost when the compressed data are uncompressed; cf. lossless compression

low level abstraction: abstraction provides little or no abstraction from a computer’s instruction set architecture
LR grammar, left regular grammar: grammar in which all productions are of the form A → wB or A → w where A and B are nonterminals and w is a regular expression

lsof: command for listing open files

LSP, language server protocol: protocol for enabling IDE features for a programming language

LSTM, long short term memory: recurrent neural network model which can learn long-term dependencies

LTE, long-term evolution: standard for wireless broadband communication for mobile devices and data terminals

LTR, left to right text: text (or script) written from left to right; e.g. Latin, Cyrillic

Lua: programming language designed primarily for embedded systems and clients
LuaTeX: TeX engine that uses Lua as an embedded scripting language

Lucene: Java library for information retrieval and full-text search
LWN.net: Linux news website and webzine

machine code: set of instructions executed directly by a computer’s CPU

machine learning, ML: field of computer science; training of programs developed by allowing a computer to learn from its experience (training data) rather than through manually coding the individual steps

MacOS: operating system for Apple computers

macro: rule or pattern that specifies how a certain input sequence should be mapped to a replacement output sequence according to a defined procedure; a part of preprocessor
mail: see email
mail list, mailing list: collection of names and addresses used by an individual or an organization to send material to multiple recipients

mail server: server that handles email
main branch, main: see master branch
main memory: see RAM

make: build automation CLI tool
Makefile: script for make
malloc: C function for allocating a block of memory on the heap

malware: malicious software

man page: documentation for a command or other software component in Unix

MAN, metropolitan area network: network that interconnects users with computer resources in a geographic area or region larger than that covered by even a large LAN but smaller than the area covered by a WAN

managed code: code that runs on a virtual machine and is managed by the runtime

mantissa: part of a floating point number which contains its significant digits
many to many, m:n: relationship between two entities where one entity can be associated with many instances of the other entity and vice versa

map: see dictionary
MapReduce: algorithm for processing and generating large data sets
MariaDB: relational database management system that is a drop-in replacement for MySQL

markdown, MD: text format for specifying basic formatting
Markov chain: stochastic model describing a sequence of possible events in which the probability of each event depends only on the state attained in the previous event

Markov property: property of a stochastic process where the conditional probability distribution of future states of the process depends only upon the present state, not on the sequence of events that preceded it

markup language: text format that uses tags to define elements within a document
masking: operation that uses a sequence of bits to select or modify a subset of bits

master branch, master: branch in a version control system which is considered to be the definitive version of the source code

material design: UI design system developed by Google
material UI: React component library implementing material design
Matlab: software for numerical computing
Matlab: programming language and GUI for numerical computing

matrix: rectangular array of numbers arranged in rows and columns; a vector of vectors
Mattermost: open source, self-hosted Slack alternative
McCabe complexity: see cyclomatic complexity

MDA, model driven architecture: software development methodology that focuses on modeling and model transformation

mean: quantity that has a value which is intermediate to the extreme values of a set of numbers

mebibyte: 2^20 bytes or 1,024 kibibytes

median: value separating the higher half of a sorted data sample from the lower half

megabytes: 10^6 bytes or 1,000 kilobytes

memory: a part of computer that is used to store information for immediate use; see also RAM

memory address: representation of the location of bytes in memory

memory allocation: process of allocating a block of memory for a variable in a program
memory footprint: amount of memory used by a program

memory leak: bug in a program that causes it to consume more and more memory over time
memory management: consists of tasks like memory alloocation, memory freeing

memory safety: property of a programming language that guarantees that a program cannot access memory that it is not authorized to access
Mercurial: distributed version control system

merge: action of adding code changes to an existing code base in a repository; usually integrating commits from a development branch into the main branch
merge conflict: situation when two commits cannot be merged automatically by a version control system since they modify the same part of the code

merge request: process of asking code maintainers and code reviewers to review code changes; cf. pull request

merge sort: algorithm for sorting
message broker: software that translates messages from the formal messaging protocol of the sender to the formal messaging protocol of the receiver
metadata: data about data; e.g. EXIF allows adding information about the camera, lens, location and date to an image file for a photo taken with a smartphone

method: function that is defined inside a class

metric: function that measures the quality of a model or a system; used for human consuption as opposed to loss
micro-benchmark: short snippet of code used to compare performance or other qualities/properties

microblogging: a form of blogging that allows users to write brief text updates
microprocessor: integrated circuit that contains all the functions of a CPU of a computer

Microsoft: company that develops and sells computer software and consumer electronics

Microsoft Office: suite of office software developed by Microsoft
mid-squared method: pseudorandom number generator

MIPS architecture: reduced instruction set computer architecture
MIPS, million instructions per second: measure of a computer’s performance
mixed type: data type that can contain values of different types
Mkdocs: static site generator for documentation
MKV, Matroska video, matroska: video container file format that can hold an unlimited number of video, audio, picture, or subtitle tracks in one file

MLC, machine learning compilation: project for running models on various platforms
MLOps, machine learning operations: practice of applying DevOps to machine learning

mmap, memory map: system call that maps files or devices into memory without copying data
MMLU, massive multitask language understanding: multilingual dataset for benchmarking large language models
mobi: ebook file format for Kindle
mock: simulated object or function that mimics the behavior of a real object or function
modal: dialog that requires users to interact with it before they can return to the main application which usually overlays the main application window

model: special kind of a general program which can do many things based on weights and its architecture
model parameters: see parameters

model view controller, MVC: software design pattern for implementing user interfaces where model represents the structure of data, view is a way of interpreting the model and controller is a way of controlling the the view and provide the model with new data

modifier key: key on keyboard that modifies the action of another key when pressed together with it; e.g. Shift, Alt, Ctrl
modular design: design principle that divides a system into smaller parts called modules which can be independently created and then used in different systems

module: part of a program that can be used in other programs

momentum: hyperparameter of gradient descent that determines how much the previous update influences the current update

monad: design pattern in which pipeline implementation are abstracted by wrapping a value in a type; also a monoid in the category of endofunctors
Mongo, MongoDB: NoSQL database system

monoid: algebraic structure with a single associative binary operation and an identity element
monorepo: repository that contains more than one project; e.g. both backend and frontend of a web application
monospace: type face in which all characters have the same width

morpheme: smallest meaningful unit of a language

morphism: structure-preserving mapping from one mathematical structure to another
morphological analysis: process of breaking words down into morphemes or assigning parts of speech to them
morphology: branch of linguistics that studies the structure of words

motherboard: main printed circuit board in a computer that holds many of the crucial components of the system as CPU, memory, connectors for hard drives, USB ports, etc.

mouse: pointing device that detects 2D motion relative to a surface

mouseless computing: human-computer interaction technique that allows users to interact with a computer without using a mouse

multicore: CPU that contains more than one core

multiple dispatch: process of choosing the right method to apply based on the number of arguments and their type; in OO programming languages the choice is based on the first argument, e.g. in Julia it’s based on the signature of the method

multiplication: arithmetic operation

multiprocessing: parallel computing technique in which a program is divided into multiple processes which are executed simultaneously on different CPUs

multithreading: parallel computing technique in which a program is divided into multiple threads which are executed simultaneously on different CPUs

mutability: property of a data type that can be changed after it has been created

mutation: operation which alters the data (e.g. in a database)

mutex, mutually exclusive flag: object (lock) which prevents multiple threads from accessing the same resource at the same time

mutt: text based email client
mux, multiplexer: function or program allowing several input signals to share one device or resource; a multiple-input single-output switch
MX record: DNS record that specifies a mail server responsible for accepting email messages on behalf of a domain
MySQL: open source relational database management system

name binding: association of data (or code) with identifiers

namespace: set of unique names used to identify and refer to objects
namespace packages: way of splitting code of one Python package into separate directories or repositories
NaN, not a number: special floating-point value that represents an undefined or unrepresentable value
NAS, network attached storage: storage that enables multiple users and heterogeneous client devices to retrieve data from a centralized disk

NAT, network address translation: method of mapping IP address space onto another by modifying network address in the IP header
natural language: language that has evolved naturally in humans; e.g. English, French, Chinese, etc.
natural language processing, NLP: field of computer science that deals with the interaction between computers and humans using natural language; also a subset of AI
NEF, Nikon Electronic Format: image file format used by Nikon cameras
negation: logic operation that takes a boolean and returns the opposite boolean
negative caching: DNS caching of negative responses
Neo4j: graph database

neomutt: fork of mutt actively maintained (as of 2023)

Neovim: fork of Vim with better support for plugins
NetBeans: IDE for Java

network: group of computers connected together

neural network: machine learning model inspired by the neuron model

neuron model: mathematical model of a biological neuron with inputs, weights, bias, activation function and outputs

newline: special character or sequence of characters that is used to represent the end of a line of text and the start of a new line

newsgroup: discussion group on the Internet

Next.js: React framework

NFS, network file system: file system over network

Nginx: HTTP server
NLTK, natural language toolkit: Python library for natural language processing

node (graph theory): vertex in a graph

node (parallel computing): computer in a cluster that is dedicated to performing computation

Node.js: JavaScript runtime built on Chrome’s V8 JavaScript engine

nondeterministic finite automata: type of finite automata where each state can have multiple transitions for the same input

nonterminal, non-terminal: symbol that can be replaced by a sequence of other symbols

NoSQL, not only SQL: database system to store non-tabular data: key-value, documents, graph (nodes and vertices)
Notepad: simple text editor available in Windows by default
Notepad++: open source text editor for Windows based on Scintilla

npm, node package manager: package manager for Node.js

NREN, national research and education network: specialised internet service provider dedicated to supporting the needs of the research and education communities within a country

NS record, name server record: DNS record that specifies the authoritative name servers for a domain

NSFW, not safe for work: content (e.g. a comment on a social network) which should be looked at in private as it may contain offensive or indecent material
nsstring: string class in Objective-C

NTFS, new technology file system: proprietary journaling file system developed by Microsoft
null: value that represents no value or a value that is not defined

null pointer: pointer that does not point to any object
null pointer exception: exception that occurs when a program attempts to access a null pointer
nullish coalescing operator, ??: logical operator that returns its right-hand side operand when its left-hand side operand is null or undefined, and otherwise returns its left-hand side operand

NumPy: Python library for efficient matrix operations

NVIDIA: company that designs and manufactures GPUs

OAuth, open authorisation: open standard for authentication and authorisation

object: data structure that contains data and functions to operate on that data

Objective-C: object oriented programming language developed by Apple for macOS and iOS

observability: measure of how well internal states of a system can be inferred from knowledge of its external outputs

Ocaml: functional programming language

ODBC, open database connectivity: standard for database access

Office 365: cloud service by Microsoft that provides access to office software

Office, MS Office: suite of office software developed by Microsoft

onclick: HTML attribute that specifies a JavaScript function to be executed when the element is clicked

ontology: formal representation of knowledge as a set of concepts within a domain and the relationships between those concepts

OOM, out of memory: situation when a program tries to allocate more memory than is available

OOO, out of office: situation when a person is not available (e.g. on vacation)

OOP, object oriented programming: programming paradigm based on the concept of objects and their interactions

OOV, out of vocabulary: word that is not in the vocabulary used by a language model; usually replaced with <unk> or some other special token

open access: publication model that makes research outputs freely available to the public
open addressing: hash table collision resolution technique where the hash function is used to find an alternative location in the table

open source: software whose source code is available to the public under a license that specifies conditions of use and modification

Open Type, OTF: font format

OpenAI: non-profit AI research company

OpenCV, open computer vision: open source computer vision and machine learning software library
OpenOrca: huge dataset of questions and answers
OpenShift: container orchestration platform by Red Hat

operand: value on which an operator acts

operation: function that takes one or more operands and produces a result, e.g. addition, subtraction, multiplication, division

operator: symbol that represents an operation

optimizer: function that minimizes the loss function by adjusting the weights of a model

or: logil operator that returns true if either of its operands is true

Oracle: database management system and a company that develops it
Oracle Cloud: IaaS and PaaS by Oracle

orchestration: process of automating the deployment, management and scaling of containers

order of magnitude: power (factor) of 10

ORM, object relational mapper: technique for converting data between a relational database and objects in an object oriented programming language
OS X: Unix based operating system developed by Apple

OS, operating system: software that manages computer and provides common services for applications run on the computer
Outlook: email and calendar software developed by Microsoft
overfitting: when a model starts to memorize the training set instead of finding generalizable underlying patterns in the data
overflow: when a number is too large to be represented by the data type used to store it
overlay file system: file system that stores files on another file system

Overleaf: SaaS for LaTeX

package manager: program that automates the process of installing, upgrading, configuring, and removing software modules

packaging: the process of preparing software for distribution

packet: unit of data that is routed between an origin and a destination on a network

page fault: when a program tries to access a page that is not in the main memory

pagination: technique for dividing a large dataset or document into smaller parts (called pages)

pair programming: software development technique where two programmers work together on the same task usually on one computer

PAL, phase alternating line: video encoding system used in Europe
palindrome: string that reads the same backward as forward

PaLM, pathways language model: large language model from Google AI with 540 billion parameters

Pandas: Python library for data analysis

parallel computing: type of computation where many calculations are performed simultaneously

parameters: sometimes used interchangeable with weights

parent class: class that is inherited from

parser: program for parsing

parsing: procedure that analyzes the structure of a text according to the rules of a formal grammar

partition: contiguous block of sectors on a disk that is treated as a separate unit

partition table: table that stores information about the partitions on a disk

Pascal: imperative programming language designed in 1968 for teaching purposes
pass: the standard Unix password manager

pass by reference: method of passing arguments to a function where the function receives a reference to the original variable instead of a copy of its value

pass by value: method of passing arguments to a function where the function receives a copy of the original variable instead of a reference to its value

password: string of characters used for authentication

password manager: software for securely storing passwords and other credentials
PATA, parallel AT attachment: bus interface for connecting mass storage devices, predecessor of SATA

pattern: see regular expression
pattern matching: technique for checking if a string matches a given pattern
PayPal: company that provides online payment services

PC, personal computer: computer for personal use

PCA, principal component analysis: technique for reducing the dimensionality of a dataset by projecting it onto a lower dimensional space

PDF, portable document format: file format for storing documents in a device independent manner suitable for printing

PEMDAS, parantheses, exponents, multiplication, division, addition, subtraction: mnemonic for the order of operations in Python expressions

pentest, penetration test: security test of a software or hardware system by attempting to exploit its vulnerabilities
perfect hash function: hash function with no collisions

Perl: scripting language
permalink, permanent link: URL that points to a specific web page and does not change over time
persistent memory: memory that retains its contents even when power is lost
PERT, program evaluation and review technique: project management statistical tool for estimating the time required to complete a project developed by the US Navy in the 1958

petabyte: 10^15 bytes or 1,000 terabytes
PhantomJS: headless browser

PHP, PHP hypertext preprocessor, personal home page: programming language for creating dynamic web pages
pi, Ludolph’s number: constant–the ratio of a circle’s circumference to its diameter, approximately 3.14159
pickle: Python standard library module for serializing and deserializing objects

PID, process identifier: number used by an operating system to uniquely identify a process
pie chart: chart that shows the relative sizes of different categories in a dataset as slices of a pie
PII, personally identifiable information: information that can be used to identify a person
PIL: Python library for image processing
ping: Unix command for testing whether a host is reachable

Pinterest: social network for sharing images
pip: Python package manager

pipe: Unix inter-process communication mechanism that connects the standard output of one process to the standard input of another
pivot table: table that summarizes data from another table
pivoting: technique for rotating a table by turning the columns into rows
pixel perfect: type of design that is implemented with pixel level accuracy

pixel, px: single point in a raster image or the smallest addressable element in a display device
plain text: file format for storing text in some encoding

platform: software or hardward environment in which applications can run

platform as a service, PaaS: cloud computing service that provides a platform for running applications; e.g. Heroku

PLOS, Public Library of Science: open access publisher
plot: graph that visualizes data
plot: graph that shows the relationship between two variables
Plotly: data visualization library

plugin: software component that adds a specific feature to an existing program
PoC, proof of concept: (software) prototype that demonstrates the feasibility of a concept

pointer: variable that stores the memory address of a value

polling: technique for periodically checking for new data; e.g. a client periodically checks for new messages; see also busy waiting
polymorphism: programming language feature where a method can be used with different types
pop: operation that removes an element from a stack
POP3, Post Office Protocol version 3: protocol for retrieving email messages
popup: graphical user interface element that appears on top of the current window

PoS, part of speech: category of words in a natural language that have similar grammatical properties; e.g. noun, verb, adjective, adverb

post: message sent to a forum or a wegpage on a blog

postfix: notation where the operator is written after the operands; e.g. 3 4 +
PostgreSQL, Postgres, pg: relational database system with ACID compliance
Postman: software for testing APIs
PostScript, PS: page description language from Adobe
power law: relationship between two quantities where one quantity varies as a power of another
PowerBI: business intelligence tool from Microsoft
PowerPoint: presentation program and format from Microsoft
PowerPoint: sofware for creating presentations by Microsoft
PowerShell: shell and scripting language from Microsoft

precision: the fraction of true positives see also recall
predicate: function that returns a boolean value

predicate logic: logic that uses quantified variables and quantified predicates

prediction: output or result of a model
preemptive multitasking: multitasking technique where the scheduler can interrupt a task to run another task
prefork server model: server model where there is a central master process that manages a set of worker processes

preprocessor: program that processes a text file before it is compiled

pretrained model: model that has weights already trained on some other (usually larger) dataset and to be fine-tuned
Primefaces: Java library for creating web applications
printf: C function for printing formatted text
private method: method that can only be called from within the class where it is defined
private variable: variable that can only be accessed from within the class where it is defined

probability distribution: function that describes the probability of a random variable taking certain values

procedure: function that does not return a value

process: program in execution which has its own memory space

processor: hardware component that executes instructions

program: set of instructions for a computer to execute

programming language: artificial language for writing programs

programming paradigm: style of programming
progres bar: graphical user interface component for showing the progress of a task
Project Gutenberg: digital library of public domain books
Project Gutenberg: digital library of public domain books

project management: discipline of planning, organizing, and managing resources (time, money, people) to achieve specific goals

Prolog: logic programming language from the 1970s

Prometheus: software for aggregating metrics from various services; designed to run on one server
promise: JavaScript object representing the eventual completion or failure of an asynchronous operation
prompt (browser): JavaScript component for interacting with the user

prompt (ChatGPT): input from the user in a chatbot conversation

property: variable that is accessed like an attribute but is actually a method

propositional calculus: branch of logic that deals with propositions and their relationships

proprietary software: software that is not open source

protocol: set of rules for communication between two entities

protocol buffer, protobuf: binary file format developed by Google
Protractor: e2e testing framework for Angular
proxy server: server intercepting requests and forwarding them to the appropriate servers and then back to users with the goal of load balancing, security, and caching

pseudorandom number generator: algorithm that generates a sequence of numbers that appear random, but are actually deterministic
public domain: intellectual property that is not protected by copyright
public method: method that can be called from outside the class where it is defined
pull (git): operation in a git which fetches and merges changes from a remote repository

pull request: request to merge a branch into another branch; see also merge request
pumping lemma for regular languages: lemma that says that all sufficiently long strings in a regular language have a substring that can be repeated an arbitrary number of times to produce a new string that is also part of the language; the lemma is used to prove that a language is not regular
Puppet: configuration management tool
Puppeteer: Node.js library for controlling headless Chrome
push (git): operation in a git which sends changes to a remote repository
push (stack): operation that adds an element to a stack

pushdown automata: type of finite automata with a stack that can be used to store data

PWA, progressive web app: web app that uses modern web capabilities and APIs to deliver an app-like experience
PyCharm: IDE for Python

pypy: Python implementation written in Python
pySpark: Python API for Spark
pytest: Python testing library

Python: dynamically typed programming language

PyTorch: machine learning framework for Python
QA, quality assurance (software development): process for ensuring quality of a software
QA, question answering (machine learning): task of answering a question based on a given context

QEMU, quick emulator: virtualization software
Qt: cross-platform framework and toolkit for developing GUI applications
quadratic time: time complexity of an algorithm that is proportional to the square of the size of the input
quant: quantitative analyst
quantified self: movement of tracking personal data; e.g. health, sleep, etc.

quantifier: logic operator that binds a variable in a logical formula; e.g. ∀ and ∃

quantitative analysis: analysis of numerical data; e.g. stock prices, sales, etc.

query: request for information from a system

query language: programming language for querying data
query optimization: process of choosing the most efficient way to execute a query

queue: data structure that stores data in FIFO manner

quick sort: recursive algorithm for sorting sequential data

QWERTY: standard keyboard layout

QWERTZ: keyboard layout used in Central Europe

R, rlang: programming language for statistical computing

RabbitMQ: message broker

radix sort: sorting algorithm that classifies data based on the digits of the numbers from the least significant digit to the most significant digit and is efficient for sorting large sets of data with a limited range of possible values

RAG, retrieval augmented generation: using a relevant document prepended to the prompt to improve the results

RAM, random access memory: memory that can be accessed randomly
random forest: machine learning architecture that uses an ensemble of decision trees
random search: random combination of hyperparameters are used to find the best solution
Raspberry Pi: small computer used in IoT

raster graphics: representation of graphical components as a collection of pixels
raster image: see bitmap image
RAW: image file format storing unprocessed data from a digital camera

RDBMS, relational database management system: database and related utilities for storing and retriewing (querying) tabular data

React: JavaScript web framework developed by Facebook
React component: UI component in React
React hook: React API for managing state

React Redux: method to share global state in React web app
README (file): file that contains information about a software; usually part of its repository

real time: system that responds to events immediately

recall: the ratio of retrieved relevant items and all relevant items; also the ratio of true positive and true positive and false negatives
recaptcha, reCAPTCHA: CAPTCHA service by Google
recommendation engine: see recommendation system

recommendation system: system that recommends items to users based on their preferences; see also collaborative filtering

recursion: technique of solving a problem by solving a smaller part of the same problem

recursively enumerable grammar: TODO

Red Hat: company that develops Linux distributions and other software

Red Hat Enterprise Linux, RHEL: Linux distribution from Red Hat with long-term support, paid subscription and focus on stability
Redis: key-value store
redo: operation that repeats the last undo operation
Redshift: data warehouse service from AWS

reduce: higher-order function that takes a binary function, a starting value (accumulator) and a list, and applies the function to the accumulator and each element of the list

redundancy: fault tolerance technique where multiple copies of data are stored

refactoring: technique for improving the design of existing code without changing its behavior

register: small amount of memory in a CPU

regression model: model which predicts one or more numeric quantities e.g. a temperature or a location; see also classification
regression testing: process of testing a software after a change to ensure that the change did not break the software; it may compare the output of the software to the previous output

regular expression, regex, regexp: pattern for matching strings

regular grammar: formal grammar where all rules are of the form A -> aB or A -> a where A and B are non-terminals and a is a terminal
regular language: formal language that can be described by a regular grammar
regularization: technique for reducing overfitting; e.g. L1 and L2 regularization
release: version of a software that is ready for production

remote repository: repository that is stored on a remote server

repository, repo: decentralized storage location for version control system
resilient distributed dataset, RDD: data structure that represents a collection of records that can be partitioned across nodes in a cluster and can be operated on in parallel

ResNet, residual network: type of convolutional model
ResNet32: ResNet with 32 layers

responsive web design: method of designing web apps that adapts to different screen sizes
REST API: API built on HTTP 1.1 protocol based on the principles of REST

REST, representational state transfer: software architecture for web services
RESTful API: set REST API

reStructuredText: markup language for writing technical documents
return: keyword that returns a value from a function
RFC, request for comments: document proposing a standard

RGB, red green blue: color model where each basic color is represented by one byte
RGBA, red green blue alpha: color model where each basic color is represented by one byte and the alpha channel is used for transparency
ring buffer: data structure that is a queue with a fixed size that overwrites the oldest element when full

ripgrep, rg: command-line tool for searching files using regular expressions; cf. grep, ack

RISC, reduced instruction set computer: computer with a small set of simple instructions

RISC-V: open source RISC instruction set
RLHF, reinforcement learning with human feedback: technique for improving on pure LLM where a human (or a better model) picks a better answer from a selection

RNN, recurrent neural network: model architecture for modeling sequential data (e.g. language data)

roll back: undo a transaction or a deployment

root node: node with no parent

router (network): device that forwards packets between networks

router (web app): mechanism for mapping URLs to views

routine: set of instructions that performs a specific task in a program; generic term for function, method, procedure

RPC, remote procedure call: protocol that allows a program to cause a subroutine or procedure to execute in another address space (commonly on another computer on a network) without the programmer explicitly coding the details for this remote interaction

RPS, request per second: metric for measuring the performance of a server

RSA, Rivest–Shamir–Adleman: cryptographic algorithm for public key encryption

RSS, Rich Site Summary, RDF Site Summary, Really Simple Syndication: XML format for publishing updated content on web

RSVP, répondez s’il vous plaît: request for a response to an invitation (e.g. for a meeting)

rsync: command and TUI program for copying files

RTF, rich text format: file format for storing formatted text

RTFM, read the fucking manual: initialism and internet slang

RTL, right to left text: text (or script) written from right to left; e.g. Arabic, Hebrew
rubber duck debugging: method of finding a solution to a problem just by articulating it in spoken or written form

Ruby: dynamic open source programming language
ruby gem: TODO
Ruby on Rails: web framework written in Ruby

runtime: instructions executed while a program is running necessary for the proper execution of the code; e.g. garbage collector in Python

runtime code: code required to implement runtime system of a programming language

Rust: low-level programming language designed to be memory-safe

rustacean: mascot of Rust

S3, simple storage service: cloud storage service by AWS

SaaS, software as a service: method of software delivery and licensing in which software is accessed online

Safari: web browser by Apple
Salesforce: CRM platform and company that develops it

salt: random data used as an additional input to a hash function

sanity check: quick test to confirm that a system is working properly or that data is reasonable

SAP, systems, applications and products: software company developing ERP systems

SAS, statistical analysis system: software suite for advanced analytics

Sass: preprocessor scripting language which is transpiled into CSS

SAT, satisfiability problem: problem of determining if there exists an interpretation that satisfies a given Boolean formula

SATA, serial AT attachment: bus interface for connecting mass storage devices

sbt, Scala build tool: build tool for Scala

Scala: open source functional programming language that runs on the JVM

scanf: C function for reading formatted input

scanner: input device for digitizing documents and images

scheduler: process that decides which process to run next

scheduling: process of assigning resources to tasks

schema: database structure

Scheme: functional programming language

Scikit learn, sklearn, scikit: Python library for machine learning

Scintilla: open source code editing component

SciPy: Python library for scientific computing

scope: part of a program where a name binding is valid

Scrapy: Python library for web scraping

screen: display device for a computer

screen reader: program that reads text displayed on a screen used e.g. by visually impaired users

screencast: digital recording of computer screen

script: program written in a scripting language

scripting language: programming language used to manipulate, customize and automate existing systems; usually interpreted; e.g. AWK, Bash, Lua, etc.

scrum: agile project management framework focused on adaptability and iterative progress through fixed-length sprints

scrum master: person responsible for facilitating the scrum process

SCSI, small computer system interface: set of standards for physically connecting and transferring data between computers and peripheral devices

SDK, software development kit: toolikit for developing software for a specific platform

Seaborn: Python library for data visualization

search engine: software system that is designed to carry out search in a large collection of documents, e.g. web

sector: smallest unit of storage on a disk
security by obscurity, security through obscurity: security through secrecy of design or implementation; e.g. sharing a link with randomly generated URL

sed, stream editor: CLI Unix utility for text processing

seed, random seed: value for initialization of random generator function

seek: low level method for jumping to a position in a file

segmentation fault, segfault: runtime error caused by an invalid memory access
SELECT: SQL statement for querying relational database
selection sort, select sort: sorting algorithm that repeatedly finds the minimum element and puts it at the beginning
Selenium: Python library for web scraping
self: keyword in some programming languages that refers to the current object

self balancing tree: tree data structure that automatically keeps its height small to guarantee fast operations

self hosting: running a service on your own (virtual) server

semantic versioning, semver: convention for versioning software in the form of MAJOR.MINOR.PATCH where MAJOR number is incremented when backward incompatible changes are made, MINOR is incremented when new functionality is added in a backwards-compatible manner, and PATCH is incremented when backwards-compatible bug fixes are made
semaphor: synchronization primitive used to control access to a shared resource in the context of concurrency

semiconductor: material with electrical conductivity between that of a conductor and an insulator
sensitivity: see recall
SEO, search engine optimization: process of improving the visibility of a web page in search engine results
separation of concerns: general principle for [software development](#software development)
sequential search: technique for finding an element in a list by checking each element in order
serial port: hardware interface for serial communication

serialization: process of converting data structures or objects into a format that can be stored or transmitted

server: computer which is connected to a network and provides services to other computers
serverless: cloud computing execution model where the cloud provider dynamically manages the allocation of machine resources
servlet: TODO

set (data structure): data structure that stores unique elements
setter (OOP): method for setting a value of a property
setuptools: Python library for packaging

SGD, stochastic gradient descent: method for efficiently finding weights for a model

SGML, standard generalized markup language: markup language for defining document types

sharding: database partitioning technique where each partition is stored on a separate server
shared library: library that is loaded at runtime

Sharepoint: collaboration platform by Microsoft

shell sort: sorting algorithm that sorts elements far apart from each other and then reduces the gap between elements to be compared

shell, sh: program that interprets and executes commands entered by the user in a command line interface (or terminal)

Shift key: modifier key that is usually used to uppercase letters
Shopify: SaaS e-commerce platform

sibling: node that has the same parent

sigmoid: non-linear function with a characteristic S-shaped curve; see also logistic function
SignalR: library for ASP.NET for real-time web functionality

signed integer: integer that can be negative or positive

significance: measure of how likely it is that an effect is not due to chance

significand: see mantissa
SIM card, subscriber identity module card: a removable card used in mobile devices which securely stores IMSI
SIMD, single instruction, multiple data: parallel computing technique where a single instruction is executed on multiple data points

similarity search: search for similar objects in a database or dataset
Simula: the first object-oriented programming language which introduced objects, classes, inheritance, coroutines and garbage collection

single responsibility principle, SRP: principle in which a module should be responsible to one and only one actor each class should have only one reason for rewriting
singleton: design pattern where only one instance of a class can be created
Siri: voice assistant by Apple

Slack: commercial software for instant messaging
Smalltalk: early object-oriented programming language

smart contract: program that runs on a blockchain and can be used to automatically enforce an agreement
SMS, short message service: protocol for sending text messages over mobile networks
SMTP, simple mail transfer protocol: protocol for sending emails
snake case: naming convention for identifiers where words are separated by underscores; e.g. num_of_students
snap (Linux): package format for Linux
snapshot: copy of a database (or a dataset at a specific point in time
Snowflake: Tor pluggable transport
SOAP, simple object access protocol: protocol for exchanging XML messages over HTTP

social network: application that allows users to connect with each other over internet, share multimedia content, etc.

socket (operating system): file descriptor used e.f. for inter-process communication
socket.IO: JavaScript library for real-time web apps

softmax: non-linear function that exponentiates numbers and normalize them to sum into 1.0; used to make the layer output a probability distribution

software: complex program consisting of source code and data, UI and documentation
software 2.0: term coined by Andrej Karpathy to distringuish the traditional software development (manually writing a code) and the approach of machine learning—letting the computer learn on its own from training data

software design: process of defining the architecture of a software

software development: process of creating software, programming
SOLID: set of design principles used in object oriented programming; S for single responsibility principle, O for open-closed principle, L for Liskov substitution principle, I for interface segregation principle and D for dependency inversion principle.

sort, sorting: operation with data sequence which re-arranges its elements in ascending or descending order

sound card: hardware device that can play and record sound and convert between analog and digital signals

source code: see code

source control: see version control
SPA, single page application: web app that loads a single HTML page and dynamically updates the page as the user interacts with it

space: set of points
spam: unsolicited email messages

Spark: framework for distributed computing built on top of Hadoop
sparse matrix: matrix in which most of the elements are zero
sparse vector: vector in which most of the elements are zero

spawn: method of creating a new process from an existing one
specification: document that describes a system to be developed: its requirements properties, user interface, etc.

speech recognition: task of converting digial audio signal of speech into text
Sphinx: Python documentation generator built upon Jinja and Docutils
splice: method of inserting or removing elements from an array
spline: curve defined by control points in vector graphics
splint: static analysis tool for C

spreadsheet: software or a data structure for working with tabular data
Spring: Java framework

sprint: interval in agile development during which a team works on a set of tasks; usually 1–4 weeks long
spyware: malware that collects information about a user without their knowledge

SQL, structure query language: domain specific language for managing and manipulating relational databases
SQLAlchemy: Python ORM
SQLite: single file database system

SSD, solid state disk: storage device based on flash memory

SSH, secure shell: protocol for secure communication between two computers over an insecure network

SSL, secure socket layer: protocol for establishing an encrypted connection between a client and a server
SSO, single sign-on: mechanism for logging in to multiple web apps with a single set of credentials
SSR, server side rendering: technique of rendering a web app on the server and sending the result to the client as HTML

stack: data structure that operates in LIFO manner
stack overflow: bug that occurs when a stack is full and a new element is pushed onto it
stack trace: list of function calls that led to the current point of execution
StackOverflow: online platform for asking and answering questions about programming
stale link: hyperlink that leads to a web page that no longer exists

standard: specification that describes how a system should behave or be structured

standard input, stdin: input stream for reading data in Unix systems

standard output, stdout: output stream for writing data in Unix systems

statement: syntactic unit of a programming language that expresses some action to be carried out; e.g. x = 1 or INSERT INTO users VALUES (1, 'John')
static binding: see early binding
static memory allocation: memory allocation where the size of the allocated memory is known at compile time
static method: method that is not bound to an object
static site: website that doesn’t contain any dynamic content (e.g. JavaScript code)

static site generator: software that generates a static site from a set of templates in plain text
static type checking: a type checking that is performed at compile time

statically typed programming language: programming language that requires type annotations of variables
statistical model: model that describes a system using statistical methods

statistics: branch of mathematics that deals with data analysis

stochastic system: system that has some randomness in it

streaming: method of data processing where data is processed as it arrives and not as a whole
stress testing: type of testing that checks the system under heavy load

string: data structure for representing text

strong type system: type system that does not allow implicit type conversions
structural patterns: a group of design patterns that deal with the composition of classes and objects
structured analysis: software design method that uses data flow diagrams to describe a system
structured data: data that is organized in a certain way and can be easily processed
structured programming: paradigm that promotes the use of well-organized, modular code composed of functions or subroutines
subdomain: part of a domain; e.g. www in www.example.com
subprocess: process that is spawned by another process

subroutine: specific type of routine which help avoid code duplication; two types: function and procedure; the term emphasizes that it is a subordinated part of a program
substring: part of a string

subtree: tree that is a part of another tree

sudo, superuser do: superuser command in Unix systems that allows to run a command as another user
sum: result of adding two or more numbers
superclass: see parent class

supercomputer: computer with high level of performance

superuser: user with the highest level of privileges in Unix systems
supervised learning: machine learning task where the model is trained using labeled data

supremum: least upper bound

SVG, scalable vector graphics: XML-based file format for representing vector graphics
SVN, Subversion, svn: open source software for versioning and revisions of source code developed by Apache
Swagger: framework for documenting REST APIs
swapping: process of moving memory pages between RAM and disk

Swift: programming language developed by Apple

Swing: Java GUI toolkit
Swing: Java GUI toolkit
switch (statement): statement that allows to select one of many code blocks to be executed
Symfony: PHP framework
synchronization (multiprocessing): process of coordinating the execution of multiple threads or processes
syntactic analysis: see parsing
syntactic sugar: a syntactic shortcut that makes the code easier to read and write but does not add any new functionality; e.g. x++ vs. x = x + 1

syntax: set of rules that defines the combinations of symbols that are considered to be correctly structured programs in a given programming language
SysML, systems modeling language: UML for systems engineering

system call: request to an operating system to perform a low-level operation; e.g. opening a file, reading a file, etc.

systemd: Linux init system
systemd timers: systemd unit for scheduling tasks

table (database): collection of related data held in a structured format within a database
table (spreadsheet): cells organized into rows and columns
tableless web design: web design method that avoids using HTML tables for layout control

tabular data: data represented in the form of a table–in rows and columns
tabular model: model which predicts one column of a table based on data in other columns of the table

tabulator: whitespace character that advances the cursor to the next position usually located at every 4th or 8th column

TAC, traffic acquisition costs: money paid to other companies for using a company’s product; e.g. Google pays Apple to be their default search engine
Tailwind CSS: CSS framework

tanh, hyperbolic tangent: non-linear function used as activation function in neural networks
tar: Unix command for archiving files

TB, terabyte: 10^12 bytes or 1,000 gigabytes

Tcl: high-level, interpreted dynamic programming language
Tcl/Tk: GUI toolkit for Tcl

TCP, transmission control protocol: network protocol for reliable communication between computers on internet

TDD, test-driven development: software development in which unit tests are written before the code itself
Teams, MS Teams: software for communication developed by Microsoft

telephony: technology for voice communication
telnet, teletype network: network protocol for remote terminal access

templating engine: software that combines templates with a data model to produce documents
temporal decomposition: breadkign down a system or dataset into components or elements based on time; used in time series analysis

tensor: [multidimensional] array
TensorFlow: machine learning library developed by Google

terminal: program that provides a text-based user interface to an operating system; not to be confused with shell

terminal (NLP): symbol that cannot be expanded further
Terraform: software for managing infrastructure as code

tessellation: the covering of a surface using one or more geometric shapes (tiles) with no overlaps and no gaps; see also tiling
test set: part of the dataset which is used only for the final evaluation of a model

test suite: collection of test cases for testing

testing: process of evaluating a software

TeX: programming language for typesetting

text format: file format that stores data as plain text
Thanos: complex software for monitoring large and complex systems based on Prometheus
The Jargon File: glossary of slang terms used by programmers

thread: part of a process which executes instructions
thunk: piece of code that does some delayed work; in React Redux context thunks are a pattern of writing functions with logic inside that can interact with a Redux store’s dispatcher and other methods

TIFF, tagged image file format: file format for storing raster graphics

tiling: see tessellation

tiling window manager: window manager which automatically arranges windows on a screen (or multiple screen) in a tiling fashion

time complexity: measure of the amount of time taken by an algorithm to run as a function of the length of the input

time series: sequence of data points indexed in time order
timestamp: sequence of characters or encoded information identifying when a certain event occurred, e.g. 2023-01-01 00:00:00
TODO, to do: when used as a comment in source code it means that something needs to be implemented

token (natural language processing): in the context of LLM tokens are words or subwords

token (network): unit of data that is transmitted over a network
tokenization: process of splitting texts into tokens

TOML, Tom’s obvious minimal language: text format for configurations
tooltip: UI component in the form of a small pop-up window that appears when a user pauses the mouse pointer over an element

Tor: network for anonymous communication developed by the US Navy

TOTP, time based one time password: method for two factor authentication using a password which expires quickly

touchpad: input device for moving a cursor on a screen by sliding a finger on a surface

TPU, tensor processing unit: special processor designed for machine learning developed by Google; see also GPU

trackball: input device for moving a cursor on a screen by rotating a ball

trackpoint: input device for moving a cursor on a screen by pressing a stick

training: see fit
training data: see training set

training set: part of dataset used for fitting the model; doesn’t contain validation set

transaction: sequence of database operations which are executed as a single unit which can be cancelled in the case of failure

transfer learning: using a pretrained model for a task different to what it was originally trained for

transfer rate: a measure of the speed of data transfer

transformer: neural network architecture based on attention mechanism

translation (geometry): transformation that moves every point of a figure by the same distance in a given direction

transpilation: process of converting source code from one programming language to another; e.g. TypeScript to JavaScript

tree (data structure): data structure that is a collection of nodes connected by edges with uni
tree (graph theory): graph without cycles,

tree height: number of edges on the longest path between a root and a leaf
tree shaking: process of removing unused code from a bundle

tree traversal: process of visiting every node in a tree

true negative: case when a model correctly predicts a negative class

true positive: case when a model correctly predicts a positive class
trunk: the main branch in a source control system
trunk-based development: source control branching approach where developers work in a single branch called trunk/master/main, and do not use other long lived branches
truth table: table that shows the output of a logic gate or a boolean function for all possible combinations of inputs

TSV, tab separated values: text format for storing tabular data

TUG, TeX Users Group: a non-profit organization for people who are interested in TeX and LaTeX

TUI, text-based user interface: user interface based on text; a keyboard is used for input and a terminal for output
tuple: immutable finite sequence of items; 2-tuple is called ordered pair or couple or tuple, 3-tuple is called a triple (triplet)
Turbo Pascal: TODO
Turing machine: mathematical model of computation that defines an abstract machine which manipulates symbols on a strip of tape according to a table of rules
Twig: templating engine for PHP
Twilio: cloud communications platform as a service

Twitch: live streaming platform

Twitter: social network and microblogging service

two factor authentication, 2FA: more secure authentication which adds another layer of security with requiring a second independent way of authentication (with SMS or a mobile app or email
TXT record: DNS record that contains text information

type checking: process of verifying and enforcing the constraints of type system

type face: set of glyphs that share common design features
type signature: description of the input and output of a function

type system: set of rules that assigns a property called type to the various constructs of a computer program such as variables, expressions, functions or modules
type-0 grammar: see recursively enumerable grammar
type-1 grammar: see context sensitive grammar
type-2 grammar: see context free grammar
type-3 grammar: see regular grammar

TypeScript, TS: programming language which is a superset of JavaScript

Ubuntu: Linux distribution developed by Canonical
UDP, user datagram protocol: network protocol for unreliable communication between computers on internet

UI, user interface: way for humans to interact with a computer programs; can be graphical or text-based

UML, unified modeling language: modeling language for software and systems
underfitting: situation when a model is too simple to capture the underlying structure of the data

undo: operation which reverts the last action

Unicode: standard for encoding characters which provides a unique number for every character in most of the world’s languages
union: set operation which creates a new set that contains all the elements that are in at least one of the original sets
uniq: Unix utility which filters adjacent duplicate lines from standard input

unit test: testing unit which tests a subroutine
universal approximation theorem: mathematical proof that shows that a neural network can slove any mathematical problem to any level of accuracy

Unix: operating system developed in 1969
Unix domain socket: socket which is used for communication between processes on the same host
Unix philosophy: set of software design principles based on Unix which emphasizes modularity and reusability and simplicity
unsigned integer: integer that can only be positive; cf. signed integer
unstructured data: data that doesn’t have a predefined data model or is not organized in a predefined manner
unsupervised learning: machine learning technique where the model is not provided with the correct answers but only with the raw data
UPDATE: SQL command which updates rows in a table
upsert: database operation that updates an existing row if a specified value already exists (in a table) and inserts a new row if the specified value doesn’t already exist; portmanteau of update and insert
URI, uniform resource identifier: string of characters that unambiguously identifies a particular resource

URL, uniform resource locator: reference to a web that specifies its location on a network and a mechanism for retrieving it
usability testing: process of testing a software with real users to see how easy it is to use, related to UX

USB, universal serial bus: standard that defines the cables, connectors and communication protocols used in a bus for computer peripherals

Usenet: network of newsgroups

userland: part of OS which is not part of the kernel

userspace: a set of locations in memory which are used by processes and kernel modules

UTF16, unicode transformation format 16: variable width character encoding standard for UNICODE compatible with ASCII; it uses 16–32 bits per character; it’s more efficient than UTF8 for Asian languages

UTF8, unicode transformation format 8: variable width character encoding standard for UNICODE compatible with ASCII; it uses 8–32 bits per character

UUID, universally unique identifier: 128-bit number used to identify information in computer systems; it is usually displayed as a sequence of 32 hexadecimal digits separated into five groups by hyphens

UUID3: UUID generated from a namespace and a name

UUID4: UUID generated randomly

uvicorn: HTTP server for ASGI applications

uWSGI: HTTP server

UX, user experience: overall experience and satisfaction a user has when interacting with a software

V8: JavaScript engine

validation set: part of dataset for training a model which is put aside (usually 20 %) and used to measure the accuracy of the model; the rest of the dataset is called training set

variable: container that holds information; it has a name and a value which can be changed during execution of a program

variance: measure of how far a set of numbers is spread out; the average of the squared differences from the mean

VB.NET: programming language which is a part of .NET

VBA, Visual Basic for Applications: Visual Basic implementation used in Microsoft Office

VBScript: programming language

vector: mathematical object that has a magnitude and a direction; a tensor of rank 1

vector database: database that stores vectors and supports vector operations

vector graphics: representation of images as a collection of geometric primitives; as opposed to raster graphics

velocity: number of points completed per each sprint

version bump: increase of a version to a higher (semver) number; a dependency version can be bumped (up) as well

version control (system): software for managing versions of source code

vertex: point where two or more lines meet; see also node

vertical, vert, vertical format: text format where every token is on a separate line and linguistic information (e.g. PoS) is on the same line separated by a tab and structure elements (e.g. sentence and documents) are encoded as simple HTML elements

Vi: text based editor from the 70s

Viber: instant messaging and VoIP application

video conferencing: videotelephony for groups

video game: game played on a computer

videotelephony: telecommunication technology for the transmission of audio and video between two or more participants in different locations

Vim: text based editor from the 90s, an alternative implementation of Vi

Vim golf: competition to solve a problem using the fewest number of keystrokes in Vim

Vim script: programming language used to extend Vim

virtual assistant: software that can perform tasks or services for an individual based on verbal commands

virtual machine: program that emulates a computer system

virtual memory: memory management technique that allows an OS to compensate for physical memory shortages by temporarily transferring data from RAM to disk

virtualenv, virtual environment, venv: tool to create isolated Python environments

virtualization: method of running multiple OS on a single computer by using a hypervisor

virus: malware that replicates itself

VisiData, vd: TUI for exploring and arranging tabular data

Visual Basic: programming language

Visual C++: programming language from Microsoft

Visual Studio: IDE from Microsoft

Viterbi algorithm: algorithm for finding the most likely sequence of hidden states in a hidden Markov model

VMware: company that makes virtualization software for x86 architecture

VoIP, voice over IP: technology that allows making voice calls over internet connection using internet protocol

volatile: keyword in some programming languages

VPN, virtual private network: network that extends a private network across a public network

Vue.js: JavaScript web framework

W3C: consortium that develops web standards

WAN, wide area network: network which covers a large geographical area
WASM, web assembly: low-level bytecode for web browsers

waterfall development: software development methodology where each phase of the development process is completed before the next phase begins
Wayland: [communication protocol](#communication protocol) between a display server and clients; also a C library implementation of that protocol; intends to replace older X11

web application: software that runs in a web browser

web crawling: process of automatically browsing the web

web design: process of creating web pages
web driver: software component which enables browsers to communicate with web servers

web framework: framework for developing web applications

web page: document on the web
web ring: collection of websites linked together in a circular structure
web scale: term used to describe a system that is designed to handle a large amount of traffic

web scraping: technique of extracting data from web pages

web server: server which serves web content over HTTP
web service: software system designed to support interoperable machine-to-machine interaction over a network
web session: period of time during which a user interacts with a web app
web socket: protocol for full-duplex communication between a client and a server over a TCP connection
webhook: HTTP callback which is triggered by an event
Webkit: layout engine used in browsers
Webpack: JavaScript module bundler

website: collection of web pages with a specific URL

weight decay: regularization technique which penalizes large weights by adding a term to the loss function

weights: values which form another input to a model; they are usually in the form of matrices or tensors; see also parameter

WhatsApp: instant messaging software and service
wheel (operating system): user account with a wheel bit (additional special system privileges)

wheel (Python): replacement for eggs

where clause: clause in a SQL which filters rows

while loop: loop which executes a block of code while a condition is true
Whisper: speech recognition open source model by OpenAI
white space: non-printable character, e.g. space, a tabulator, a vertical tabulator

widget: GUI component

wifi, Wi-Fi, wireless fidelity: wireless communication protocol for LAN
Wikidata: collaboratively edited knowledge base
Wikipedia: collaboratively edited encyclopedia
Winamp: media player for Windows

window manager, window system: software that manages different parts of display screens

Windows: operating system by Microsoft

WIP, work in progress: unfinished work, something which is a subject of change in the future

wireless: communication protocol or technology over the air; e.g. wifi, bluetooth
Wix: WYSIWYG website builder platform
WLAN, wireless LAN: wireless LAN
Word, MS Word: word processor by Microsoft
word embedding: embedding of words

word processor: software for creating and editing documents in WYSIWYG fashion

word size: number of bits that can be stored in a register
WordNet: lexical database of English with ontology organized in a graph structure

WordPress, WP: content management system for web sites written in PHP
WPF, Windows presentation foundation: GUI framework for Windows
write-only code: code that is hard to read and understand;
WSDL, web services description language: XML-based language for describing web services
WSL, Windows Subsystem for Linux: compatibility layer for running Linux binaries on Windows

WWW, world wide web, web: system of interlinked hypertext documents accessible via the internet

WYSIWYG, what you see is what you get: text editor that shows the final output during editing
X (company): see Twitter
X Window System, X: see X11
X.org foundation: non-profit organization that develops X11 and other software

X11: windowing system for bitmap displays common on Unix systems

x86: CPU architecture developed by Intel
xargs: Unix command for building and executing commands from standard input

Xcode: IDE for macOS by Apple

XeTeX: TeX typesetting engine supporting Unicode and modern font format as Open Type

Xfce: lightweight desktop environment for Linux

XML, eXchange markup language: SGML markup language
XMLHTTPRequest: JavaScript API for making HTTP requests
XP, extreme programming: methodology that emphasizes close collaboration between developers and customers, continuous feedback, and practices such as test-driven development, pair programming, and continuous delivery
XPATH, XML path language: query language for XML documents
XSD: XML based language for describing the structure of XML documents
XSL: family of languages used to transform XML documents

XSLT, extensible stylesheet language transformations: XML based language for transformation of XML documents
Xubuntu: Linux distribution based on Ubuntu with Xfce as the default desktop environment
XXX, x-rated content: content which is indecent
Y’UV, YUV: color model found in the PAL analogue color TV standard

YAML, YAML Ain’t Markup Language: text format for representing structured data
YAML stanza: group of lines in YAML file
Yandex: Russian technology company
YouTube: video sharing platform
z index: property in CSS which sets the order of a positioned element on z axis

ZB, zettabyte: 10^21 bytes or 1,000 exabytes
zero day: type of vulnerability of software or hardware unknown to the vendor which has zero days to fix the issue
zero shot learning: machine learning method where a model is altered to predict classes it has not seen during training
zeroth-order logic: see propositional logic
Zig: low level programming language with memory safety features
Zigbee: wireless communication protocol used in IoT applications
zip: file format for compression
zlib: library for compression
Zoom: proprietary software for videotelephony
zstd: lossless data compression algorithm or the program implementing it
ZX Spectrum: 8-bit personal computer released in 1982

About

This glossary is WIP.

When a term is used in a definition it must link to its own definition.
Basic terms are not defined in this glossary: analysis, average, execution, finite, function, job, notation, operation, principle, sequence, system, tool, … only if they have a special meaning (e.g. method).
All non basic terms must be defined in this glossary using the most specific terms possible.
Abbreviations should be defined together with the full term.
All terms should be (proper) nouns.
Multiword terms should be included only if they don’t satisfy the compositionality principle. E.g. there is no point in having term Linux kernel when it can be deduced from the two terms Linux and kernel.
Avoid using hyphens in lexemes.
Only the first usage of a term within a definition is hyperlinked.
Proper noun lexemes should be external links.
Do not use links to Wikipedia, this glossary should be self-contained.
Polysemous terms should be defined separately and distinguished with a contextual term in parentheses. The contextual term should be the most general term.
The jumps between related terms (see) should be as short as possible.
Avoid using cyclic definitions.
Avoid using rare abbreviations in definitions.
The text after ; in a definition is not part of the definition. It may contain additional information about the term, examples and references to synonyms or related terms.
No determiners at the beginning of definitions.
Definitions should not contain external links.

published: 2023-06-30
last modified: 2023-12-01

#list #dev #code

https://vit.baisa.cz/notes/code/vocabulary/