Concept · Tech Jargon
Software development jargon decoded: repos, builds, releases and sprints
How software is built, tested and delivered, and what the words in a project update actually mean.
Beginner6 min read

Key takeaways
- Most development jargon describes how a team keeps its work organised, tested and safely released.
- Version control is the memory of a project, and a pipeline is its assembly line.
- Staging is the rehearsal, and production is the live show.
- Words such as agile, sprint and technical debt describe habits you can ask about in plain terms.
Project updates from a software team can read like a different language: “we merged the PR, the pipeline is green, and we will cut a release to staging before the sprint ends.” Behind the jargon is a simple, sensible routine for building software safely with several people. Once you see the routine, the words make sense.
This article follows a small change from an idea to a live release, naming the pieces on the way.
Keeping the history: version control
- Version control. A system that records every change to the code, who made it and why, so that nothing is lost and mistakes can be undone. Git is the most widely used.
- Repository (repo). The project’s folder, together with its entire history.
- Commit. A saved set of changes, with a short message describing it.
- Branch. A separate line of work, so that someone can try something without disturbing the main version.
- Merge. Combining a branch back into the main one.
- Pull request (PR). A proposal to merge a branch, which colleagues review before it is accepted.
- Conflict. When two people changed the same lines and the system needs a person to decide.
- Fork. A personal copy of someone else’s repository.
Writing the code
- Library. A bundle of ready-made code for a particular job, such as drawing charts.
- Framework. A larger structure that tells you how to organise your whole application.
- Dependency. Another piece of software that yours relies on. Updating or losing one can break your project.
- Open source. Code that is published for anyone to read, use and often modify, under a licence.
- Refactoring. Tidying the inside of the code without changing what it does.
- Technical debt. The accumulated cost of shortcuts. Like financial debt, it is fine in moderation and ruinous if ignored.
- Legacy system. Old software that still matters, and is hard to change.
Checking the code
- Bug. A mistake in the software.
- Testing. Running the software to see whether it does what it should.
- Unit test. A small, automatic check of one piece. Integration tests check that pieces work together, and end-to-end tests simulate a user.
- QA (quality assurance). The practice of finding problems before users do.
- Code review. A colleague reading your changes before they are accepted.
- Regression. Something that used to work and broke after a change.
- Linting. Automatic checks for style and common mistakes.
The assembly line: building and delivering
- Build. Turning source code into something that can be run.
- CI (continuous integration). Automatically building and testing the code every time someone makes a change.
- CD (continuous delivery or deployment). Automatically preparing, or even releasing, tested code.
- Pipeline. The automated sequence of steps from change to release.
- “The pipeline is green”. All the automatic checks have passed. “Red” means something failed.
- Artifact. The built, packaged output of a pipeline.
- Container (such as Docker). A sealed package of software and everything it needs, so that it runs the same everywhere.
Environments and releases
- Development (dev). Where programmers work and experiment.
- Staging (or test). A close copy of the live system, used as a rehearsal.
- Production (prod). The real, live system that users depend on.
- Deployment. Putting a version into an environment.
- Release. A version made available to users. Version numbers such as 2.4.1 usually signal how big the change is.
- Rollback. Returning to the previous version if the new one causes problems.
- Hotfix. An urgent fix released outside the normal schedule.
- Feature flag. A switch that turns a new feature on for some people, or off, without a new release.
- Downtime and maintenance window. The planned period when a system is unavailable for work.
How teams organise the work
- Agile. A way of working in short cycles, with frequent feedback, instead of one long plan.
- Sprint. A fixed work period, often one to two weeks, with an agreed goal.
- Backlog. The ordered list of work that is waiting.
- User story. A short description of a need from a user’s point of view: “As a field officer, I want to see my villages on a map”.
- Stand-up. A very short daily meeting to share progress and blockers.
- Retrospective. A meeting at the end of a sprint to discuss what went well and what to change.
- Scrum and Kanban. Two popular ways of organising agile work.
- MVP, PoC and prototype. Increasingly complete early versions, used to learn before building fully.
- Roadmap. The longer-term plan of what will be built.
Words that mean different things to different people
- “Done”. Written, tested, reviewed, deployed and documented, or just written? Agree a definition of done.
- “Quick fix”. Often a shortcut that creates technical debt.
- “It works on my machine”. It runs for the developer, and may fail elsewhere. It is the reason environments and containers exist.
- “We will refactor later”. It may never happen. Ask when.
Questions to ask your software team
- Where is the code kept, and who owns that account?
- How are changes reviewed and tested before they reach users?
- How often can we release, and how do we undo a bad one?
- What does “done” mean for this project?
- How much technical debt do we carry, and what is the plan for it?
- Is it documented well enough that someone new could take over?
Frequently asked questions
- What is the difference between a library and a framework?
- You call a library when you need it. A framework calls your code, and sets the structure. A library is a tool, and a framework is a template.
- Why are releases not instant?
- Because testing, review and careful rollout protect users from mistakes. Automation makes frequent, safe releases possible.
- Is agile just a way to avoid planning?
- No. Done well, it plans continuously, in small steps, and adapts to what is learned. Done badly, it can become disorganised.
- What is DevOps?
- A set of practices that bring development and operations together, so that software is built, released and run as one smooth process.
Software jargon describes how people keep a complicated thing under control. When you hear a word, ask what problem it solves: remembering, checking, packaging, releasing or organising. The word will nearly always map to one of them.
Join the conversation
Reactions & comments
Comments are opening soon.
In the meantime, tell us what you think by email at info.istem.lab@gmail.com.



