Practice That Actually Builds the Skill
Git is a habit, not a body of knowledge. Habits need a project and about four weeks, and without them the whole of this book fades to a vague memory that there was something about branches.
So here is what to actually do, in order, starting today.
1. Put everything under Git today
Every project folder on your machine, including the messy ones and the ones you are slightly ashamed of.
cd whichever-project git init # write a .gitignore first git add . git commit -m "Add the project as it stands today"
The point is not that these projects deserve version control. It is that the habit only attaches to real work, and this is the real work you have.
2. Use branches even alone
One branch per change, merged and deleted. Yes, even for a ten-minute fix on a project nobody else will ever see.
Working alone is the safest possible place to practise, and the loop is what you are practising. By the time somebody is waiting on your pull request, you want the branch-commit-merge rhythm to be something your fingers do while you think about the actual problem.
3. Break things deliberately
This is the exercise most people skip and the one that pays best. On a throwaway copy of a project — clone it twice if you like — do each of these once, on purpose, while calm:
Commit something wrong and revert it. Commit too early and fix it with --amend. Create a conflict by editing the same line on two branches, then resolve it. Then create another one and --abort it instead. Delete a branch you should not have and find it again in git reflog.
Each of these takes two minutes by choice and half an hour by accident, and doing them deliberately is what turns Chapter 4 from something you read into something you have done.
4. Make one contribution
Find a tool you actually use, read its documentation until you hit a sentence that is wrong, unclear, or missing a step. Fix it.
Fork, branch, commit, push, pull request — Chapter 9's loop, nothing new. Your first pull request to a stranger is the largest single step in this book, and it is one afternoon.
Then go deeper — but not yet
There is a moment when the deep dive becomes worth starting, and it is not today.
It arrives when you find yourself wanting a straight-line history and knowing merging cannot give you one. When you are curious about what is actually inside .git. When somebody hands you a repository in a state you cannot diagnose. When a team asks you to rebase and you would rather understand it than follow instructions.
Those are the questions the deep dive answers. Have them first and the book takes half as long, because every chapter lands on something you already wanted to know. Read it before you have them and it is a list of features.
And keep going
Sandpiper is not finished. The club still wants a photo gallery, a contact page, and a layout that works properly on a phone. Every one of those is branch, commit, merge — the loop you now have.
That is the honest end of this book: not a finished skill, but a project you can carry on with, and a habit that makes carrying on cheap.
- "I should re-read the book to remember it." Six commands, used daily for two weeks, will stick better than three re-readings. This is a skill rather than a subject.
- "I need a big or serious project to practise on." A folder of notes under version control teaches the same loop as a large codebase. The size changes nothing about the commands.
- "Breaking things deliberately is a waste of time." It is the difference between meeting
reflogwhile calm and meeting it while panicking. Two minutes now against half an hour later. - "I should start the deep dive immediately to keep the momentum." It answers questions you have not had yet. Have them first and it takes half as long.
- Skills learned and not used disappear within a month, and this page is what decides whether the book was worth your time.
- Practising recovery deliberately is the only way to meet Chapter 4's commands while relaxed, which is the state in which they are easy.
- The sequencing advice protects the next step too: readers who arrive at the deep dive with real questions finish it, and readers who arrive without them do not.
Knowledge Check
What does this topic suggest doing first, today?
- Put every project folder on your machine under Git, with a
.gitignore - Start the deep dive, while the material from this book is still fresh
- Re-read the chapters you found hardest before touching anything else
- Build a new project specifically designed to practise Git on
Why does the plan include breaking things deliberately?
- So you meet each recovery command while calm rather than while you are panicking
- So you can find out which commands are dangerous by observing the damage
- So the reflog fills up with entries and becomes useful for later recovery
- So you learn to repair a repository that other people have already broken
When does this book say the deep dive becomes worth starting?
- When you have specific questions it answers, like a straight history or
.git - Immediately after finishing this book, while the material is still fresh
- After a year of professional experience, whatever questions have come up
- Only if a job requires it, since the daily loop covers everything else
What is the honest end of this book, according to this page?
- A project you can carry on with, and a habit that makes carrying on cheap
- A complete knowledge of Git, sufficient for any situation you will meet
- A portfolio strong enough to apply for a junior developer position with
- A finished skill that needs no further practice or reading of any kind
You got correct