Overall, it’s not perfect, but I suspect that it will continue to improve over time. Sometimes the quality hasn’t been there either - it may have worked, but it wasn’t clean. On more than one occasion the suggestion that it’s given me has been bizarre. This is true of code you find anywhere though! You still need to understand what’s it doing. Sometimes the code it suggests is off, even by a subtle name change. This may not totally be a bad thing - but you shouldn’t just blindly trust what it suggests. Here are some things that I’ve noticed about Copilot that aren’t as good: You Spend More Time Reviewing Code It doesn’t have suggestions for everything I write, which is better than it trying to fill in code that I don’t need. I’m using Copilot in Visual Studio Code with Python - and I’ve found it’s been very easy to use and that its previews are easy to review without being overbearing or obnoxious. You know that feeling when you know what you need to do, but you can’t quite remember the best way to do it or the syntax of how it starts? Copilot has helped with that! The logic is still all mine, but it’s kept me from having to Google syntax more than once. Copilot has been a great help at writing entire easy functions that help me keep my code clean and neat. Sometimes, you need boilerplate code - or code that is very similar to code that you’ve already written. Here are my favorite things about Github Copilot so far: Good at Boilerplate However, when it launched in late June, I immediately noticed the interest it stirred up - both on Twitter and within my personal network of developers, so I decided to give it a try. I thought that it sounded rather gimmicky - like a hyper autocomplete. I have to admit that I wasn’t that excited about Github Copilot when I first heard about it. Photo by Roman Synkevych □□ on Unsplash
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |