Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Three libraries, a framework and an API - how ContentTagger got built in 7 hours (jaggeree.com)
25 points by danw on April 3, 2009 | hide | past | favorite | 2 comments


I still meet people, both in real life and most recently in #startups, who don't use any sort of 3rd-party libraries or frameworks; they hand-code everything. I've been trying to explain for years how completely ridiculous that is for 95% of use cases, but still people fight me.

This is a great example of taking a bunch of existing tools and data then stringing them together to create something new. Granted, we don't have "The next Google" here (whatever that means), but it seems like people who still roll their own everything rationalize it by saying things like "it makes it more flexible" or "blah isn't scalable enough" or "I don't really like the conventions".

These are all pretty stupid reasons. Chances are, you're not going to need that flexibility anyway; if you can't name a half dozen things X doesn't do that you need it to, just use it. You probably won't ever have to worry about scaling, so forget it. And if those conventions save hundreds of hours, who cares?

We have numerous languages now, Python especially, where there is probably a library for anything you need to do. There might even be 3rd-party APIs for all the data you need. Embrace the work that others are doing for you: stop coding and start stealing!


Personally, I think your attitude is laughable. Why should I use 3rd-party APIs? I don't even have the time to learn them, because I'm too busy debugging by own private programming language, tool chain, and operating system. I'm surprised I'm even posting this on the internet that everyone uses, as opposed to my completely solipsistic one-person internet. It's really heaven, trust me.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: