Steve "spez" Huffman of Reddit on The Total Rewrite Decision | Hacker News Reader

Steve "spez" Huffman of Reddit on The Total Rewrite Decision

reddit.com

22 pointsstaunch9 comments

I don't get it. I can't seem to find any reason why they wanted to rewrite it, and that's interesting in and of itself.

Given the language change of the past, and now a complete re-write, neither of which had any major impact on the end user, I'm wondering if we're not witnessing some sort of infatuation with coding for the sake of coding over there.

I can think of a myriad of reasons why one would want to re-write code, but they're certainly not sharing theirs.

>the old code was essentially two years of hacks and cludges. The new version is much cleaner, and the new back-end is much more malleable. The idea is that once we get caught up with ourselves, the actual new stuff will come pretty quickly.

+1 for reading TFA. More of that FTW.

>the old code was essentially two years of hacks and cludges.

That basically describes every piece of software ever written. It's not always a justifiable excuse to re-write the codebase though.

Chris's comment is interesting...

http://reddit.com/info/2h8kd/comments/c2hccv

So I guess Reddit is now on Pylons, using Mako as the templating engine.

Yep, I'd say that's a pretty good choice. Pylons+Mako+SQLAlchemy makes an excellent framework.

Raldi's comment is hilarious.

A rewrite only tastes sweet after a long process of freaking out that you just threw away all your code.

And since time is the most precious resource in a startup then rewriting is the worst decision you could make, if that logic is correct.

However, Reddit isn't really a startup anymore, so they can probably afford it.

definitely, they may (should) be looking at what they will need back-end wise for the next few years. if building on a dodgy base is too risky, you have to build the better foundation. they're not a start-up anymore, it's a new set of challenges

Reply on news.ycombinator.com