Painfully Obvious.

Comments

1. Ryan Christensen Apr 16th, 2007 at 9:32 pm

   If anything it sure makes for good entertainment!

2. whatnot Apr 17th, 2007 at 12:51 am

   It’s obviously intended as a damning criticism and in no way should be construed as potentially humorous

3. Andrew Apr 17th, 2007 at 12:55 am

   @wntd: As if Mark ever uses one without the other.

4. whatnot Apr 17th, 2007 at 10:25 am

   It is a criticism, but it’s not so powerful as to warrant responses like:

   Mark, I wish you had thought twice before posting this one. While David may not be the programming savior of the world, he is very upfront about his faults. Your dissection of things he didn’t say says more about you than it does about him.

   This is just viciousness.

   And so on.

   It doesn’t help that the Twitter developer used P-word, and then insinuated it might be better than Ruby in a certain fashion. When used in the presence of Rubyists that has a unique ability to allow them to troll themselves. It’s like shooting fish in a barrel, if the fish attempted to leap out of the barrel and pull the trigger for you.

   It also doesn’t help that DHH conflated optimization with scalability, and then pulled out (excuse me, had someone else pull out) a solution to the multiple-databases problem, when from the interview it’s clear that it’s not just the database interface but large parts of Rails and Ruby itself, and the 75 LOC “response” is only a partial solution.

5. Andrew Apr 17th, 2007 at 3:55 pm

   It doesn’t help that the Twitter developer used P-word, and then insinuated it might be better than Ruby in a certain fashion.

   I’m the first to admit how easy it is to get a Rubyist riled up. But I find the Ruby/Python language war no less tiring than language wars in general. I generally agree that people should just calm the hell down about this.

   It also doesn’t help that DHH conflated optimization with scalability, and then pulled out (excuse me, had someone else pull out) a solution to the multiple-databases problem, when from the interview it’s clear that it’s not just the database interface but large parts of Rails and Ruby itself, and the 75 LOC “response” is only a partial solution.

   Well, DHH can’t solve the “Ruby is slow” problem by himself, and that wasn’t really what Alex was asking for. Anyone who’s worked with Rails knows that you’re making a trade-off up front — more hardware to compensate for fewer LOC. (I’m not going to get into whether that’s a worthwhile gambit, since smart people can disagree on this, but it’s a gambit that any Rails user chooses to make.)

   Put more bluntly, I think that any Rails optimizations to Rails would be a drop in the bucket compared to the larger issue: Ruby is fucking slow, and anyone who denies it (or tries to sidle around it with rhetoric) is deluded. There are people working on making Ruby faster.

   So the general speed issue has been talked about before. Alex was specifically griping about the lack of a multiple-database interface to Rails — the absence of which doesn’t really surprise me, since Twitter is the first Rails site that has needed to scale to the point where the database is the bottleneck. The 75 lines of code someone else wrote… they’re beside the point, I think. He was griping about the fact that someone else hadn’t written code to solve a problem that nobody had encountered yet.

   Now, having said all that, I think this whole thing was blown way, way out of proportion, and that Alex was just venting to a person that happened to be writing down everything he said.