5 ms·
"Let's try an experiment. Think of a project you use all day. Maybe it's Rails or Python or something. Now, name 4 people on the core team without looking them
by plinkplonk 16y ago
"Let's try an experiment. Think of a project you use all day. Maybe it's Rails or Python or something. Now, name 4 people on the core team without looking them up. I can't do that for anything I use. Alright, let's say you can do that. You know a myriad of things about the people who make your tools, but can you honestly say you know as much about them as you do about the tools they made you? Be honest with yourself and really look at how much you know about the people behind your gear as you do about the gear itself."
This is very bizarre. Isn't this true for all tools/man made artifacts we use? I have no idea who exactly designed my car, my guitar , my cellphone, or even the apartment I live in. And when I do know their names I certainly don't know them better than I do my tools. Why should I want to?
"I still have to do programmer interviews like everyone else. No matter how much code I put out, I still have to solve stupid puzzles about coconuts and manholes. No matter how many web servers or email frameworks or database servers or chat servers or assemblers I write I still have to prove I can code. No matter how many copies of my software get deployed I still have to prove I can make reliable software."
I wouldn't want to comment on what Zed's personal experience is , but I know many programmers who wouldn't have to "prove that they can code". "Fame" sees to correlate inversely with having to jump through hoops. I doubt if anyone really wants to test Linus (or Carmack or DHH or any other "famous" programmer) for "ability to code". Outside the MegaCorp, and especially in startups, a reputation for Open Source contributions helps you avoid stupid questions/tests etc, at least in my limited experience. It is often the undistinguished guy with the undistinguished cv that has to go through the technical nitpick interview.
I think this experience may be somewhat unique to Zed. No harm in that of course. Just pointing out that is far from universal.
Just my perception, but Zed seems to get weirder with every post he writes. I don't mean that he is crazy or anything, just that the logic in his posts seems increasingly frayed.
- lelele 16y agoOTOH, many people think that every cool innovation from Apple comes straight out from Steve Jobs' brain.
- troygoode 16y ago"This is very bizarre. Isn't this true for all tools/man made artifacts we use? I have no idea who exactly designed my car, my guitar , my cellphone, or even the apartment I live in." I think you're making part of Zed's point - better than he did, in fact. There are no famous car/guitar designers, cell phone engineers, or construction workers either. People like Linus, Carmack, and DHH aren't famous because of their code - they're famous because of their products. Of course then he dives off into this strange diatribe about being forced to interview the same way we peons have to despite his previous works - nevermind the fact that most of the time there is about a 0% chance the hiring manager has read the code to mongrel (most of the devs using it haven't either, for that matter). I also think he is wrong on this latter point as - even though I'm completely non-famous - my meager contributions to the open source world have opened a lot of doors and given me more credibility than I otherwise would have had. "Just my perception, but Zed seems to get weirder with every post he writes." I'll agree with you on that one.
- stcredzero 16y agothe fact that most of the time there is about a 0% chance the hiring manager has read the code to mongrel Just wait a minute here! Isn't this weird? It's like hiring designers and refusing to look at their portfolios. The interview process puts people through all these weird contortions so we can (among other things) get some indirect indications of how they code. Why not just read their code? We want to know if they can collaborate on a project, why not look at the result of projects they collaborated on? I think in the corporate world at least, programming has become a commodity.
- troygoode 16y ago"Why not just read their code?" I agree. Unfortunately I think interview processes are designed to accommodate the majority of applicants - and in the corporate/enterprise world, the majority of applicants have zero open source work. I think/hope this will change over time.
- vijaydev 16y agoI highly doubt that hiring managers are going to have the time to read code. Very few people go through a candidate's code before an interview.
- Quarrelsome 16y ago"Just my perception, but Zed seems to get weirder with every post he writes." I don't think this one is that weird. It is a logical extension of the rant about not getting paid: http://zedshaw.com/blog/2009-07-13.html http://zedshaw.com/blog/2009-07-13.html when he found out how many people were using Mongrel commercially without even putting him in the credits.
- evilduck 16y agoWhich is a childish rant (like most of his posts). He chose a license that explicitly permitted what he's complaining about, and then cries that what happened within his own terms of usage...isn't what he had hoped for. If you want money for your code, you explicitly ask for it via an employment agreement or a software license. He's weird because he's obviously intelligent, due to the code he's demonstrated, but at the same time, he fails so hard at anticipating the obvious repercussions of his actions (i.e. his "Rails is Ghetto" debacle he's trying to pass off as a joke post).