Live data from Hacker News

Trouble with Diaspora

blog.steveklabnik.com

161–166 of 166 posts

Re: Trouble with Diaspora

#161

This is the problem: college students are terrible programmers. There aren't enough consequences for writing bad code in college. In industry, you learn very quickly that everything you learned in college is minuscule compared to what you actually need to know to work. I knew guys who only studied databases or only studied HTML+JavaScript+CSS. And we all expected that this level of specialization was common and even…

This all sounds very wise...until you remember that Facebook itself was written by Zuckerberg when he wasn't yet out of college. Linux was written by Linus Torvalds when he was in college. Firefox was written by Blake Ross when he was a high school stduent. Chatroulette was written by Andrey Ternovskiy, another high school student.

Some people are just really good. And a startup makes them even better, and much faster than if they had decided to be "apprentices" or "interns" forever.

The thing about Diaspora is that UNLIKE Facebook or many other startups, its source code is on the internet for all to see. Do you think the original Facebook (written in PHP) didn't have security holes that would be transparent upon reading the source code? For that matter, do you think today's Facebook source would withstand attack if posted on the internet tomorrow?

The question answers itself.

If it achieves its potential, Diaspora will be hardened to a considerable extent because of the scrutiny that open source provides. This is not the time to slam these kids as wet behind the ears undergrads. Their code is pretty good for something written in 3 months. And if you really are that much wiser than them, they are accepting patches.

Re: Trouble with Diaspora

#162
post #19
post #17

"Release early, release often" (an open source mantra). Here on HN i also find the 'getto launch' being preached -- "if you not embarrest by your product at lauch-time, you should have released earlier". Judging from the noise their release makes here on HN i think the diaspora guys did well opening up their repo 'early'. And to steve: i think you hold 'professional programmers' (which i interpret as programmer that…

There are a few reasons why this is worse: first, they are launching to media attention and a rabid community. BCC sucked at launch, but no one saw it, so yay. (And by sucked, I mean looked ugly, not "Anyone can delete all your documents at will.") Diaspora has thousands of end users already. Some seeds have 200+. They are launched. Not prealpha. Launched. Their entire reason for existing is "Facebook but private". A…

i think you miss that diaspora is an open source project. once they release a 'developer preview' people _can_ actually use go an use it. i would not call that 'launched' or '1.0' (which is the open source analog to launching in my opinion). sitting on top of yr code will not spawn discussion, attract devs/tester to have a look. therefor the opensource mantra is release early+often.

it took me 2 minutes to find out what BCC means. for those also wondering: http://www.bingocardcreator.com/ sorry but BCC is not open source so i call it a bad comparison. commercial software is different. the thousands of diaspora end-users you talk of are probably quite aware they are guinnee-pigs of a facebook counter project.

"facebook but private", sorry but again i dont think you get diaspora as it is more like "facebook but distributed".

calling diaspora a "flop" is like calling mozilla before 1.0 a flop: it's open source (something you dont seem to get), its out there for _anyone_ to play with freely, to collaborate on, to learn from...

Re: Trouble with Diaspora

#163
post #70

Okay first of all, I'm glad that a bunch of undergrads from my school were able to raise $200k, get a lot of press, and build something. This alone should get a community of people around the project fixing bugs, etc. I've always been saying that making a distributed social network is much easier than "solving" privacy and security for such a thing. First of all, try even defining what it means to privately share thi…

Is your school specializing on building secure distributed social networks?

Re: Trouble with Diaspora

#164

This is the problem: college students are terrible programmers. There aren't enough consequences for writing bad code in college. In industry, you learn very quickly that everything you learned in college is minuscule compared to what you actually need to know to work. I knew guys who only studied databases or only studied HTML+JavaScript+CSS. And we all expected that this level of specialization was common and even…

Every programmer I've known thought he was a super hacker by the time he got out of college. Wasn't Linus Torvalds a college student when he released linux? For that matter, Facebook was written by a college student. Some college students can do amazing things. Also, this is just a pet peeve of mine, but don't diminish college students by calling them "kids." Many may be young, but they are adults.

Yes, but an advanced graduate student, and Linux was his thesis project. Though he was ~21-23 at the time, the experience and maturity level was indubitably higher than the average CS undergrad in American universities.

Re: Trouble with Diaspora

#165
post #70

Okay first of all, I'm glad that a bunch of undergrads from my school were able to raise $200k, get a lot of press, and build something. This alone should get a community of people around the project fixing bugs, etc. I've always been saying that making a distributed social network is much easier than "solving" privacy and security for such a thing. First of all, try even defining what it means to privately share thi…

Is your school specializing on building secure distributed social networks?

No, man. I was studying math at NYU. Now I do web development.

It's just a coincidence that these guys are also from NYU.

Re: Trouble with Diaspora

#166

Earlier quoted context omitted.

How is that a fundamental security flaw if there's an easy one line fix with no likely side effects?

If it's that easy, why not just implement it in the first place and give people less of a reason to lambaste your application that supposedly centers around privacy?

It's ALPHA software. They're a few months in. They're releasing early and often. I think it's great that they're releasing it now and that people have the opportunity to work on it with them, rather than waiting for them to get everything up to production (or even beta) quality.
Post reply on HN