Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Trouble with Diaspora

blog.steveklabnik.com

Post reply on HN
Print view View original

31–40 of 166 posts

Re: Trouble with Diaspora

#31

Post by jarin »

I'm not even sure it's worth submitting patches to Diaspora, both because of the fundamental problems with the code, but also because of their "Open Core" licensing scheme (AGPL + contributor agreement): http://www.ebb.org/bkuhn/blog/2009/10/16/open-core-shareware...

Looking through the code, it looks like Diaspora is really just putting a front end and "aspects" on top of OStatus. I think it might be good at this point to just scrap the Diaspora code and start over from the basics with a good OStatus-based reference implementation in Rails.

Re: Trouble with Diaspora

#32

Post by com »

There's a lot of negativity and "I told you so" snickering floating around Diaspora on Hacker News.

I'm putting it down to envy: these guys have shipped some pre-alpha code that's interesting to a large number of people.

Excellent marketing in the open source community for developer eyeballs, perhaps not so good in terms of end-user experiences, but that's not the point at this stage, numbers will be low and the perceptions of the dumb early-adopters (of pre-alpha distributed social networking code, ffs) shouldn't leak too badly into the mainstream.

However, people are now eating the dogfood, and I expect to see fairly rapid improvements in the code: not unexpected for an alpha drop in my experience.

To the people who are moaning, would you like others to see your alpha code and laugh bitterly about you being a young (or old for that matter) upstart?

Re: Trouble with Diaspora

#33

Post by mfukar »

cilantro wrote: post #3

This code was released to developers as an incomplete preview. I'm not sure why people are holding it to the same standards as a finished product that's being released to end users. Seems like a pretext to talk trash.

First impressions are important. Just as I was excited to hear about a decentralized, privacy-aware social network back in May, I am now worried when it's made available for us to watch its course.

The fact is, the implementation is obviously problematic. I won't go as far as saying that the design is flawed, although some security vulnerabilities are certainly pointing that way. It's obvious that Diaspora's developers are inexperienced, and therefore could use all the help they can get from us. This is one point in favour of an early release, in my book. On the other hand, early design and implementation decisions have profound implications during the lifespan of a project (this is especially true when it comes to security), and I will not trust my personal details to Diaspora in its current form; the project still has a (very) long way to go before being technically up-to-date.

Re: Trouble with Diaspora

#35

Post by mayank »

T_S_ wrote: post #30
jlgbecom wrote: post #13

Earlier quoted context omitted.

And more importantly, if you're going to rewrite, why help Diaspora, and not a more mature option?

What more more mature option were you thinking of?

Here you go! There's not just one, but a whole bunch!

http://gitorious.org/social/pages/ProjectComparison

Re: Trouble with Diaspora

#36

Post by citricsquid »

gloob wrote: post #22
citricsquid wrote: post #18

Earlier quoted context omitted.

The issue is you shouldn't build (or ship) code like this with such major security holes, you build security at the start, it should be an integral part of the application. You can't just dick out some insecure application then add in security, it doesn't work.

Microsoft didn't "build security in at the start", nor did Apple, nor did Twitter, nor (I suspect) did Facebook or YouTube. It's a pre-alpha of an open source project. Of course there will be problems.

Yeah and how much code is being used now that existed when they didn't have security? You can't compare this and what Microsoft, Apple or Twitter have done, I'd be very surprised if any of those companies continue to use code that was developed at a time when they didn't consider security. Although thinking about it, the software industry is questionable... maybe I'm wrong, it just seems a very bad start.

Twitter has had major security problems, same with many other "big" companies, shouldn't this be a lesson that security is the primary concern especially for an open source project?

Re: Trouble with Diaspora

#37

Post by kennedywm »

jlgbecom wrote: post #13
mayank wrote: post #2

It's great that they're getting so much open-source help, but I'm going to ask the obvious question: if a "complete overhaul" is what's needed, as the author seems to imply, and the FOSS community performs said overhaul, then what of the $250k that was given to the Diaspora guys? Is it still even "Diaspora" anymore, as opposed to a FOSS project?

And more importantly, if you're going to rewrite, why help Diaspora, and not a more mature option?

Because momentum and attention are more important than maturity. The qualities that come from maturity can be built with work; momentum and attention aren't as much of a function of hard work and are far more difficult to capture.

Re: Trouble with Diaspora

#38

Post by pkulak »

First people bitch and whine because they are working "in secret" and not releasing code. Then they release and everyone freaks out because it's not finished. I guess the only way everyone would be happy would be if they released a completed project after a week.

Re: Trouble with Diaspora

#39

Post by ohyes »

This code was written by a bunch of undergraduate college students.

This is hiring a bunch of interns (with near zero experience) to implement your product, giving them three months, and then being shocked -- SHOCKED, when the code is not professional grade quality.

I was shocked when everyone and their brother was willing to shell out money to a group of completely unproven college students to produce a distributed open source 'Facebook Clone', that is also 'private'. My inclination is that the 'distributed' and 'private' parts of the description push it into the oxymoron tier of product specifications. I would have expected this alone to give people pause about what the architecture would be (somehow it didn't).

Honestly, at least they have produced something, and for the most part, it works. Hopefully they haven't burned through too much of the $250,000 that they started with. 3 months of development time is honestly nothing.

Presumably, they could get comments on this, throw the entire thing away, re-write while fixing the various issues, and be well beyond where they are now in another three months. (If this is as bad as Steve says, I hope this is the case). Presumably the development will go faster because previously they were learning and developing at the same time (presumably).

Hacking together a prototype that you then throw away is a perfectly reasonable development model. I'm impressed (and pleased) that they have produced anything.

Re: Trouble with Diaspora

#40

Post by cmars232 »

jarin wrote: post #31

I'm not even sure it's worth submitting patches to Diaspora, both because of the fundamental problems with the code, but also because of their "Open Core" licensing scheme (AGPL + contributor agreement): http://www.ebb.org/bkuhn/blog/2009/10/16/open-core-shareware... Looking through the code, it looks like Diaspora is really just putting a front end and "aspects" on top of OStatus. I think it might be good at this po…

At its core, the real value in something like Diaspora is its protocols. If you disagree with the license, or the language they chose, why not make your own implementation from scratch that can federate with it.

If it survives, I might do this... I think their approach is fine for rolling out a private enterprise FB or creating an alternative hosting solution, but I'd like to see an even more decentralized solution. Ruby isn't a good choice of language for mobile, or native compiled "IJW out of the box" deployable solutions.

Post reply on HN