Slashdot Mirror
Creating a Better Facebook
Fed up with Facebook's insatiable need to continue to expose your personal information to ever widening circles, four NYU students have decided to build an open source, distributed competitor to the social networking behemoth. They've raised a few grand, but I imagine it will be harder to convince your mom to log in.
Unfortunately Facebook's power is in that everyone uses it, and that is what they use to get new users too. Alternative projects are a humble goal, but especially with social networks you are quite much locked in to a single existing network just because everyone else you know uses it, and they in turn use it because you use it too.
Interestingly creating a network like this means you have convince everyone to forget about Facebook and move to this platform. Even if it would become successful, once these four students have millions of people in their social network, they most likely will change it the same way that Facebook did. Remember that Facebook also was a hobby project made by students.
Am I part of the core demographic for Swedish Fish?
These guys just want extra cash for the project, without giving out a clear view about how the platform will work or run?
A facebook-clone in 3-4 months? Very unlikely.
That is, essentially, how Facebook began. The only thing that is different is greed. As college students, they might want to protect privacy. As fresh out of college students, they might look at their massive college debt and start weighing their options. Before you know it, they're paying lip service to advertisers at the expense of their user base. But hey, at least they'll have Ruralville!
Pay no attention to the amount of data they let loose upon their facebook pages. Nor do they care, as long as they can access their online farms. They're already giving out their credit card numbers to buy fuel for their tractors.
Saw this article this morning. Don't overlook the "dirty Unix joke" on the blackboard. ;)
I only post comments when someone on the internet is wrong.
First everyone hosted their own site themselves (I believe this was the case? I didn't really do that part)
Then everyone had sites hosted elsewhere (geocities)
Then everyone had a page on a single site (facebook)
Soon everyone will have their own facebook (diaspora)
And then everyone will have their own... everything on their own server... kinda like Unite by... Opera! Always two steps ahead
They hit slashdot. They will become major in a few days. Don't worry. This one will actually succede!
Right, because everyone knows that Slashdot posters are social dynamos, followed by hundreds of fans who will willingly follow them onto the new network.
Please apply 5 seconds' thought before getting all distributed up in my hizzizzy.
For this service to be popular, Real People will have to use it, not just you, me and him over there.
For Real People to use it, it will need to Just Work, First Time.
To Just Work, First Time, it needs to rely on having a reliable server/seeder/aggregator/gateway present 100% of the time. Let's call it a metaserver, although it's just semantics. There needs to be one place where every peer goes to find out where other peers are.
Who's going to run that default metaserver? Well, duh. The authors will run it.
When - not if, when - they go Dark Side and release a client that injects ads or collates data, who's going to switch to a fork clients and a different metaserver and protocol version? That's right: you, and me, and him over there. Not Real People.
If this takes off, then 99% of users will treat it exactly as they do Facebook, as a service that can (and will, eventually) do pretty much what it wants to them. Its success is predicated on being used by Real People, not you, me and him over there.
You may now commence your explanations of why this time, it will be different, and Real People will care about the things that you, me and him over there care about. I apologise for the interruption.
If you were blocking sigs, you wouldn't have to read this.
ctd.
This is why distributed approaches like Diaspora/Retroshare/... will fail:
- You have a problem publishing new versions of the software. You can't force new versions out, there will be incompatibilities between nodes, things will not 'just work'.
- Privacy aside, you don't add value that Facebook hasn't.
- Quality of the service: The development team or community will not provide a continuous, mature program version.
* unless they have some business model on how to generate revenue from it.
- No inspiration, or higher goal they strive to. They just do something existing a little bit better. But there is nothing fundamental about why one should use the new service. It is better in features, it is logical to use it. But that is not satisfactory.
- Original developers will at some point stop maintaining the project, and not have gained enough other developers around them that continue development, maintenance and infrastructure on a high quality level.
Please, Diaspora* team, prove me wrong. Read this and prove me wrong.
If you can't, it is not the fault of your expertise, or skills as a programmer or software engineer. There is just more to it than developing a superior product.
NB: The message above might reflect my opinion right now, but not necessarily tomorrow or next year.
Almost, except for the the fact that not at all.
That's a bad thing?
This Easter I was BSing with the various family members after dinner. And my sister started getting on my teenage niece's case about some boy on her Facebook page and the teen-related shenanigans mentioned. Minor shit -- a kiss.
I finally looked at my sister and asked her if she recalled being that age. I recall my sister at that age, and let's just say our mom would have been elated if she could have kept her activities down to raunchy (as opposed to nasty).
Teens need liberated from Facebook. No one needs their goddamned parents breathing down their neck just because last night their boyfriend was breathing down their neck.
I scream. You scream. I assume that means we're both acquainted with the problem. We proceed.
There are quite a few projects to create this:
http://www.elgg.org/ though it is not distributed (they are working on it)
http://onesocialweb.org/ is xmpp based, i have set up my own instance.
http://groups.fsf.org/wiki/Group:GNU_Social has just started and is a gnu project.
There are some standards to help this kind of thing but most are not complete.
you may want to look into foaf for storing a social graph for example.
Please chat with other people if you find this interesting.
IRC chat: #social on freenode
To me, software freedom is being able to choose and customize the software I use without limit. With applications like Facebook, I cannot of course do any more customizing than the Facebook allows me to. The FSF tried to address this problem with the AGPL and many web applications have rightfully chosen it as a way to give users freedom online. Unfortunately there's a rather big part of the equation that the AGPL and the four traditional freedoms miss. It's that our data is often stuck inside even AGPLd applications. If we want to have true freedom online we need The Freedom to Migrate and it seems Diaspora is trying to provide.
Because first you did indeed host your own site, but YOU were a university or other large institution as they were the only ones to have access to the net.
If you mean the unwashed masses with everyone then the first hosting was the home page, provided by your ISP.
Geocities and the like came after that, when ISP's turned more towards low-cost and provided to little flexibility or capacity.
Next up was the blog, the home-page re-invented.
Myspace took a look in.
I think that with ip6 we might actually indeed get something like Opera's unite instead. I already use the same Opera thanks to its sync feature, why not host my profile from my web browser as well and people connecting to my own router which has its own IP and can be reached by anyone? The next facebook would possibly not host the content, but index it instead. It would allow far more freedom as to how you publish information, but also make it less standarized.
Basically, since you list is far longer, it seems we get something new, or something old in a new coat, every couple of years. So presumably the perfect method has not yet been found.
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
Drupal has been providing open-source community plumbing for years.
The real trick here is how they build the communications and privacy models - how much data is open, how do they share it, what can you do with it - and how that affects what kinds of features you can build using it. You could build something like Livejournal on a pretty tight system with no central data storage, but it'd be harder to find your ex-girlfriend's cousin's third-grade-teacher's dog's picture and send it a cute icon of a fire hydrant. Or if you build a system that's really good at both of those, then you'll have tradeoffs in how much data you have to ship around, so your DSL connection is 98% full of encrypted packets for your friends' friends' friends' searches, and your query gets you a dialog box about "your posting may cost the net hundreds or thousands of dollars."
And that flexibility is important, not only for the kinds of marketing people who want to monetize everything, but also for the people who want to maintain the community and keep all of those users around and interested, as opposed to having them disappear like Friendster or Orkut users who had their fifteen minutes of fame and six months of friend invitations from cute guys in Brazil. Livejournal seems to be doing ok with it, but Facebook gets a lot of social involvement out of all of that Farmville and Mafia Wars stuff, and the question becomes how to facilitate the social networking effects of it without also the mass information-leakage.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks