Burned by Twitter, Developers Launch Distributed Microblogging Service

http://www.readwriteweb.com/cloud/2011/03/burned-by-twitters-api.php

by rwwmike • 15 years ago

207 90 15 years ago

benwerd• 15 years ago

This is a great thing. Status.net is an awesome project, and Evan is brilliant - but having a bunch of inter-operable services and projects just makes OStatus that much more useful. I can't wait for more.

I hope everyone's creating machine images for these projects. It should take two minutes to start your own microblogging site - ideally with as little technical involvement as possible.

tomkarlo3• 15 years ago

Sounds great, looks great, but actually trying to sign up with a new account using email address (because using my Twitter or Facebook login on this seems well, kind of inappropriate) returns "Internal Server Error".

Fail.

rsoto• 15 years ago

What's up with the downvotes? I also feel like the best option to sign up to a twitter's competitor service is by NOT using twitter.

The signup button goes to a "Internal Server Error" as described above. He wasn't bashing anyone.

Vivtek2• 15 years ago

I think it's the concluding "Fail". It's a bit over the top in terms of tone.

ddagradi9• 15 years ago

We're aware of the email issue, and we're working to fix it. Please keep in mind that this is an alpha release from a small team in a little under two weeks, and there's a long way to go.

wyclif• 15 years ago

Same here. Server error upon using the email link.

ddagradi9• 15 years ago

And actually - this is because our Heroku sendgrid add-on ran out of emails so....herp derp. It's being taken care of; I'm not sure anyone imagined as much attention as it's gotten today.

petervandijck2• 15 years ago

"In order to follow someone from Identi.ca, just paste the ATOM feed from their profile into rStat.us."

That's when it all falls down, of course. And you can't follow people on Twitter?

generalk• 15 years ago

I can see this being replaced with something automated. As it stands it smacks of "hacker put a text input field to test functionality and it works." I mean, worst-case, they could screenscrape the user's page and pull the Atom URL that way.

ddagradi9• 15 years ago

It's only an initial implementation. If you can't follow users between sites with a single click in the final version, we've failed.

petervandijck2• 15 years ago

Cool, looking forward to more iterations. Congrats on the launch!

jmathai2• 15 years ago

Unfortunately, the only people who care about Twitter's recent actions are the nerds. At this point Twitter has too much momentum for the nerds to have much (if any) influence. Building "open" technology isn't an effective method of social engineering here. I wish I knew a better alternative but I don't.

That being said, I never understood why everyone (including pg) referred to Twitter as this great new protocol. It's 100% proprietary and this type of decision should have been easily predicted. SMTP and POP are protocols, Twitter and Facebook are websites with APIs.

But good for the developers. The seem passionate about the idea and are building it. Kudos.

pumpmylemma• 15 years ago

I agree with your tone, but disagree with the conclusions.

   Unfortunately, the only people who care about Twitter's recent actions are the nerds.
   ...
   SMTP and POP are protocols, Twitter and Facebook are websites with APIs.

A protocol is literally a standard or convention for communicating. In the case of twitter and facebook, the tools they provide for declaring social networks enables (arguably new) forms of social communication. Comparatively, SMTP, POP, HTTP, etc didn't require social scaffolding, or at least, it was not the central element. Without the social scaffolding, twitter and Facebook are useless; without social scaffolding, SMTP and POP still let me communicate with friends I already know.

(That being said, I would have preferred that Twitter was a distributed and open protocol from the start; however, I'm not going to switch to an open product just because it's open. I want the more accessible and wider cast social network.)

kirubakaran4• 15 years ago

What about Twitter's actions that are yet to come?

And why not just try? What if the open platform succeeds? Even if the chance of success is 0.01%, it is worth trying. Also, a million such attempts by million such teams can fail. Only one team needs to succeed. Once.

peteforde• 15 years ago

On Hacker News, we don't downvote someone just because we can't find the right words to describe how radically we disagree.

There really are two types of people in the world: lottery players and everyone else.

jamesaguilar2• 15 years ago

> Even if the chances of success is 0.01% . . .

If chance of success * probable value of outcome + (the converse of failure) - cost > 0, it's worth it. Is that really the case here? Personally I think you are overestimating chance of success by a wide margin.

jfischer• 15 years ago

Hmmm. If we look at some past examples of cases in the consumer internet space where a new entrant overtook an entrenched player (e.g. Google vs. other search engines and Facebook vs. Myspace), I think you need to have something new and compelling about your application to drive adoption. I don't know of any cases where just cloning an existing app was successful for a startup. Has movement to a new player every been driven by openness concerns?

What can you do that is different from Twitter and cannot easily be copied by them?

rch3• 15 years ago

In the general scheme of things, Twitter and Facebook have only just demonstrated that a market exists around their type of product. They'll be replaced at some point, but not by the services in the article.

Niche players might be able establish a de facto standard for shared profile data though.

jmathai2• 15 years ago

> What about Twitter's actions that are yet to come?

No telling. That's the point.

> And why not just try?

Because human hours are the most valuable asset on this planet. If you have a .01% chance of success then it's not necessarily worth trying. But that's up to the person who's putting in the time. In this case some people do believe it's worth it and by all means they should try.

I just wonder if there's an option that's more than .01%.

ddagradi9• 15 years ago

Twitter also started out as a niche product "for the nerds": early adopters drive mass adoption. The fact that Twitter is a firmly entrenched player doesn't stop us from competing with a better, simpler product.

I completely agree that "open" is not the great differentiator that will attract endless droves of users. While it makes the platform more attractive to developers and those who share this mindset, the average user's priority is neither owning their data nor using "open" technologies. What we do have, however, is the opportunity and enthusiasm to build something new and exciting with a community of greatly talented individuals.

DennisP• 15 years ago

Twitter is doing this because at some point they have to monetize, and that probably means showing ads. Some users will get annoyed by ads, and that will mean a bigger market for an ad-free alternative.

elviejo• 15 years ago

I think Jabber has had a reasonable success in a very closed space.

First there was ICQ, Messenger, Yahoo Messenger, AOL, etc. And now it seems that you can talk to anybody in any network. And jabber provided a way to create your own network.

I think in the end Openness wins. It's just a marathon not a sprint.

mindstab• 15 years ago

Didn't this already happen, and last time we got identi.ca? Why spin off a new microblogging site when there already is one rebel site full of open source devs? What does rStat.us get us that identi.ca didn't?

BenSS• 15 years ago

Precisely what I was going to post. Both Identi.ca and rStat.us need a compelling reason to switch. Either doing twitter better, or something you can't do on twitter.

'Because Twitter is mean' doesn't pull many users.

drinian• 15 years ago

As mentioned in the article, all of these services use the OStatus protocol for interoperability.

It's pretty great stuff -- I have a StatusNet instance on my website, and people can follow me from other StatusNet instances, I can follow them; it really is decentralized Twitter.

ddagradi9• 15 years ago

Our goal is not to compete with identi.ca, but join them in promoting OStatus as an open alternative to Twitter that encourages a thriving developer community to build new and exciting apps and services without fear.

eoghan• 15 years ago

Meanwhile on Twitter...

http://twitter.com/#!/search/%23stoprightthere

None of these people care, I'm afraid.

w1ntermute• 15 years ago

People are going to be a lot more hesistant about this after the Diaspora fiasco.

andresmh2• 15 years ago

is it officially a fiasco?

ddagradi9• 15 years ago

I'd call it a -gate

JoelMcCracken• 15 years ago

Of course, Klabnik was one of the main dudes who brought this to the community's attention. Not that this would be well known, exactly, but it was enough for this mention:

http://www.kalzumeus.com/2010/09/22/security-lessons-learned...

SkyMarshal• 15 years ago

Trying to register using email gives "Internal Server Error".

(I prefer to register for sites using my email address, and then link to my FB and Twitter accounts. In case I ever decide to delete either of the latter, other services like rstat.us won't be affected.)

ddagradi9• 15 years ago

We have at ticket out that involves better account management that's going to get addressed this weekend at the latest. Being able to add and remove services easily is a priority: https://github.com/hotsh/rstat.us/issues#issue/44

(See other comment re: email)

andresmh2• 15 years ago

written in Ruby to attract more developers than identi.ca that is written in PHP... really?

OSButler• 15 years ago

That line irked me too. I don't believe anyone would take offence about the underlying technology, if the product is of excellent quality. The OP might want to add some more explanation to that point, as it seems more like the usual php bashing than being based on substantial data to validate that point.

MatthewPhillips• 15 years ago

I took it to mean that there are some languages that certain developers won't touch, and php is one of those and ruby isn't. Can't we agree on that?

shuri• 15 years ago

The question is how can you get UberMedia on board.

anon114• 15 years ago

P2P twitter is absolutely necessary but it has to be done right.

First of all, ensure privacy with public key cryptography. Sign tweets for authenticity. Retweets can just be additional signatures.

If a distributed microblogging protocol was interoperable with twitter and user friendly, it would probably be able to siphon people off of twitter proper. Certainly it would be an attractive alternative to anyone who NEEDED the service, and that's the important thing, right? Hopefully work out a way so that tweets on twitter.com can be captured and distributed in this P2P network. These tweets could be unsigned since if you include a link to the original tweet they can be verified.

Defining protocols instead of providing services democratizes a layer of the OSI model. We need to think deeper than that, though. We need to democratize the physical layer as well. Luckily we've proven that you don't need a high bandwidth link to be useful in a crisis situation. Twitter will do. To that end, I suggest that this project make an even more lofty goal:

Create a small embedded device interoperable with this P2P microblogging network. The device can communicate with peers over a Software Defined Radio. The device should be capable of bridging to a wi-fi or 3g network. This would accomplish the democratization of the physical layer which is so important to combat censorship and oppression.

You can't monetize democracy. This is why these projects will only happen on a volunteer/charity basis.

cabalamat2• 15 years ago

> The device can communicate with peers over a Software Defined Radio.

For extra points, allows several such devices within range of each other to automatically create an ad hoc mesh network.

dfj225• 15 years ago

I agree with you in principle, but I don't ever see something like this becoming mainstream, even if it does interoperate with Twitter and is as easy to use as Twitter. Simply put, there will be no selling point. Switching would give the user the same capability of subscribing to short messages from other people as they already have on Twitter. So, why switch? Even if they can continue to follow everyone they already follow, what's the incentive aside from technology they don't understand anyway?

im_dario• 15 years ago

My final degree project in Computer Science was an attempt to build a P2P twitter http://code.google.com/p/qantiqa/

Due to personal issues and a tight deadline it was not featureful as I wished but it worked, although there is a central point that works as gateway to the overlay.

Anyway, after this point, it works fully decentralized.

dchest• 15 years ago

Looks interesting. Is there a human-readable description of the protocol somewhere?

dmoney• 15 years ago

Lots of people have thought of writing "twitter but open". Congrats on actually doing it. When I was kicking around the idea but never acting on it, one thing that occurred to me was that, in order to make it really take off, you have to create some way that everybody already has it. Some way to follow and be followed by twitter users, for example, or a way to make it so that your e-mail address is your microblog address. Just my two cents.

marcamillion2• 15 years ago

I hate to say it...but like Diaspora, I expect this to come out with a bang and slowly peter out and die.

It's not because Twitter's technology is any more superior. It's the simple thing called 'network effects'.

Unfortunately, as others have pointed out, I think Twitter is too far along for their momentum to be stopped.

Valiant attempt, and kudos to the developer(s) for actually getting it launched - but this is definitely like spitting in the wind....imho of course.

cabalamat2• 15 years ago

> It's not because Twitter's technology is any more superior. It's the simple thing called 'network effects'.

I agree, and the solution must be to provide a service that is useful even when not many people are using it. For example, could an rstat.us account be set up so that everything published on it is automatically republished on the user's Twitter and facebook accounts, and any replies from those places published back to rstat.us?

Hawramani• 15 years ago

This is a great idea! (So long as Twitter doesn't block it)

recoiledsnake• 15 years ago

That will make the incentive for Twitter users to get rstat.us accounts a round zero. Exactly what Buzz tried, remember how badly that failed?

_frog• 15 years ago

It already posts updates to your twitter feed if you sign in with that account and I've heard they'll be adding that feature to standard accounts soon.

InclinedPlane4• 15 years ago

Yup, at this point you can't get people to move away from twitter just by providing a nearly identical service. Because the network effect makes them no longer identical. You need to add a way for people to either proxy through the competition (e.g. parallels on a Macbook, skype integrating with the existing phone system, etc.) or you need a featureset that is so superior enough people will switch anyway to catalyze a larger migration.

wilkie• 15 years ago

Perhaps. But the code will be there, and somebody can start up their own site on their own server at any point of time.

And the beauty of it: the way the technology works (OStatus), every time somebody does this, the overall network gets larger. In my opinion, it's really only a matter of time.

marcamillion2• 15 years ago

I have to disagree with you on this one. It's not about the technology. I think this is what we techies get mixed up.

I would wager to say that by now 95% of all Twitter users don't give a crap about the technology.

They just want Twitter to work.

Twitter solves a problem that people didn't realize they had before. The ability to easily see what people are currently doing - the 140 character limitation is a godsend for brevity.

Because Twitter invented that space, it is going to be VERY difficult to unseat them.

At this point, only Twitter can unseat Twitter - quite frankly.

lparry• 15 years ago

diaspora came out with a bang? heck? is it even out? if it is, it's "coming out" was more like a wimper

evgen• 15 years ago

"Network effects" cut both ways. I am interested in participating in the network that most of my peers are on, but I don't give a shit which network Ashton Kutcher, some C-list blogger or a random hashtag spammer is on. A non-twitter microblogging network could actually gain a lot of early momentum just by giving some thought to what it would take to make twitter suck less and then letting people know that this is not the service that Oprah or their mom will be on.

nikcub• 15 years ago

These guys have that going for them: they aren't in the NYTimes

If they get to the point of providing users a viable alternative, then it has worked. It doesn't have to be 10M users.

erehweb• 15 years ago

From my May 2010 post on Diaspora* and Facebook - "Billion dollar companies that are paying any kind of attention don’t sit quietly and let themselves be dismantled." http://erehweb.wordpress.com/2010/05/17/why-diaspora-is-grea...

InclinedPlane4• 15 years ago

Twitter is not "microblogging", thinking of it that way will only cause you to misunderstand twitter and come to incorrect conclusions. Twitter is closer to facebook and IRC than it is to blogging.

jmreid• 15 years ago

That's your opinion. It's too generic of a service for it to be used in one way.

InclinedPlane4• 15 years ago

It's not identical to facebook or IRC either, that's my point as well. It's a new sort of thing that requires careful thought to properly understand, applying inaccurate labels doesn't help.

TheSwede754• 15 years ago

I agree 100%, twitter is 'SMS for the web' and NOTHING else! 140 character limit should be enough to make this point.

benatkin• 15 years ago

Has anyone managed to subscribe to an rstat.us feed from identi.ca?

TheSwede754• 15 years ago

I LOVE the fact that you can 'login with twitter' to rstat :-)

TheSwede754• 15 years ago

Internal server error when trying to sign-up.

ddagradi9• 15 years ago

There's an email issue we're aware of. Twitter and Facebook logins are working correctly.

TheSwede754• 15 years ago

Thx, I did notice. Testing out with twitter login :-)

bergie• 15 years ago

I wish there was a OStatus implementation to a Jaiku-like service (see for instance http://qaiku.com )

The threaded microblogging model supports far deeper interaction than the way Twitter works.

elrodeo2• 15 years ago

http://identi.ca/group

pathik• 15 years ago

The next Diaspora. Doomed to fail.

Tichy• 15 years ago

I recently had this thought that even with a distributed system, probably some centralized monopoly would emerge - the search engine, like Google for the web. Following people works OK in a distributed way, but search and analytics would be a problem. Unless an efficient distributed search engine can be built, too.

elrodeo2• 15 years ago

I'd suggest that all people who want to try out rstat.us post something with hash tag #hackernews, such that we can connect there.

dclaysmith• 15 years ago

I think that it is possible (tho unlikely) for an alternative to Twitter to emerge but it would require:

1) All clients start implementing the "open" protocol next to the Twitter. So send each message twice. 2) Someone takes the Twitter firehose and screws it into the "open" protocol.

You would have a bit of noise but you would then have a legitimate choice: "open" or twitter. If the number of people using the "open" alternative/protocol achieved critical mass you would have something.

Maybe this is why Twitter is clamping down on clients. Undermining the popularity of clients prevents the first item.

And Twitter would never allow anyone to use the firehose for this purpose.

So, pretty unlikely.

markkat2• 15 years ago

I think they should consult some people with strong experience in marketing in order to present this in the most concise and approachable manner as possible.

Their biggest hurdle is going to be apathy due to a lack of comprehension. The more noob-friendly rstat.us is, the better their chances.

They've actually done much better than most, but it really can't be dumbed down enough. There should be a clear barrier between text for developers, and text for everyone else. I really wish them luck. It's a noble effort.

ddagradi9• 15 years ago

We've got some great designers ready to help us simplify the experience and provide a clearer and more comprehensible message. It can't be stressed enough how important great design is to a project like this, and I'm looking forward to crafting a truly elegant experience.

markkat2• 15 years ago

Cool! I'm looking forward to seeing what you guys come up with. Like I said, what you have gives the impression that it's on your mind.

The underlying concept is actually a pretty heady idea to non-tech folk. It might not hurt to get a bunch of them and bounce things off of them until you see their light bulb turn on. Best of luck!

rch3• 15 years ago

Strange to see a reference to couchdb/couchbase so awkwardly shoehorned into the article.

modernerd• 15 years ago

Distributed microblogging: connecting the masses by making them further apart.

Great to see OStatus gaining traction.

Is there space in the OStatus spec for an email-like username/service format? I'd like to be able to add someone on identi.ca by typing theirhandle@identi.ca, for example, rather than having to copy and paste a link to their atom feed. Or would that be too confusing? If so, is there a more user-friendly alternative?

jamesll• 15 years ago

This is an additional spec used by statusnet called webfinger. we're working with the OStatus folks and should have support soon.

ajays• 15 years ago

If I could draw an analogy: Twitter is like AIM. Can we put together an open service, like XMPP, which interfaces to TWitter, but is distributed and free?

Burned by Twitter, Developers Launch Distributed Microblogging Service

Related Stories