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.
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.
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?
If enough of us "nerds" get pissed of by the actions of Twitter, can't we ourselves provide the initial momentum for an open alternative, with HN having 80k daily uniques and all? Does it have to have better features? (although that will help)
It's all about critical mass. If techies who care about it being open switch over and then it becomes the most popular amongst HN readers, then it can grow out from there.
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.
That is a good method to make personal choices. But in the 'startup ecosystem' (pardon me) or the 'free market' the cost is privatized and the benefit of success is socialized.
The overall equation looks like this: success = (team1_success or team2_success or ... or teamn_success)
I was assuming that the calculation was from the standpoint of all society, considering most free software advocates consider themselves altruists. So our functions are the same in this case, and it is still possible for the project to not be a good idea based on expected value.
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.
> 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%.
I am not exactly worried about getting downvoted but thanks anyway :-)
I guess I should elaborate on my previous comment: I am saying that there is no reason to look at someone actually trying something big and say "Dude, give up! It won't work!" no matter how rooted in reality you think your choices are.
IMHO it is not helpful or interesting to view the world in binary terms as "lottery players and everyone else". In the risk profile continuum, you're drawing a line at your max risk tolerance level and viewing it as a binary choice. But it is a continuum. Someone else may view everyone not working for the government with pension benefits as a lottery player.
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.
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.)
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.
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.
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.
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.
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.
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?
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?
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.
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.
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.
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".
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.
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.
As much as I'm sympathetic to a small team getting a spike in traffic, I stand by that comment... It's a pretty big "fail" when a huge rush of users trying to register results in... nobody being able to register. At the very least they could have been capturing the emails for later follow-up. How many of those people will remember to come back later and sign up again?
> 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?
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.
"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.
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.
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.
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.
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.
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.)
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
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.
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.
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!
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.
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.
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.
It's clear what twitter is: it's a website, it's a SMS service for phones, messaging, service, etc.
There is plenty of vertical development that can be done by third parties, no need to focus on creating your own client.
In programming, twitter is like the IE mini-browser control you can drop into a window and create your own 'custom' browser. Clients like that aren't needed. Instead use the data in a new way, or provide a new data stream by tracking something in real life.
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.
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?
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.
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.
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?
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:
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.