[pinaceae]

it is a big conundrum if your core product, through its simplicity, is really great.

you hire all these product people, have all these investors, but any direction you can take the product actually makes it worse against its initial, great core use.

twitter as a protocol is on a level with smtp - a lucky strike, hitting a need, something for the ages. journalists, media, etc. love it. RSS on a whole new level.

but twitter as a product company? smtp is a not a profit model, you need to have real, closed products - hence the API limits, hence all this other bull. they have a narrow scope hit product and will kill it by making it broad. a little bit like google and search, put ads on it, done, the rest is noise driven by boredom and/or panic (we need to justify our existence!).

twitters design team is bigger than most startups - and for what? the whole slack team fits into the twitter reception area and covers how many platforms, apps, use cases by now?

you threw a lucky punch with a communication channel/protocol, but now you're stuck. aren't we happy that the smtp or unix guys as a whole didn't try the same. "monetize".

[pionar]

Twitter isn't a protocol. If it was, there'd be Twitter (the protocol) based services everywhere, and it'd be open and federated.

Twitter is a service. Services need money. It's a simple as that.

[arrrg]

Thing is, Twitter should be a protocol. It clearly fills an existing need (I see it as very similar to email in that regard, just filling a different need – people want to chat publicly with each other and in a very informal way inform the public about things), but it doesn’t really make sense as a service that wants to make money.

Twitter should be just like email, not this one monopoly controlling everything. It’s sad that that’s the way it is.

[patrickk]

I feel exactly the same about Facebook. There should be a protocol for social networks and should be federated with no one controlling everything, similar to email.

If fact, I believe it will go this way at some point in the future, despite how impregnable Facebook seems now. At one point Bebo and MySpace seemed unassailable too.

[majani]

Twitter was open to the protocol idea once, when they were even allowing Twitter clones to be built using their API. Chances are they simply ran the numbers and saw that that route was not big enough to fulfill their lofty ambitions.

The only way the social network protocol can be realised is if you have a lucky band of developers for whom everything just clicks into place without funding, like Craigslist. Unfortunately that's really hard to achieve in today's internet landscape where anyone with a large hosting bill gets an offer they can't refuse from a VC.

[chton]

I'm not sure if it's even possible to build a twitter-like system with distributed servers in the vein of e-mail. It's possible for e-mail because it's a one-way system. Twitter has the concept of following a user, and that user's tweets appear to me. It's a 2-way system that needs to constantly be kept in sync. Even if it's possible to do, it won't be easy, with loads of extra technical challenges that SMTP doesn't have.

[steveklabnik]

There was already an open standard to do everything but the private messaging part: http://en.wikipedia.org/wiki/OStatus

[freehunter]

App.net is like Twitter, the protocol. Not quite the same, but it's basically Twitter as a platform, where the core business is federation and an API for services to build off.

Still centralized, still controlled by one company, but it's a step in the right direction. It's a shame they're struggling to survive against worse services.

[brandoncapecci]

Twitter is a private company so they should continue to do whatever they please. If you want to create a decentralized Twitter go for it... my guess is it would be about as popular as Diaspora but all the praise to you if you can disrupt them.

[arrrg]

They can do whatever they please. I can wish that they would do something else. They obviously won’t. It’s not in their own self-interest.

Saying they should remain this cynical monopoly that does shitty things just seems nonsensical to me. No, they should not. They can. They do. I think they shouldn’t.

I mean, none of this will actually happen. Twitter will continue to have a monopoly and I see no realistic way to change that. I don’t think there will ever be Twitter as a protocol. I’m just saying there should be. It did work out for email. And, yeah, email is imperfect in so, so, so many ways, but at least it isn’t under control by a single company.

[brandoncapecci]

You're allowed to think whatever you want obviously; my point was that stating you'd wish they'd gone a different direction is much different than seemingly positioning an argument for what they should and shouldn't do as realistic or in any way beneficial for Twitter. I'm glad you cleared that up but your personal opinions on their direction seem in contention with your admittance that it wouldn't be possible or advantageous for them to follow what you'd like them to do. In regards to email, the oligopoly of decent clients act in a way really no less perverse than Twitter does (which I don't think is much at all mind you).

[pinaceae]

it started as such. lots of 3rd party clients, cool new use cases and UIs. all squashed by now. API caps, etc.

they need money, sure. but not just to operate and make a nice profit, they need to grow, they need blowout quarter after blowout quarter. impossible with this simple and clean core product. hence doom on the horizon. just look at all the bad feedback they're getting for their feed changes. core audience hates it. won't bring new users. stuck.

twitter could/should be run like craisglist or reddit. hardcore maintainence mode. neat little business. but no VC will ever allow it.

[chton]

Here's what the problem is: twitter isn't a product company. It's a service company. They didn't just invent a protocol, like the SMTP guys did, they provide a service for free. That means they need money to hire people, maintain servers, and survive in general. Even if they didn't want to make any money, if they never cared for profit, they would still need a way to have their service generate income.

I agree that it's a conundrum for them, but with this solution at least they're trying to do it in a non-obtrusive way that doesn't jeopardize the entire platform.

[ksk]

> unix guys as a whole didn't try the same. "monetize".

um.. UNIX started out as a closed-source commercial OS.