Hacker News new | comments | show | ask | jobs | submit login
Twitter to developers: no new Twitter clients (groups.google.com)
451 points by samstokes on Mar 11, 2011 | hide | past | web | favorite | 172 comments

It's the old casino analogy: When you're betting on a platform you don't own, the house always wins.

I don't know much about Twitter politics, but we've got:

- Developer hostility this week

- User hostility last week, with the dickbar

So maybe this is just how it goes now that Costolo is in charge. "Fuck the loyalists, we're here to make some money. Getting sick of these third party tools charging coin for our damn API while we get nothing."

If so, the question becomes: who creates more value on Twitter? Is it the old guard, who use it as a communications and community medium? Or is it the mainstream, following celebrities and talking about sandwiches they're eating, eyeballs to be sold to the highest bidder?

(edit: My hunch is that, too late, the Pareto principle will be discovered hard at work: 20% or less of Twitter users actually generate 80% of the value. I just can't see it as a bland, empty consumption tool. There's surely peril in optimizing toward that.)

It sounds to me like Twitter wants to round up its user-cattle and drive them on down to monetization gulch. Anyone who gets in the way of this is going to be flattened.

>It sounds to me like Twitter wants to round up its user-cattle and drive them on down to monetization gulch.

Uh, yeah. Is that really surprising? We're all being driven to Monetization Gulch to escape the absolutely fatal Winter of Unprofitability.

> Uh, yeah. Is that really surprising?

Your sarcastic grunting aside, it's utterly shocking. Imagine Google, circa 2001, with crass, interstitial banner ads between search and SERP. It would have destroyed them. The opportunity for heaps and heaps of money, gone in a single, vulgar puff of smoke.

Twitter has the opportunity to leverage a critical mass of enthusiastic users. This is a rare, golden gift.

Last time this happened, Facebook minted some billionaires while discovering a sustainable, user-virtuous business model. There were growing pains, especially with the early, spammy apps, but overall, Facebook has managed to balance the needs of a lot of people without screwing up the magic of their product.

The time before that, Google did the same thing.

Sadly, rather than pushing toward a business model that makes the most of the very impressive ecosystem they've cultivated, it looks as though they want to break down the most interesting portions of it and then find money in whatever turns out on the other side.

The lasting virtue of Twitter, to me, always seemed to be that it would always be whatever any given user needed it to be, thanks to the wealth of apps that dip into its API. If that goes away, Twitter's shelf life is curtailed.

To me, it seems nutty, and not terribly imaginative. Ask Digg: pissing off your core constituency so you can hopefully make money is not the same as actually making money.

pissing off your core constituency so you can hopefully make money is not the same as actually making money.

So, after all this ranting about Twitter's change - what do you suggest they do to "actually make money?" Isn't that the end-game for every company? Is twitter even close to justifying their private valuations with revenue? I haven't seen anything from you that suggests an alternative approach that actually makes them more money, other than "don't do that."

You also previously mentioned that twitter has a rare golden gift of leveraging a critical mass of enthusiastic users.

My thoughts about that? Show me the money! So far, they aren't doing that.

Easy, have normal accounts for personal and non profit use, anyone that is trying to either sell things or building a brand can pay for access, including all the celebrities.

I don't mind the current promoted trends/tweets/follows, I guess the only thing I would change is call the promoted trend something else, because it isn't really a trend.

This makes no sense. How is it in any way tractable for Twitter to bucket a user into a brand, a person, a person who is also a brand, a person who is trying to build a brand, a person who is just really enthusiastic about a product, or a astroturf account pretending to be really enthusiastic about a product?

It's not. That's not easy, that's a fundamentally realignment of the product.

What if you're a normal guy and you get famous from some youtube video? Now you're a celebrity. Pay up - but oops, you're broke? Fuck it, you're off Twitter.

Not gonna work. :/

Okay make it simpler, your posting as directly as a for profit company and not under your own name you pay, doesn't have to be a large amount. I think it's better model than advertising, if your reaching a customer base through the service and see value in it for your business you pay. I personally like it a lot better than ads.

They could offer premium accounts with new services like analytics, search ads like google does, charge brands for an analytics tool showing anyone talking about their products, etc.

There are definitely options.

I think they should use their large engineering team to build something new and separate from Twitter.

Wait, the Twitter engineering team should build something separate from Twitter? I don't understand.

On a similar note, though, I am surprised that this hasn't sparked more discussion about the distributed open source social networking movements. I know Dave Winer's been talking about quitting Twitter working on some sort of alternative. There seem to be some interesting answers to Facebook that could orthogonally solve similar problems to the ones Twitter does. Anyone know of any other interesting work along these lines?


> Uh, yeah. Is that really surprising?

Why yes it is. Twitter went out of their way to say how much they supported developers and considered them key to their success. That wouldn't seem to include making Titan off limits.

80% of Twitter's users probably won't care, it'll be Bieber and #bootyappreciationday as usual for them.

If we start seeing a lot of Dickbar-level gaffes from Twitter though, it may end up that the other 20% starts flocking to something else (kind of like the slow exodus from MySpace to Facebook).

Totally agree. I think that's the danger Twitter faces with these kind of moves.

Since the nerds founded the Twitter community, I wouldn't be surprised to see the next move be some sort of open, distributed, federated Twitter-clone that mirrors the existing APIs. Easy win for the client developers, no one to be a buzzkill, and it's not like spam can be any worse than it is now anyway.

This has existed for quite some time now and is called identi.ca

I've used identi.ca almost as long as twitter, and my preferred client Twidroyd on the Android supports both of them (thanks to the fact that identi.ca has an option to use an API made to emulate the twitter-API).

I highly prefer identi.ca to twitter, and the only thing keeping me on the twitter-network at all is the fact that most of the people I follow only are on twitter, and for some strange reason it's very hard to convince people to change, even though, in my opinion, identi.ca is superior.

I really hope they soon manage to get a twitter-pull up and running (they have been talking about that for a long time now). If they managed that I would convert fully to identi.ca.

Edit: Got to update my browser more often... Sorry bingaman

Why do you prefer identi.ca to twitter? I don't use either, but I have been to both sites. Twitter looks nicer. I wish people did prefer identic.ca, it concerns me that so few popular websites are open-platforms.

The fact that middle-click-to-open-in-new-tab doesn't work anymore in the Twitter web client makes identi.ca at least less crappy to my tastes.

Mainly because it's open, but also because it seems to be more stable than twitter and because I find the twitter web UI horrible (not that the identi.ca web UI is nice, but it's still better).

Generally I almost only use free software, and prefer open-platforms whenever possible. I actually can't remember the last time I used non-free software.

Freedom is killer feature.

For people who consider FOSS a killer feature.

By rights it should be, but it's demonstrably not. Sadly the industry is plagued with a critical mass of people (though I question whether they are actually still members of our species) who don't object to being told how they may use their tools.

Identi.ca: http://identi.ca Status.net: http://status.net

With all due respect to the open source Twitter clones, Facebook didn't beat MySpace by being open source. It beat MySpace by being an exclusive, opinionated, clutter-free alternative to MySpace, and later on a developer platform.

Are identi.ca and status.net distributed? Their website doesn't make it obvious what it's all about.

Status.net is the Free software that power Identi.ca - Identi.ca is the showcase instance of Status.net and the biggest to date... But there are many others : http://status.net/wiki/Sites - And you can deploy your own !

Yes, using the OStatus protocol suite, see http://status.net/2010/03/07/understanding-ostatus

No, they're centralized like Twitter. StatusNet is like the Wordpress of microblogging. Each Wordpress installation is its own isolated island. It's the same with StatusNet.

Not true, see my reply above. Users on different StatusNet instances can subscribe, reply, and fave each other.

You can see this in action at https://identi.ca/conversation/60337615 (which also points out some of the rough edges).

People have been talking about this for years now, since the early grumblings of Twitter performance problems. There was no mass migration then, and I doubt we'll see one anytime soon. There's no viable alternative to Twitter at this point like Facebook was for Myspace refugees.

The problem is for a migration to happen, it can't just be a little better than Twitter. It has to be a LOT better than Twitter.

>There's no viable alternative to Twitter at this point like Facebook was for Myspace refugees.

well, if one didn't put him/herself into Myspace in the first place, the one wouldn't need to seek refuge.

Exactly. MySpace has shown that these kinds of moves can be fatal to a social network with low switching costs for users, even in the presence of network effects. Twitter's strength is not technology; from a technical standpoint, duplicating Twitter is trivial. Twitter's strength is its community, and they erode that at their own peril.

I think the key thing here is 'what are we betting on?'. The analogy to Facebook doesn't work, as the developers have bet on different things. The apps that Twitter are calling out are ones that are trying to replace Twitter itself as a consumption device for end-users. Can you name one major Facebook app that does that? I can't. When Facebook saw a good pure social innovation in an app, they didn't hesitate to copy it. However apps that have built on top of Facebook to add value in different ways have been actively supported by Facebook for years. Twitter themselves in the announcement make this distinction - do something different than just trying to replace us for end users (e.g. a business management tool, a cool data analysis tool) and we'll support you; otherwise get lost.

I don't really think this is hostility - I think it makes sense for Twitter and for developers; people who've just been trying to replace Twitter themselves were always fighting a losing battle.

The apps that Twitter are calling out are ones that are trying to replace Twitter itself as a consumption device for end-users.

Until today, "Twitter itself" referred to the network, not the clients.

"When you're betting on a platform you don't own, the house always wins."

Except when it doesn't.

It's not a zero-sum game-- the house winning doesn't mean that the bettor loses. Zynga, Rovio, and FourSquare are all pretty fat on platforms they don't own.

The house always wins because the house always has final say. If they want to put you out of business, they get to. The game is rigged.

The point is that in the end, you're never going to be in a better position than the guy who runs the show. You can often win alongside them, and that might be the bulk of developers, but if you're enjoying a better position than the platform operator, that's not going to last forever.

For now.

are they really? Rovio might be the exception, but how fat are Zynga and FourSquare right now, compared to their debt?

Uh, what?

Zynga is making $50M per month: http://www.engagedigital.com/2010/05/14/zynga-revenue-estima... (they've raised $500M).

Foursquare is a bit farther from monetizing... Not sure if I'd bet on them but they would be nowhere without the FB/Twitter platforms under their feet. But they've raised ~$20M on a rumored $100M valuation.

That isn't debt. In both cases, the companies sold a small minority stake of ownership for a huge pile of cash into the company coffers.

That post is nearly a year old. I'd be curious to see if anything has changed.

Not debt, but it's still money owed.

"Not debt, but it's still money owed."

False, and inanity.

> If so, the question becomes: who creates more value on Twitter? Is it the old guard, who use it as a communications and community medium? Or is it the mainstream, following celebrities and talking about sandwiches they're eating, eyeballs to be sold to the highest bidder?

> (edit: My hunch is that, too late, the Pareto principle will be discovered hard at work: 20% or less of Twitter users actually generate 80% of the value. I just can't see it as a bland, empty consumption tool. There's surely peril in optimizing toward that.)

Sounds like what Digg tried. On the other hand, while we can see how that one went Digg had competition.

What's Twitter's competition? What other almost-but-not-quite-as-used (say within an order of magnitude) microblogging service is there out there?

Facebook, for one.

tumblr, for some demographics.

And Convore, though nowhere near the same scale. But that might not matter. The founding members of the Twitter community enjoyed its budding intimacy once upon a time, I suspect.

Convore is infinitely more optimized for community than Twitter was, though it remains to be seen if the structure helps or hinders organic community growth.

Some of us in Finland use http://www.qaiku.com/

This service was built when Jaiku was being run down by Google, and features microblogging with threaded comments, embedded media and language filters

The trick my father has used to succeed in various limited duration markets is to a) realize when you're in one and b) get out before it's too late. Both, especially the latter, are difficult and perhaps impossibly so in today's very well connected world, but he did it for e.g. 8 track tapes in the '70s (for a long window the law didn't require giving the RIAA a cut, just the artists) and C band satellite home video in the '80s (what the pros use(d) for distribution, now replaced by K band DirecTV and Dish Network).

A lot of people also made money in these areas but clung to their companies' obsolete business models for so long their net gain was modest or negative by the end as they rode them down to the crash.

All tech companies, companies really have the tendency to become evil after the become successful. Flattening the opposition or anyone perceived as a treat to their business model, really their bottom-line, will get "flattened". This can include faithful users and developers. This is the basis of capitalism. Microsoft, Apple, Google, Amazon, all media darlings at one time or another, have all exhibited differing levels of douche-baggery at some point.

Twitter, like Apple, right or wrong, wants to have complete control of their platform. It certainly is within their rights and not at all shocking that at a certain point, they would reset certain parameters. Whether they go too far remains to be seem. We probably don't need any more Twitter clients, but who knows. Maybe the slickest, most awesome. voice-activated, built-in hipness meter, Twitterlishous(tm) has yet to be written.

Well said. Twitter's got a lot of goodwill but seems intent on running it down.

When companies started releasing awesome APIs for free, we kinda wondered "what's the catch?". Years later, we're finding out.

The hype surrounding free APIs without formal agreements is the biggest farce in the Valley. We are now in that phase of the cycle where this will become more and more apparent as these companies with awesome APIs get serious about making money(and the free API impeding their revenue plans).

Free APIs--especially the powerful ones--should be seen as "cute things" with little purpose beyond experimental side projects. If you ACTUALLY want to build a company off someone's API, get a formal agreement with that company especially if that company insists you don't need one.

So much has changed since the 90s when access to APIs meant paying huge $. And yet, so little has really changed.

> When companies started releasing awesome APIs for free, we kinda wondered "what's the catch?". Years later, we're finding out.

H.264, anyone?

H.264 was never without licensing fees and I don't think it can be considered a bait-and-switch. The catch is right there in the licensing terms.

I think they were attempting to pursue a bait-and-switch strategy but ended up backing off. H.264 originally came with a "free for 5 years" license for certain kinds of non-subscription online video, which was renewed for a second 5 years in 2010. It was widely assumed that the intent was to start charging once everyone had been lured into being reliant on it.

But partly in response to competitors like WebM, they made the royalty-free license permanent in August 2010: http://arstechnica.com/media/news/2010/08/mpeg-la-counters-g...

It's currently offered "for free" to many people. That's one of the big arguments people always make in the H.264 vs. Google debate — H.264 is just as free for most users, and surely the MPEG-LA would never turn around and change that.

I think that's the bubbliest part of the startup bubble, a lot of cool startups could be crushed overnight if the big player who they orbit changes the free API they sprung up to use.

I slightly disagree. Many people have minted quite a bit of coin off these "cute things" and some started real companies that grew a brand name because of them.

Its fine to rely on them, but you need an exit plan.

"Quite a bit" is very relative. I think there are a bunch of individuals who have developed apps and sold it for some decent $.

But on a larger scale, there are few venture-backed cos without formal API agreements who have not run into trouble. Those are really the ones I was addressing. Like you said, ones who succeed have an exit plan in which case they are not really relying exclusively on the APIs so it becomes a moot point.

A guy building an app on top of an API and the app taking off is still building "cute things" in my book. And yes, every now and then you can make a quick buck from the cute thing.

Twitter doesn't want you to build a twitter client that automatically filters out ads in the twitstream, or doesn't have ads on the sidebar like the official client, or in some other way is superior to the official client.

That is, Twitter is planning to monetize by making their product worse, and they don't want anyone offering a service on the level of what Twitter used to offer.

What is particularly galling is that Twitter has been saying for years that they want to monetize without slapping ads over everything. Now they're monetizing by... slapping ads over everything. And ensuring that no one can threaten the business model they supposedly didn't want to use.

All of this would have been better had Twitter just started off by slapping ads on their service and making it clear how they were going to protect it. Instead, they've grown their userbase (at least among the technorati) by asserting that they weren't going to pursue the exact model they seem so ardently chasing now.

Sure, 3rd party app devs might be "suckers" for having latched on early, but I don't think that excuses Twitter for such a (to use a Gruberism) dick move.

I was about to upvote you until you used the word Gruberism to describe a phrase that is probably older than he is.

Gruber refined the meaning so that it applies to such a move by anyone other than Steve Jobs.

Of course its older than he is. I was referring to his apparent affinity for inserting "dick" in front of everything.

Pedantically nitpick much? ;)

People love useful, friendly accounts on Twitter. My wife and I built up a community of 250K followers just by following that simple formula. If Twitter had a several useful, friendly accounts that every user was forced to follow, they could run the occasional ad, if no single account overtweeted.

While there is no proof that this would achieve a required level of success, had Twitter been doing it for years, they would have learned a ton, and made a lot of money.

Using Twitter accounts to connect to their community and run ads could still be tried, and a level of balance found. (They can respond to inquires from a separate account that accepts no followers so it's replies don't show in anyone's timeline.)

That is, Twitter is planning to monetize by making their product worse, and they don't want anyone offering a service on the level of what Twitter used to offer.

IMO if they have to add some inconveniences to their product to make some dough, I have no problem with that. They have a company to run and sometimes you have to make hard decisions.

I just wish they were more upfront about it than their present passive-aggressive demeanor.

they're trying their best to spin it in a way that makes it look like they have the user's best interests at heart.

The galling thing in this, is that much of the innovation that happened on the client side was initiated by 3rd party developers. Applications like Tweetie, Birdhouse, Twitterific, etc. probably wouldn't have sprung up inside Twitter.

Indeed. The most interesting Twitter client for me at the moment is Flipboard. Now they are stopping the next Flipboard from developing.

Where is my ability as a user to pay for the API and use whatever heck I want to access it?

"Twitter is planning to monetize by making their product worse" - exactly. I wrote a similar blog post about this sentiment a few days ago:


I don't get it. If twitter already provides the top 5 twitter clients, and 90% of users interact with twitter through those clients, where's the fractured landscape and user confusion coming from?

And if the organic trend is towards official twitter client adoption, why are they taking the risk of coming out and bitching at developers to stop making clients? Everyone already noticed the market for these sorts of apps drying up, and has started wondering how serious twitter actually is about keeping a robust 3rd-party platform for the long haul. This just adds fuel to that fire, for no gain whatsoever.

I get what their goal is. They're positioning to make a serious push to get ads/promoted tweets/etc in users' faces, and they want everything in place to be able to shut down or prevent the 3rd party clients that pop up to filter all that out. Okay, fine. But why make the big announcement like this? Why explicitly freak out all of your developers while you're still on a trajectory of solving the problem organically? Why not wait until it's actually a problem, when you're actually seeing increased adoption of 3rd-party clients that ruin your plans?

It's always interesting to watch a company work hard to solve a problem they perceive from their perspective that's completely at odds with the problem the rest of us see from ours. See also: record/movie companies and DRM, Rupert Murdoch and Google, MS and Internet Explorer, etc, etc, etc.

A Friday afternoon release of bad news is a typical PR move to blunt the blow.

Too bad only nerds will care and nerds will be online over the weekend.

"More specifically, developers ask us if they should build client apps that mimic or reproduce the mainstream Twitter consumer client experience. The answer is no.".

How can a company who's user base has grown to such an amount because of third party clients say something like this? Talk about showing a little appreciation. As someone who develops a Twitter client, it is a huge kick in the teeth.

I thought the next paragraph was much more interesting:

"If you are an existing developer of client apps, you can continue to serve your user base, but we will be holding you to high standards to ensure you do not violate users' privacy, that you provide consistency in the user experience, and that you rigorously adhere to all areas of our Terms of Service."

Translation: "we will be looking very hard for any excuse to shut down existing unofficial Twitter clients."

So are they pushing the next wave of developers to resort to scraping instead of playing nice with the API?

I can't imagine that Twitter's service adds enough value to any 3rd party developer that scraping would make any sense. I think Twitter's doing irreversible damage to their developer community.

Wether or not that backfires on them is yet to be seen.

Not only is it a kick in the teeth to developers, but the reason Twitter has traction, spread and growth is because of the diversity of tools around Twitter. The user experience is not all in their hands, and isn't it ironic that this comes a week after the "dickbar" incident? Twitter hasn't really proven they can keep the user experience the way it should be. There was a post last week here on HN about how Tweetie used to never crash and the lastest release was awful... Get your act together Twitter or don't be the elephant in the room, or act that way.

The conspiracy theorist in me thinks that Dickbar and the backlash it received contributed to this. For instance, the iOS official app is the best Twitter client so far despite shoving promoted trends down our throats. One could come up with a better client with no dickbar and everyone would be flocking there. By discouraging new clients, they are discouraging our ability to turn off dickbar.

>shoving promoted trends down our throats.

And are they also "shoving" an extremely popular, free service with ongoing development "down your throat?" I am a fan of free software etc. (even set up a diaspora account... lol), but I don't understand how people expect businesses such as twitter to run forever without a viable revenue stream.

Of course, they shouldn't kill the goose that laid the golden egg, but after a while you have to wonder if these eggs aren't just tin someone took a can of spray paint to! What "gold"? Tons of users are nice insofar as they lead to money. Users are not an end in themselves.

Sorry for the rant but please, people, stop acting like twitter owes you something. (My comments are vis-a-vis users, I do think what they are doing to developers is not right, from a courtesy point of view.)

Where are these developers that ask this?

If you're going to roll on new draconian policies, have the balls just to lay it down as it is. Not wrap it up as if they're doing the developer community a favor when clearly no developers have been asking this at all!.

(Yeah, I know that 'we've been asked' is usual a PR spin mechanism, it's just a shitty way to communicating with people)

The thing is, if you look at the new terms of service, they aren't prohibiting new clients, nor are they shutting down old ones without cause. They just think that this particular niche is already filled, and suggesting that developers pursue other areas. It's perhaps not the friendliest thing to do, but I don't see it as a kick in the teeth.

they aren't prohibiting new clients

If you ask me, that's actually worse. Why such ambiguity? It is pretty apparent Twitter is no fan of clients because of revenue reasons. They will do all of us a service by clearly acknowledging that and simply banning apps of specific nature.

It would sting, sure, but I'd respect a company openly stating financial realities versus a company giving developers a runaround.

Yeah, I keep reading the post, looking for "no new clients" statement that is in OP's headline, and not finding it. I read it as you do; they suggest there's not a whole lot of opportunity in building more clients, but it doesn't read like they'd deny you an API key for it.

Why not let the market decide this though? If someone wants to make a new client because they either see a profit there or have some new ideas I don't think they should be telling them not to.

This highlights an interesting loophole in the libertarian "magic markets" theory: a truly free market includes the right to create closed, authoritarian ecosystems, such as the App Store, or the downloadable game markets on all three major consoles.

Sometimes the model turns out to be inefficient and unsustainable, and such companies are crushed by more open competitors, but not always.

A healthy third-party client ecosystem is the only reliable hedge against future dickbar offenses. It makes sense that they'd want to shut it down. But this move is about what's best for Twitter's business, not the users.

That's why it feels so wrong.

Wow, this is written in a really weird tone, especially considering it's coming from an official Twitter representative. It doesn't feel clear at all what they want developers to do. Either way, it feels very hostile.

It seems like Ryan Sarver, along with the senior and/or PR people who most likely collaborated with him, is trying to channel the tone of Steve Jobs’s “we don't need any more fart apps” speech.

If nothing else, I think it’s advice developers should seriously consider heeding if they want to create value (and generate revenue) using the Twitter ecosystem.

Regarding whether the tone is “hostile” or the message represents “a kick in the teeth,” I think people need to keep some perspective: A hundred and how many million tweets a day?! That's a lot! If I were responsible for the big picture at Twitter, supporting your junior high science fair project Twitter client would not be at the top of my to-do list.

It's hostile because it's purposefully ambiguous. They would catch a lot of bad PR for coming out and stating that they are terminating all unofficial Twitter clients, so they are trying to dance around the bush. There is now a lot of grey area with respect to what is kosher and what is not.

Ironically, it is the ambiguity that is causing so many problems in the first place. It is ambiguity over where they plan to extract revenue that is causing 3rd party devs to choose Twitter-based business models that later threaten Twitter's new revenue model. It is ambiguity in their rules/TOS that is getting 3rd party devs banned (see: Twitdroyd). Etc.

At this point, Twitter needs to realize that the ambiguity isn't helping their PR, its the reason they have a PR problem in the first place.

EDIT: Minor grammatical correction.

> supporting your junior high science fair project Twitter client

What are the support costs? I would think you publish the API and it's up to the client to manage it from there.

Which reminds me, a Twitter client is almost a standard programming tutorial example now. What will replace those?

No, it's perfectly clear what they want developers to do: Fuck off.

Yeah really. It's hard to see this working out well for Twitter.

As an end user (and by no means a power user) of Twitter, I have to say that I don't want a consistent user experience. I want the best user experience for how I use Twitter.

Right now that means TweetDeck, but if something better came along, I'd jump to it. It lets me do things I can't do in Twitter's tools (or that I don't easily see how to do).

So from an 'outsider' on this topic: boo!

Ugh, that tears it. I'm not doing any more Twitter development unless Twitter acquires me. Not hires me, acquires me.

I love Twitter and have several friends that work there, but I am losing all confidence in developing on the platform when I am not a big-name company with an official partnership.

Signed, A 3rd party twitter dev since 2007

Just a heads up, the website in your profile, tweethook has an expired SSL certificate.

yeah... Since the service is now in violation of Twitter API TOS (tweet data resyndication, now only samctioned through Gnip), I've decided to shut it down shortly... no need to renew cert I'm afraid.

As a tweethook customer, I'm sad to hear it.

As the creator I'm sad to say it :( however I will give plenty of notice before this happens. Still running for now.

"in the name of user(experience)".

Back in the Soviet Union a lot of things were happening "by request of the workers", for example an unpaid [and mandatory] additional day of work on Saturday sometimes.

That, and I don't know why they can't just be straight out and say what they mean, instead of wrapping every quasi-statement in 'developers ask us if they can..'

Being straight out when planning a unpopular move would do them more damage then lying. Pretending to do the exact opposite of what you plan to do has been a standard tactic in sports, politics, warfare or any kind of games for ages.

* http://en.wikipedia.org/wiki/Head_fake

"Thanks for all your help getting us where we are. Now, get out and shut the door behind you".

Man, TapBots just can't catch a break. First they are about to announce TweetBot[1], when Twitter for iPhone (a free, first-party alternative to competitors) is announced, and they recently announced the coming of a re-imagined TweetBot[2].

On another note, I don't understand why Twitter are so callous to throw away their community goodwill on a whim. They already have a lot - just look how they took a stand when the data for one of their users was requested by the U.S. government. And then they do something like this. (After [this](http://news.ycombinator.com/item?id=2313152).) And the dickbar!

[1]: http://tapbots.com/blog/business/say-hello-to-tweetbot

[2]: http://tapbots.com/blog/tweetbot/tweetbot-is-coming

According to our data, 90% of active Twitter users use official Twitter apps on a monthly basis.

I notice the omission of the words only or even mostly. I'm curious about the raw percentage of tweets from official/non-official apps.. and the percentage of users who use official applications say, 90+ % of the time.

Presumably, twitter.com is by definition an "official Twitter app."

So if you follow a web link to a tweet while logged in or authorize with oAuth, you've used an official Twitter app.

Given that possibly expansive criterion, 90% is on the low end of what I'd expect.

Good catch. How many people included in this 90% use, say, the official web client from the Apple store in a pinch or perhaps from work on the sly?

From the announcement:

- Display of tweets in 3rd-party services. We need to ensure that tweets, and tweet actions, are rendered in a consistent way so that people have the same experience with tweets no matter where they are. For example, some developers display “comment”, “like”, or other terms with tweets instead of “follow, favorite, retweet, reply” - thus changing the core functions of a tweet.

While I don't like the idea of 3rd-party services treating Twitter as a white-label medium, it's hard to believe this is coming from the same service that is famous for letting its users establish convention, and then supporting that convention. (@, #, etc)

(Edit: s/email/announcement/)

March 11, 2011 - Twitter was fatally injured by a gunshot wound to the foot. Initial reports are indicating this was not an accicdental shooting. Twitter will be mourned and missed by a wide variety of tech enthusiasts.

Interesting that it caught fire during one sxsw and then poisoned itself during another one.

Yup, and there's a dozen hungry startups who are ready to eat their lunch.

I think the biggest material difficulty this will present Twitter is in hiring engineering talent.

I used to get the feel that they wanted to let the ecosystem develop naturally so it could realize it's full potential. That "build great stuff and we'll win" mentality. Now it feels like they've got a "strategy" that they're executing. Without that light touch it just doesn't seem as fun a place to work.

     build great stuff and we'll win
Does that strategy work at all?

To me it seems companies like Google that did that are rather the exception that confirms the rule - building a great business model should be a top priority.

You know what I consider a fun place to work? A company that doesn't have to borrow money to stay alive that also happens to work on cool projects. Even Mozilla is profitable; and even if their partnership with Google goes away in flames at some point, how fucked up is that?

Does that strategy work at all?

Sometimes! Twitter has done pretty well with it so far. Depending on the specifics of your product you might fail. But my intuition is that making stuff people care about, and only monetizing where it's painless is a pretty airtight strategy, yes.

A company that doesn't have to borrow money to stay alive that also happens to work on cool projects

So you think the only two options are "sacrifice your community's goodwill" and "having enough money to stay alive"? I think there's a middle ground where you can leave some money on the table in order to stay focused on the stuff you care about, but still pay your bills.

"As we point out above, we need to move to a less fragmented world, where every user can experience Twitter in a consistent way. This is already happening organically - the number and market share of consumer client apps that are not owned or operated by Twitter has been shrinking. According to our data, 90% of active Twitter users use official Twitter apps on a monthly basis."

If this is true, then what's the point of this announcement. They make it sound like they are already getting what they want...people moving at a rapid pace towards the official Twitter applications. Yet in the same release they come off as scared that the fragmentation is growing. I don't get it.

I think what we are seeing is quite a bit of chasm crossing from the platforms. When that happens, the early technical users (who they don't make much money from, btw) get pushed out or at least _feel_ pushed out.

Yes, we helped build the ecosystem and make it a viable company...but if you read "Crossing the Chasm", it is almost a blueprint for what these platforms are going through and doing...they rely on early technical folks to build a base...but eventually need to reject those early users or marginalize them for sake of profit...very interesting read.

Yeah, as much as this post makes me not want to touch Twitter with a 10-foot pole, I have to admit that they're not going to go any further as a company by catering specifically to developers concerns. I am, however, counting the days until Evan's departure.

Fred Wilson pretty much announced that the status quo was over for Twitter API developers in an article published last April, "The Twitter Platform's Inflection Point": http://avc.com/a_vc/2010/04/the-twitter-platform.html

Notable quote:

"I think the time for filling the holes in the Twitter service has come and gone. It was a great period for Twitter and its third party developers."

Does anyone think this could rejuvenate something like Status.net (http://status.net/) or any other open system for status updates? (I always hopeful about that)

Yep. Last paragraph says it all: Use the API for something besides clients.

Can't wait for Twitter to join Myspace in the slow death club. This move is certainly a step in that direction.

I don't understand why people are so upset about this. It's simple: Twitter Inc. doesn't like apps which are the same as Twitter, or worse. (I don't mean to say that Twitter is bad already: just that Twitter Inc. doesn't like 1. apps recreating Twitter and 2. apps which give a low-quality UI.)

In other words: they urge devs to develop a client with added value. Wether that's apps for "Company Tweeting" or Real-Time Data, it's adding something to the core experience of Twitter.

Most importantly, I think Twitter Inc. still likes "normal" Twitter apps, as long as they have added value: a superior UI. So get devving, and make the new and better Tweetie!

I'm not quite sure about that. Check this part:

"More specifically, developers ask us if they should build client apps that mimic or reproduce the mainstream Twitter consumer client experience. The answer is no.

If you are an existing developer of client apps, you can continue to serve your user base, but we will be holding you to high standards to ensure you do not violate users’ privacy, that you provide consistency in the user experience, and that you rigorously adhere to all areas of our Terms of Service"

This sounds to me like they don't really want "normal" Twitter apps anymore, but have just grand-fathered in the existing ones, while opening the door to closing them down if they don't follow the guidelines.

It's simple: Twitter Inc. doesn't like apps which are the same as Twitter, or worse.

From this announcement, it sounds like they also don't like clients which have different UIs than Twitter now.

I don't think that Twitter Inc. likes any 3rd party Twitter apps anymore cause these don't let them show their Dickbar and whatever will come to the official apps.

You know what I think? There is ONE solution to pass this problem: Integrate such things into the API so EVERY client has to show the Dickbar. This sounds like the logical next level.

As bad as it sounds I think they have some good points about that cause, you know, they want to make some money and integrate some annoying stuff to the official apps. 3rd party apps don't have these kind of annoying things and that means they don't earn a PENNY from those "assholes" BUT they are earning money from their apps which uses the Twitter API.

I know this sounds crazy but I think that's exactly what Twitter is thinking about. They want people to use their official Apps BUT if they use other Apps these MUST HAVE the Dickbar (or whatever) too so Twitter earns money.

Lets see in which direction this will go BUT I have to admit that this "letter" to developers is fucking sharp.

I'm not so sure they are not soliciting alternate UI:

"We need to ensure that tweets, and tweet actions, are rendered in a consistent way so that people have the same experience with tweets no matter where they are."

Then I read further down and they talk about different areas of the ecosystem. Inconsistent message. Do they want outside development or not? Are they intending to release a mandatory graphics toolkit?

Let the market decide. If developers are making clients that are worse than the standard, people won't use them. What is Twitter's product here?

If the market was completely transparent, that would be the case, but that's not how it is. For example Yfrog: it looks absolutely terrible, but still it's very popular. If I made Twitter I wouldn't be happy about that.

A new normal Twitter-user (some who doesn't read HN, basically) might think something like this: "Oh, so this is where you post images on twitter? Doesn't look great. How do I login? Ah nevermind, I'll go back to Facebook." At the same time, TwitPic adds something to Twitter, in such a way that it fits in with it.

Bottom-line: users will connect 3rd-party apps and experiences to the API provider, so Twitter is right in making clear that they don't want crappy apps.

A Consistent User Experience. I believe I've heard that somewhere before... http://developer.apple.com/library/mac/#documentation/UserEx...

Translation: Look, the VCs are expecting a payout at some point and the only way we've worked out for making money is through ads. Now that's not going to fly if people can just get some sort of 'ad free' client from you guys. Our hands are tied.

PS. Thanks for helping us get this far!

There is no surprise with the aid of #dickbar. I am not going to use Twitter for any serious purpose anymore. This bottom line has violated by basic requirement of information freedom.

Just seeing this issue I really think that there is still spaces for even more Twitter clients. Last months or so there is one called littlecosm.com which is a client+game type.

It is like they simply don't want people to avoid looking at their public timeline without promoted tweets and dickbar. What a shame.

I have to believe that this means that Twitter is looking at turning on the revenues and likely with in stream "sponsored tweets" or some other similar type of nonsense and they don't want third parties stripping out these ads.

Its always been strange that they'd let 3rd parties be a primary interaction model since its very hard to monetize other peoples clients.

>since its very hard to monetize other peoples clients

Google API key seems to be at least one way to do it.

I just remembered an article written by Alex Payne (al3x), former engineer at Twitter (and used to work in the API):


It definitely sounds hostile, but I can see where they're coming from. They want to encourage more creative uses of their API, rather than just flooding the market with hundreds of mostly subpar client apps. They've basically reached critical mass as far as third party clients go. Any additional clients are not going to increase their user base anymore. What they really need is more use cases for the Twitter stream. I don't particularly agree with this strategy of leaving a lot of developers feeling like they just got slapped across the face, but it will be interesting to see where the API goes now.

Twitter became twitter thanks to its community that always created value and contributed most to its future path. The invention of hashtag and @ were also contributed by community. the power users that always contributed high quality content and conversations to twitter.

The developers that contributed best applications for twitter ecosystem.

but now that it has justin bieber and other celebrities along with millions of their followers twitter feels like it does not need that core contributing community anymore.

To me it looks like start of demise for twitter.

Well, I'm done using Twitter. Their engineered experience sucks, but I do like third-party clients.

As much as this sounds like the beginning of the end, I do have to agree with one gripe:

> For example, some 
developers display “comment”, “like”, or other terms with tweets instead of 
 “follow, favorite, retweet, reply” - thus changing the core functions of a 

Very true. I have accidentally tweeted by “logging in with Twitter” and then “commenting” on comments. Despicable behavior, and it should be stopped. (Alert! Previous statement is narrow in scope!)

On the contrary this helps quite a bit. There was a cloud hanging over anyone doing anything with Twitter whether they would find their product in competition with the Mother ship itself.

It also explains apps suddenly losing access to the API and then regaining it. Twitter was asserting its control over standards.

I think if you aren't building a client, following the rules and adding value you don't have anything to worry about. To me that is great news.

To expect enduring openness from a privately-owned medium or enduring stability from a single point of failure is naïve.

This is an interesting turning point for Twitter and for API's of free services in general. Building apps based on an ecosystem that you have no control over has always seemed like a substantial risk to me, and this assumption is starting to take shape. Twitter's recent move is an example of what happens when a company is ready to monetize their content and wants to cut out the "middle man" so to speak. It's pretty clear that they don't want third parties encroaching on their ad revenue, and while sad, it was inevitable.

I see this as a very disappointing move by Twitter for the fact that they are alienating the very developers that helped build their brand. This is a glaring warning sign for all applications that are based on an external platform. If nothing else, this should serve as a lesson to all developers that free API's should be utilized with discretion.

Surprising: "According to our data, 90% of active Twitter users use official Twitter apps on a monthly basis."

You can read that as, 90% of active twitter users visit the twitter.com website at least once per month.

Not sure how this matches up to the 90% number:


This is based on tweets composed.

They put out some statistics on this last year: http://blog.twitter.com/2010/09/evolving-ecosystem.html

That's a shame, because I am still in the market for a good Android client. Guess there won't be one, ever.

Tweetdeck is pretty good.

Thanks, will try.

Can anyone point to language in the revised API Terms of Service (http://dev.twitter.com/pages/api_terms) that implies "no new Twitter clients"? I'm reading it, but I can't find the smoking gun.

I believe that was pulled from Ryan's email:

"... developers ask us if they should build client apps that mimic or reproduce the mainstream Twitter consumer client experience. The answer is no." ... "If you are an existing developer of client apps, you can continue to serve your user base"

That's fine, but that's just the author's commentary. The blog piece is just an announcement about the new ToS, which is the defining policy. We need to look for the language in there, not his blog piece.

While I understand Twitter's motivation for doing this -- and no, I don't believe it's about consistency -- I think they could have expressed at least some token of appreciation for the devs who helped get them to where they are now.

It was bound to happen ... and about time, hopefully it gives someone enough inspiration to come up with something to compete against twitter.

I want to see someone give twitter a run for their money. They're sloppy.

I read it this way:

There's now 75k registered Twitter apps. You're not going to be successful making a generic client. Think bigger: there's plenty of room for success in other Twitter verticals.

(and boooo for the fake FUD headline)

And yet Twitter interaction needs to be the "same way everywhere", "consistent", "same experience", rigorous adherence, etc. Seems anti-innovation to me.

I think it's rather that they want to own the client space, and are setting a "high bar" - i.e. a way of removing the competition by fiat. But if you supply content that they can make money off of, by all means, continue to sharecrop.

Not all twitter apps are clients. I would guess that most are not clients.

Agreed. I've got a few little Python and shell scripts laying around that have talked to their API at one point. By a loose standard, each is a client. Perhaps there are 74,800 little shell scripts or integration adapters out there, but not what an ordinary person would call a client app. That I'd believe.

Twitter claims preventing user confusion as a motive yet the move from old to new Twitter on the web is the biggest source of confusion. I mean, come on. Teaching people an interface isn't rocket science. Do something simple, such as put out a video a week showcasing use of the new interface. Heck, have user contests to come up with videos showcasing the new interface. Or go all out and develop an awesome training and help system, and integrate it into the interface. If users have a problem, do something about it!

Anybody who in the first place developed and designed a product around a third-party's ecosystem and API has what they have coming

in principle, you're right.

In practice, i must bring MS as a positive (!) example. Not that they were saint or anywhere close to it, yet they didn't even come close to use legal sh!t like EULA/Terms of service to hit developers on their platform. MS provided their own competitive products and used [sometimes dirty] tricks to win.

This has got to be at least partially about UberMedia.

Still makes me glad I don't depend on Twitter's platform at all.

This is why Twitter is soon to be bankrupt. They don't understand how to stop insulting the people that actually provide the content and value. The value isn't the technology, but the people developing and using it.

On one hand, I agree in terms of providing a consistent user experience.

On the other hand, I disagree and think people should be able to build whatever experience they want.

This is becoming an increasingly common mistake for tech CEOs: adopt Steve Jobs' attitude without being Steve Jobs.

But I don;t want consistent experience. I dislike the experience and want it done my own way!

i guess that's what investors call 'feature' as opposing to a standalone product - turns out that many of these clients funded by them are features of twitter


The precedence of things.

That is the deafening thud of the other shoe dropping

has google not bought twitter yet?

Funny timing considering how there's been a noticeable degradation in the Twitter app quality since Atebits got Borg-ified by them. Their plan thus far seems to be:

1. identify best client out there

2. buy it

3. ruin it

4. outlaw all other (well, new) clients

5. ...

6. profit!

consistency and ecosystem opportunities

couldn't he just write "you're fucked, we're the boss" without all this corporate BS speech?

So if no new twitter clients are allowed, but existing ones are "grandfathered" in, does this mean there will be a market of buying and selling these grandfathered twitter clients?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact