back
73 comments
Twitter's image would be better served by the bandaid-removal approach. Everything bad - all at once, rather than this drip-drip-drip of changes perceived to be hostile towards its developer community. You have to figure they've got a road-map of where this heads, so why not do it all at once and consolidate the pain?
"Severities should be dealt out all at once, so that their suddenness may give less offense; benefits ought to be handed ought drop by drop, so that they may be relished the more." - Niccolo Machiavelli
I'm not sure that Twitter can afford to be feared - they need to be loved. Twitter doesn't lord over a Renaissance Italian fiefdom; their serfs can migrate to other city-states at will.

But first, there needs to be another city-state.

Because incremental change is way easier to manage and deploy and test than massive ones.
In technical rollouts, yes. But twitter's ticking people off with a bunch of new policies...
Actually I agree with you in this case, given the massive amount of attention on Twitter now, but in situations of lesser scrutiny, the 'drip-drip-drip' is often the the more cunning move.

See also: http://en.wikipedia.org/wiki/Creeping_normalcy and all the related concepts, eg boiling the frog etc.

Except you know, if you boil a frog slowly, it will still jump out.

http://www.fastcompany.com/26455/next-time-what-say-we-boil-...

IE needs more actual examples, instead of metaphors that turn out to be completely wrong in practice.

Most of what is creeping these days is way too debatable...
What's more interesting is that each of these little drips is turning into front page news material. Every time the giant shifts in its sleep, the villagers run amuck.
It was about time, always wondered how 'via' improved user experience in any way.

EDIT: since I'm getting downvoted because some of you don't agree:

There's like 140M active Twitter users (as of March 2012) and like, 30 clients (guesstimating). If I assume there's like 1M developers on Twitter (which I don't think it is), this feature is still useful to less than 1% of their user base. Not exactly a "must have" feature IMHO.

It helped people discover new Twitter clients by seeing what other people were posting with. I'll admit that this benefit may not have been anywhere near worth the added clutter and screen space, but it was good for something.
Exactly. I found out about new clients quite a few times because of the via stamp.
Not only that, but it helps them discover new apps. For example, 'posted via Instagram would be useful', as would 'posted via Camera+' etc.
and it helped me to remove useless/automated tweets from timeline using a home-made greasemonkey/user script.
Just like Google’s “I’m Feeling Lucky”, its purpose was more to communicate something about the brand rather than to serve some useful functionality. In Google’s case, it communicates playfulness, in Twitter’s case, it was geeky openness. Clearly this is not true for Twitter anymore.
According to Twitter's own numbers, there were over a million registered apps... last year. By their own growth numbers, assuming they didn't increase (very unlikely), there would be almost 25 million registered apps today.

http://blog.twitter.com/2011/07/one-million-registered-twitt...

How many of those apps are people using T or running ThinkUp instances?

It's doubtful that many of these apps offer any functionality that could qualify them as a client.

The via stamp told you whether a tweet was posted via desktop or mobile, which can be useful to know.
It's not that we don't agree, it's that we (or, personally, I) don't see how you wouldn't know what its benefit could be. That you would think "it's about time" was worth posting at all based upon not knowing the benefits of the feature, even if those benefits only accrued to other people.

But hey, I suppose it's easier to say we don't understand your point to the degree necessary to agree with it, than it is to suspect a fundamental flaw in your own reasoning.

It certainly encouraged innovative and creative developers.
And how exactly is that?
In terms of social interaction it adds quite a bit. Imagine the tweet: "This party's awesome".

Seeing if it was published from Twitter for Blackberry, Tweetbot for Mac or EventualTweet tells me a lot about yourself as a person (and believe me, stuff like this makes/breaks relationships).

Are we still at the stage of judging people based on what smartphone they own? Good grief.
I thought 140 characters making or breaking relations was somehow frivolous. I did not realize that those clients had a relationship weight associated with them as well.
What are we in grade school?
Yeah, it's of no interest except to nerds, who are always a minority.
Regarding the downmod, did I say something false?
I'm sick of this kind of twitter stories. This is just a hype about Twitters' internal affairs. Their business. Why on earth this is top news or even on frontpage? They have just skipped showing 2 words. If the story was about "how twitter hypothetically screwed and made 2 man hours work of deleting two words from tweets 2 man days of work" I'd be more willing to read it.

Please - this is not about hacking, this is not about technology, this is not even about entrepreneurship - I would really love stories about twitter's opensource projects, twitter's technology choice, twitter's software development methodologies, twitter's devels' thoughts about samsung-apple case. But this is not news worthy. At least not news worthy for Hacker News.

Tell ya what I'm sick of.

This kind of comment.

Why on earth should I care what some anonymous person thinks is newsworthy? You've got a counter right there showing you exactly how many people find it worth reading.

If it's interesting, read it and/or comment. If it's not, scroll past. Flag if you want. But the only thing less newsworthy than a given story is your assessment of its newsworthiness.

Look, there is something wrong about this.

This story has 97 point at time of my writing, 44 comments. It's top story. Learning C with gdb story has 297 points, 80 comments and it's on third row.

So why are you not just skipping my comment and wasting your precious time on answering me?
> Please - this is not about hacking, this is not about technology, this is not even about entrepreneurship

Sorry, but it is. It's a masterclass in how to do a bait and switch. Start out completely open, let open developers build your community, then shut them all off once you don't need them any more.

This is not even about entrepreneurship

At some level consider it is? (1) symbolic the evolution of strategy; or (2) its part of a larger narrative on openess vs trusworthiness; or even (3) its might ve PR issue.ETC.

Also, its always interesting to look back and put this in context: like this http://dashes.com/anil/2009/12/the-twitter-api-is-finished.h...

I really liked the via tag for one purpose.

Buffer.

I liked to know when folks were live tweeting versus time delayed tweeting as it gave me context. If they were live, I'd respond. If they weren't I'd just consumer the content (if interested) and move on.

"First-party clients" is certainly a new euphemism in this context.
This is very clearly the end of Twitter as a "platform". Now that Twitter is huge, they don't want or need developers or 3rd party clients. They want to centralize everything into a monetizable service. That would've been fine if that was their plan from the get-go. The bait-n-switch they're doing disingenuous at the least.

So what now? Status.net hasn't gained traction, and it's obviously too early to say anything about tent.io, but I am very, very hopeful that eventually a new open, decentralized platform/protocol will emerge as the next-generation social communications platform to replace both Twitter and Facebook simultaneously.

dramatic much? it's not like 3rd party clients stopped working
You'd have to be silly to not expect what Twitters next move will be.
"We've always been at war with third-party clients."

Twitter needs to shit or get off the pot. Either they want a nice big ecosystem and they make money by leveraging their ridiculous amount of 3rd-party traffic, or they listen to the biz dev guys with their nice haircuts and just force all traffic through their official points. It's so wishy-washy, with the Tweetbot Alpha token pull, and this it's very clear they're saying this is the new way but through so many layers of subtext it's like reading a management presentation on blue-sky thinking.

This is an unfortunate move. I found it very useful to see who was tweeting using an app that suggested it's the real person, versus tweets coming from large, paid enterprise apps using Twitter's API -- big "social media management" software that a marketing organization would use to schedule tweets at pre-designated times.
The API changes stated that they wanted all clients to look the same and to have a similar experience. Could this be a hint that they want to rule out 3rd party clients slowly but surely?
I think they are more concerned about users avoiding ads than they are about a consistent user interface.
Not too long now before twitter releases pricing?
Pricing for what? Users? That would be an unequivocal death knell. Developers? Possibly. However, it's doubtful that Twitter stands to gain any significant revenue from charging for an API. Their audience for advertising is so much larger and more lucrative, a paid API would be of trivial importance to their bottom line and a waste of time.

I don't think they'll charge for anything. Doing so would relegate them to the same, minuscule potential user base as App.net has.

This should not be surprising to any of you.
one more step closer to monopoly