Why #fixtweetie is important

Firstly, it’s not about fixing one iPhone app. Sure, Tweetie was my favourite iPhone Twitter client. But there are others. And that’s the point of the apps ecosystem: if you don’t like one any more, you go and find another one. I’d paid for it, but it’s not the three quid either that makes it important.

This is why.

Go back to an original, pure concept of Twitter. A great big cloud of 140 character tweets. You can do anything you like with those characters. And those tweets; using a range of Twitter clients (note: all, apart from Twitter’s own web client at Twitter.com, are independent products). The clients help you filter down the bits of the cloud that you want to see.

Add some conventions: that a word preceded by @ is a user name, and a word preceded by # is a way of labelling content that can then be easily linked together. Importantly, these conventions could have been generated from the user community. Other conventions, like the use of the slash ‘/’ (as in /via) – are in their infancy. “Slashtags”. Love that. With a little ingenuity, other character-based syntax conventions could emerge: ^zipcode, to make up an example.

If the convention is picked up and adopted, great. That’s evolution. Everyone’s happy. In time those client applications can be tweaked to get greater value out of the convention (just as most now automatically return search results based on clicking a hashtag).

Twitter does of course do more than just host a big cloud of tweets. They save you the bother of creating a register in your own client application of the users from whom you are interested in seeing tweets. Though this is theoretically possible, it would be a pain each time you changed client (or switched from PC to mobile device) having to synchronise those lists.

So Twitter helps you, by running some basic user registration functions, including the ability to create a record of those you follow. Because these registered user identities live within Twitter’s domain, they can also support some other functions, like direct messaging, blocking and so on.

And lately, they can be assembled into ‘lists’, which also live alongside your profile, in Twitterland.

And the last thing that Twitter do is make pretty much all of the above information freely available to be sucked out and interpreted by these aforementioned client applications, through an API (a way of easily getting information out of Twitterland to do clever things with it).

The last big rumble in the Twungle was when Twitter tweaked the API so that instead of being able to see all tweets emanating from a user, you’d only see their @replies if you were also following the user they were @replying to. Although this definitely makes for a neater way of seeing content, there were some benefits to occasionally viewing an entirely unfiltered stream of information (such as a serendipitous comment from someone you followed to someone you’d never heard of, which might prompt a conversation with them). Yada yada yada.

Some, like me, argued that this was a violation of the ‘purity’ of the information; it would be perfectly easy for your client application to make that @reply filtering choice for you, assuming it could be supplied with the “whole picture”; but Twitter decided to take away that choice. The whole argument was captured on the hashtag #fixreplies.

Taking away choice = bad, IMHO.

But never before did Twitter cross the line into specifying what format those 140 characters might take (other than some forays into metadata – already used in any case for datestamps, client application identifiers, and so on – and latterly for geocoding). But the way the core 140 characters worked? No, that was up to you. If you wanted to put RT… at the start of your tweets, good on ya.

That’s changed with the new RT feature. Read the Twitter line here: (as reposted by @atebits – Tweetie’s makers).

Interesting. Yes, another play on cleaning up your stream – as with #fixreplies. But what’s all this stuff about ‘dictating’ etiquette? What happened to the evolutionary adoption of things that worked? Surely if a long stream of identical tweets was annoying, client applications would evolve that could suppress these at the client. Even I could code that… And if they weren’t identical? Well, that would be because people put in little personal comments along the way with their RTs. So you’d lose those, obviously. (Or have to throw them away depending on how you tuned your duplicate tweet suppression on the client.)

So perhaps this is all a bit of misdirection. Instead of focusing on just how helpful the new RT format is, try working out for yourself what’s really behind new RTs. It shouldn’t take long.

And the fact that hooey like that post from @ev gets bandied about, instead of the honest answer about Twitter’s move into content shaping (and it won’t be the last one), is why I felt strongly about #fixtweetie*, and why this has become a blog post.

So, did you get there? Hear the jingling dollars? And as Dan Moon wisely points out – we can’t overlook that Twitter is a business. Of course they can impose content parameters and design controls if they want. If we don’t like it we can all go back to, erm, Facebook and MSN. Or the Next Big Thing (am working on that…he he).

I’d just ask for a little more transparency about all this – saying “it’s a nicer experience” doesn’t really cut it for me. What do you think?

*a hashtag I began in order to argue that client apps should allow the user as much choice as possible in the formatting of their tweets, so that etiquette could continue to evolve, rather than be imposed.