Hugh McGuire

Evan just launched an open source twitterish thing called identi.ca. He’s got tons of traction in a few short days (that surely have been long for Evan and the rest of the crack team at Controlez-Vous), including lots of interest from luminaries such as Dave Winer, Tm O’Reilly and others. So: first, a big kudos to Evan.

Identi.ca also has its pooh poohers, including the knife sharpeners at TechCrunch, who wrote a lukewarm piece called The Problem with Identi.ca Is That It’s Not Twitter. And that’s been pretty much the line of those less than impressed: It’s got fewer features, why bother, everyone is on Twitter, why would they leave, and: who cares if it’s open source, it still needs to be good. etc.

Which speaks of breathtaking short-sightedness, not to mention, a total erasure of the last, oh, half-decade plus of the most recent Internet history.

So here is my take: Identi.ca is not an alternative service to Twitter; it’s an open microblogging platform. That’s a huge difference.

(OK, identi.ca is an alternative service, and laconi.ca is the open microblogging platform behind it, but for the sake of this article lets say they are the same thing).

It’s not difficult to find salient parallels, either. Biz Stone and Ev Williams’ pre-Twitter project (before Odeo) was Blogger. A great platform for making blogs. But it turned out that an open source version, WordPress, was far more powerful, versatile, and compelling. Bloogger is still popular and still a good solution for many people. WordPress though turned into something different, and arguably much more important.

Will identi.ca be as successful as WordPress? Who knows, but if you think that microblogging is important, then *something* like identi.ca will be successful, and it’s the best candidate so far, that I know of. Again, it’s not a service; it’s a platform (and an open one at that).

Let me give two small examples:

Mobile Microblogging in the Developing World

I met Joel Selanikio, a doctor, epidemiologist, and software developer, at the Stockholm Challenge, where Joel’s project, EpiSurveyor won in the Health category. Here is a short description of EpiSurveyor: an open source mobile phone platform for collecting health & epidemiological data, which is being implemented by the World Health Org among others (which gives great cost and efficiency improvements over both paper/pencil- the usual method – and expensive commercial software & consulting).

Joel and I had some great discussions about mobile as a platform in the developing world: ie, why spend money on OLPC in the developing world, when every teacher already has a computer in their pocket … a mobile phone. The smart thing to do is to develop applications for the “network-connected minicomputers” people already have, namely: phones. Let’s develop for the tools that exist rather than the ones we’d like to imagine.

We also talked about Twitter as a web platform for mobile communications; interestingly, Joel thought Twitter was puzzling (I’m putting that mildly, I think he said it was a waste of time!), whereas for me – other than the time-wasting/communication aspect, Twitter is compelling as a platform for developing web-based/mobile enabled communications, the specifics of which I can’t put my finger upon. One example that I provided was the Tower Bridge Twitter stream in London. This is a trivial little project that scrapes the web for info on when the bridge is opening/closing and what ships are sailing through. The example itself is irrelevant; the point is that one can imagine useful bits of information being transmitted to your mobile device in such a way.

Here are some interesting facts:

• many/most people in developing countries have mobile phones
• many/most people in developing countries DON”T have: computers & high bandwidth net access
• mobile phone-enabled microblogging tools might be the perfect platform for information distribution/communication in such a place
• a microblogging tool could be used for any number of useful things, beyond “just” the run-of-mill social communication, eg:
  • price discovery, for say exchange rates, market prices
  • boil water alerts
  • traffic reports
  • education (say, informing parents of homework? still wondering about this one…)
  • health alerts
  • news headline distribution
  • who knows what else?

Identi.ca can become a development platform to do all this, and much more that you and I can’t think of. Luckily there are 5 billion people on the planet who will be able to take identi.ca/laconi.ca and build/improve upon them. While Twitter, Plurk, Pownce and all the rest are constrained because they are just closed services, that do only what their owners wish them to do.

Archiving Links, and Search Rank

Here is another area of significant interest. I wrote a while ago lamenting that Twitter has replaced del.icio.us for me as a place to archive interesting links. While Twitter does a good job of letting me share interesting links with friends immediately, it doesn’t serve as a useful archive in the way del.icio.us does. So that means:

1. unless I post twice, I lose a structured archive of links I found useful
2. because of ubiquitous use of URL-shortening services in Twitter, the web is also losing the significant work of URL-sorting/ranking that we used to do by blogging about interesting links, and putting them into del.icio.us (etc).

The other night, I had dinner with Larry Sanger (thanks for the invite, Mike), and Larry was batting around some compelling ideas about opening up the search space.

And that had me stewing about things, thinking about Identi.ca and my problems with Twitter and (no longer) archiving my links. It would be “easy” to do this in identi.ca, by specifiying:

• that this identi.ca post contains a link (this can be inferred by the existence of a url)
• that i want to structure it somehow – eg using #hashtags
• that i wish to archive this – ie an RSS stream of my categorized links, that could easily be fungible with a more centralized or decentralized bookmark depository (del.icio.us or other) …

You’d also want the system to keep track of the true link, rather than the shortened on.

This is not just more useful to me, but Important in how the web/google/searches assign value to different URLs.

Now, theoretically all this could happen at Twitter. But Twitter is a company, with a few guys and (apparently, gasp) ONE mysql database (with two slaves). They have enough problems just keeping the fail whale at sea.

Identi.ca, on the other hand, belongs to us all … and if I had the chops and the interest (I have the latter but not the former, and not the time) I could code something up that would do the trick, and pitch it to Evan, or install my own identi.ca instance on my server doing what I want it to do.

Summary: Identi.ca is important because it is a microbloggin development platform; not because it is an alternative to Twitter. Whether or not identi.ca and the open source codebase laconi.ca succeed I can’t predict. But something like this *will* succeed because mobile-enabled microblogging might just be the most compelling new communications space, especially in the developing world where access to mobile phones is almost ubiquitous, while access to computers and bandwidth is limited.

[Incidentally, and as an aside, all this has much to do with why I thought Steve's comments on my iphone post were off-base ... there may be many people who lament that their shiny gadgets are too expensive, but given all this above, it's clear that there is much exciting work to be done in mobile web, much of it important, and with crappy data plans Canadians are excluded from this area of innovation, which is what pisses me off - luckily, tools like identi.ca mean we webbers have a new development tool to do interesting things in the space].

A service that listened for url’s in your micro-entries and auto-added them to del.icio.us (or other service) shouldn’t be too difficult to build.

Figuring out tags might be tricky, but the content of the entry (and its permalink) could be put in the “Note” field.

Adding options would also be difficult – it’s easier just to auto-add everything.

tags can be hacked using hashtags in the body of the post… “#politics #video” etc… can be extracted from the post easily.

alternately you could add tags to your posting mechanism, and keep them only in the links rss feed.

Also, there is valuable information in the sharing of URLs. A lot of Google’s algorithm is determined by people giving token nods in the direction of certain sites that have no social value. In this way Google still gives the advantage to established brands and the status quo.

Often my students would link to things just because they were a brand they recognized – not because they had any inherent social value. And I see this a lot in the blogosphere too.

Whereas URL’s in a microblogging community have a social value. They are social objects rather than footnotes. (Because there is no space for footnotes).

I sometimes wonder why TinyURL/snurl/whoever hasn’t used their data to give us indices of the most social URLs that day, etc. It’s the kind of thing O’Reilly has been looking for for ages.

part of the identi.ca conglomerate is an open url shortener, ur1.ca which plans to make that data available… so one hopes that it’ll go in the direction you suggest.

[...] Go read Hugh McGuire’s post, “Why identi.ca matters” – this is basically the post I wanted to write. Also of [...]

[...] Hugh Mcguire has a thoughtful essay on Open Microblogging and what it could mean for the development of mobile computing in   Micro Moblogging or Why Identica Matters. [...]

[...] – Identi.ca: à découvrir, le premier service de micro-blogging en logiciel libre (Open source). Alors que toutes les plateformes sont fermées, Identi.ca offre la possibilité d’installer le logiciel sur son serveur et d’y apporter ses propres configurations. A surveiller car cette approche pourrait être bien utile pour les entreprises ou même les écoles. J’y reviendrais dans un futur billet mais en attendant vous pouvez lire les commentaires de Dave Winer et Hugh McGuire. [...]

mobile computing nowadays is not yet very powerful compared to netbooks but time will come that it would become like that.*’~