Wanna grok XBRL?

I’ve got my XBRL consortium hat firmly on here.

I know we are starting to get the attention of the technical community when W3C folk like Dan Connolly start posting del.icio.us comments like this one:

XBRL specifications

nice use of URIs… their schemas resolve, they have good persistence policies in their schemas. nice looking test suites. Looks like quality work. Hmm… their use of taxonomies goes beyond XML Schema and uses XLink; I wonder if it maps nicely to RDF/OWL

Thanks Dan! There will be a lot of people that have worked really hard over a number of years glad to hear you say it. And these posts here and here from Tim Bray, really help the cause.

But like Tim says, XBRL is not exactly the simplest thing to get your hands around. We set out to capture the semantics of business reporting. Not a subset of it, and that means that 80/20 has never really worked for us. We didn’t think that investors or corporations would be very grateful if we had said “XBRL lets you instantly, accurately and unambiguously define and disseminate some of what you want to disclose to the market.” So the specification turned out to be a bit bigger than anyone would really have liked.

Furthermore, real users working in real projects, from Korea to Spain, to the UK, Canada and the US want more! Like versioning, formula, functions and (gulp) rendering. So, in April the International Steering Committee announced the creation of the XBRL Standards Board, charged with improving the quality, consistency and stability of XBRL International IP. XBRL’s technical product has always had some formal guidelines to help things along, but now that the standard is in such extensive use, it’s far more important that we consult at each step, take account of what the market has developed independently and really focus on how easy the work is to implement.

So right now, inside the consortium’s mailing lists, we are consulting on a technical roadmap, a brand new, shiny set of policies and procedures (modelled on what we thought were the best bits of OASIS and W3C practices) and are calling for participation in a number of brand new working groups.

Not a member? Join! It doesn’t matter where you are, your contribution can make a difference.

Work at one of the 480 organisations that are already members and want to get stuck in? Great!

Either way, sign-up for one or more of the Working Groups. There is no better way to really understand a standard than to be part of its development (even if you are just lurking in the background). While the XBRL 2.1 specification is stable, these new work products (like formula, functions and versioning) are all new (optional but very useful) modules to XBRL. You can get involved from the ground up. And learn about the base specification while you go along. See you soon in a working group conference call! Or come along to Philly!

Now means Now!

I can’t help enjoying this post on Information Arbitrage… even though it dissed some of my mates. But, err, Roger? I’m not a luddite, however unless something very funky happens to the way that Atom/RSS works (and yes, Tim has chimed in with a potential solution) you really, really, really need material news to be distributed simultaneously. As in… at exactly the same time. I pointed it out a bit earlier this week, over here. That (together with identity authentication and persistence) are some of the goodies that make wire services important.

REALLY fast disclosure…


Jonathan Schwartz is writing about Regulation FD, which governs the release of material information by companies in the United States (and, in Europe at least, the rules are pretty similar). He points out, (with Sun General Counsel, Mike Dillon talking the same point up yesterday) that for most individuals, the internet is far more accessible than an audio-conference or press release. The problem is the notion of simultaneity.

It's absolutely true that if I, as an individual, am considering investing in Sun or any other company, subscribing to company blogs would be a great way to understand the business. As long as I know which of the 4000 or so blogs at Sun I should concentrate on (and I realise that Eve and her fellow identity experts can ensure that I know with complete certainty which they are), presumably it would be a great way for me to notice material statements.

With Atom or RSS, I'd be pretty sure of getting that information within a minute… or at least minutes of the announcement. Depending on latency.

Within a minute? Pretty sure? Depending on latency? Hang on!

That might be fine for individual investors. But what about professional investors? If a professional investor in a mutual fund, a hedge fund or proprietary trader on Wall Street or in the City was faced with the mere possibility that they will probably get the information at probably the same time as their competitors…

Well. Ummm. There might be a few problems with that.

The thing about the "anachronistic press release" is that it can get information to thousands of media outlets, professional investment houses and investment infomediaries absolutely simultaneously. And I mean simultaneous… which is to say, within a tick or so of an atomic clock.

So, if a trader (or her algorithm) is sitting next to a terminal watching material news flash by, she can be totally confident that the only difference between her and her competitor is the time it takes to act on the news. All market participants get the news at exactly the same time.

In the interests of full disclosure, I should mention that we are part owned by Business Wire, who run the proprietary NX network that drives nearly half the press releases that contain material news in the United States and a big chunk of the rest around the world. They are great people, who take their role in the information supply chain very seriously, and run a bunch of smart technology that ensures that simultaneous disclosures really work.

I believe that in a world that is increasingly transmitting material information in computer readable form, especially XBRL, the need for true simultaneity is only increasing. So I, for one, just don't see how the rules can change to allow what Jonathan is asking. Not that it isn't an admirable goal. Just that the internet can't deliver on it just yet. Thoughts? john DOT turner AT corefiling DOT com

[Updated] Tim Bray weighs in.

XBRL Italy kicks off

Did it seem like all the XBRL news last week was coming out of Washington? As usual, there were developments everywhere else. They were meeting in Rome last week to kick off the official Italian jurisdiction, with Infocamera, the companies registrar, set to co-ordinate the development of taxonomies for collecting financial information from companies all over the country. According to this communique, the work will be completed by March next year. XBRL in Italy has been slow to incubate, but I’d say they have their timing just about right!