XBRL as Platform

I don’t often go off on a bender on this blog – I reserve that for my Friday night curry and cider binge. But here I’m going to noodle a bit about XBRL as a platform. For those of you who are already lost (e.g. Fashionistas – But, like, how can XBRL be on the bottom of my shoes?) here’s my definition of a platform in software industry terms:

A platform is an application development and delivery ecosystem.

So in this post, I’m going to outline the history of platforms in order to get to my point: that XBRL is a platform too.

An application development and delivery ecosystem implies a number of characteristics that define a ‘platform’:

l          the endgame is to enable people to use applications to help them do stuff better

l          you need a solid foundation to start from otherwise the platform is built on sand

l          it’s collaborative and involves a wide range of stakeholders – developers, users, distributors, documenters, influencers etc.

l          it needs ways to promote and deliver apps (e.g. today’s appstores)

l          it isn’t all about making money (e.g. Open Source movement)

Platforms build on each other or to put it another way: stand on each others shoulders. Conceptually the development of platforms is not that different from Maslow’s hierarchy of needs. Up until the Internet came out of the shadows in around the mid 90’s – there were 3 types of platform that built on top of each other:

1. Hardware platforms

Without some form of hardware. a platform doesn’t exist. So we had mainframe, mini and then PC platforms. As each platform became cheaper and less proprietary, the software development and delivery ecosystem grew. And today we have mobile phones and tablets as the latest hardware platforms.

2.  Operating system (OS) platforms

Arguably DEC’s VMS was the first true OS platform that attracted independent software developers onto a hardware platform but UNIX, MS-DOS, Mac OSX and now LINUX all qualified/qualify as OS platforms to provide the foundation for software to be built to work on specific hardware.

3. Application standard platform

I class MS Windows as an application standard platform because it was more than just an operating system platform to run on a hardware platform (the PC). Windows also defined the look and feel and some of the core functionality and behaviour of the applications that were built to run on the Windows platform.

But what characterizes all the platform implementations outlined so far (except the mention of UNIX/LINUX) is that they were proprietary and essentially non-collaborative. The first crack in this proprietary non-collaborative platform ecosystem was the arrival of the Internet, what we now recognize as a vast and largely ‘open’ platform that has proven fundamentally disruptive in platform innovation terms by leading to:

4. Social platforms

The Internet has spawned a new generation of social platforms. Social platforms have seen the phenomenal growth that they have because they tapped into the pent up demand from ordinary users to share their user-generated content via the Web and from a new generation of software developers/hackers/dabblers who could finally create ‘theirapp’ by leveraging the key element of a social platform: the application programming interface or API. Although they were great ideas and well-executed – Facebook, Flickr, Twitter and the rest – have flourished as platforms by building communities that now act as the consumer base for a universe of apps built on their APIs.

Although they don’t quite fit into my classifications Wordpress is a platform and so is Google. Individual tools and applications like the ORACLE database or the SAP ERP system also qualify as platforms. But now I want to get to my point.

5. Data platforms

While millions of words have been written about the Internet driving the growth of social platforms, it’s often forgotten that the Internet is also driving another kind of platform  - the data platform. I’m not sure if anyone tracks this stuff (please tell me if you know someone who does) but I would guess that a significant new data repository comes on stream on the Internet at least weekly if not daily. Arguably the semantic web is nothing more than an attempt to turn the Internet into the ultimate data platform (which makes me wonder if Sir Tim has ever read The Machine Stops by E.M. Forster?)

Some of these these data repositories are large enough or interesting enough to justify being called platforms in their own right. One of my recent posts mentioned the UK Government’s COINS data repository that has just put millions of data points online about public spending in the UK. In fact at their heart, Facebook, Flickr, Twitter et al. are all fundamentally data platforms – the crowd creates the data via social networking – but the end result is millions of pokes, photos and tweets stored in the cloud and more or less accessible directly via an API or indirectly via the apps that leverage the API.

Which is why XBRL is a platform or more specifically a data platform. All around the world XBRL is powering the availability of structured data online that is more or less publicly accessible and by doing so enables an application development and delivery ecosystem to build around it.

Yes, I know. Even after more than a decade, the ecosystem is relatively immature. But this is where the really disruptive opportunity lies for XBRL, as a platform not as some esoteric technology play, to make a fundamental difference to the way businesses control and communicate their financial, sustainability and other operational data.