The following post is a facsimile (with minor edits) of a position paper/provocation paper presented at the Open Source Schools Think Tank held in London on the 26th March 2010.

Just in case you were wondering, all opinions are mine and not those of my employer.

These are not fully realised or worked-through thoughts, but hopefully contain some useful / contentious principles to provoke some discussion, or at least make the reader think. While reading please bear in mind that it may be unfinished!

What's the problem?

As someone who started working in school-level education systems ten years ago, the one thing which has remained constant throughout has been my amazement at how a certain system has dominated and, in some schools and in many ways, has shaped the way children of all ages are taught and how teachers of all levels of experience work. I am of course referring to the dominant MIS system used in the UK. I have no particular beef with the system , but the fact that it is in some spheres so universally reviled by staff (and hence used ineffectively or in ways which are less than optimal). However, the tortuous process of moving from the MIS system in question to any comparable system are such that most schools appear to feel content (compelled?) to stick with the devil they know. Such a situation can only lead to a lack of innovation. Consider how this landscape would look if the opposite was the case - i.e. that schools could migrate from one MIS system to another with no more than a month of upheaval, which consisted of a certain amount of data transfer (possibly through open XML schemas) and some appropriate retraining of key staff. If this was the case, innovative and (let's face it) better tools would be rewarded with schools flocking to them - the "the school down the road thinks it's great so we're having a look" modus operandi is well developed in terms of technology in schools. Since these tools would operate on data (which would be held in a common, open standard) then it would be possible to use a single tool to access all of the data, or even (for those schools who really wanted to cherry pick the best bits) to use different tools to carry out different tasks - for example, one element of a particular MIS to "do" attendance and another from another provider to "do" assessment. Such an environment would nurture innovation - indeed, innovation would be the main driving force in this sector, which isn't something that can said to be true now. The market leader is by parts counter-intuitive, lumbering and nowhere near as agile as its competitors. However, it's still the market leader, which gives you an idea of how difficult it must be to migrate from it.

Where is the role of Open Source in this? OSS-based school MIS tools don't currently come close to the worldwide reach and use of a tool such as Moodle for any number of reasons - they are either fragmented, the efforts of a small group of individuals, or in the case of the leading contender (SchoolTool) so beholden to a complete insistence on being run on a particular Linux distribution that this immediately disqualifies them from use in most schools-and almost certainly rules them out of a wider market in the UK.

Again, here an insistence by a body such as Becta on Open Standards for MIS data would be a way to make this work and could also stimulate innovation in the MIS market. SIF does some of this, but is primarily concerned with ensuring that data can be piped between systems and doesn't, from this user's perspective, have true interoperability at its heart. The discussions (verging on squabbles) and splintering of vendors and other parties around SIF (see SALTIS for an example of the issues here) shows the potential for factions over any standard, but for the good of schools, learning in general and the MIS industry's reputation. There are examples of excellent interfaces between the existing proprietary systems and Open Source tools - the limiting factor in these is the closed and obfuscated data in the existing systems. Think what could be done if the data was open and easily (and securely) accessible. The arena of school Management Information sorely needs a ringmaster with a whip and a set of house rules - such an approach could be a shot of adrenaline to a lethargic industry and an opportunity for the creativity & innovation afforded by OSS to flourish & benefit schools in a measurable way.

A comparable solution to a different issue?

The issue of ePortfolios has waxed and waned in importance as online parental reporting has lumbered over the horizon, however there was (to my mind) a chance to have opened up the realm of how ePortfolios could be supported and developed on a national, and even international, basis. Here's how it could have worked...

• Becta / the DCSF / whoever commission what could be The World's Largest Slice Of Cloud Storage - with resilience, backup (possibly devolved to local data centres on an LA or RBC basis). Hey, if they want, that commissioning could be done through a procurement process...
• The storage is accessed via a common (and Open) API - an Application Programming Interface which means that, like many online services, data could be written to and read from the storage using a variety of tools - whether web-based, mobile applications, or functions built-in to desktop tools such as the MS Office or OpenOffice suites. Flickr is an example of a tool which, in some ways, exhibits characteristics which might be deemed desirable in an ePortfolio system of this type.
• Any "vendor" or "supplier" would need to offer a tool which used this API to both write data to and retrieve it from the hosted service. This would open the market up to a range of providers - in the knowledge that the stored data would be accessible

This openness would deal with many of the issues around the portability of ePortfolios. For example, it's fairly clear that there's considerable mobility across arbitrary county & other administrative area - in 'my' County of Buckinghamshire (for example) many parents choose not to put their children into the selective secondary system - so there's an immediate issue of a child's work being "held" in Buckinghamshire and then being inaccessible in a form - other than an as a read-only archive - once the child ends up in Oxfordshire or Hertfordshire. In the current paradigm an ePortfolio is "for life" as long as a pupil's "life" stays in one administrative boundary.

Such an approach would see the "traditional" providers of learning tools (including those 10 9 companies on the Becta Learning Platform Services Framework) offer ePortfolio tools and allow any open-source tool which could interface with the open API to be used on an equal basis. Such an approach fosters innovation, creative approaches to learning and would be a healthy, strong, competitive environment which would exist in significant contrast to the MIS market described above. As everything was immediately transferable - since a portfolio could be accessed and modified using any of the tools - and it might be the case that a whole host of best-of-breed tools are used by an individual user or institution, if each of the tools offered something unique and innovative. The role of Becta (or a similar organisation overseeing the storage) would be to ensure it was resilient and maintain the API.

This last element - agreeing what the API does or how it works - looks like another opportunity for the HTML "standards" wars of the late 1990s, when Netscape and Microsoft decided that "innovation" meant "a standard of ours not supported by anyone else". Well, with a little thinking and the carrot of "the funding goes to where the standards are" this could potentially be dealt with. Becta's Learning Platform Services Framework was supposed to have interoperability - the principle that meant that (in theory) an institution could switch providers almost seamlessly, or that a learner could move between institutions who were using different Learning Platforms and retain control and access to their data. It's an open secret that none of the providers was able to offer that true interoperability - see the case of StudyWiz's demise for more discussion on that - however having to fully meet a common API could ensure that interoperability was at the heart of such a project.

Truly portable portfolios for truly lifelong learning would be no mean achievement - but what would it require from both Becta, the Open Source sector and suppliers of proprietary software? A first start would be a commitment to open, competitve innovation with an acknowledgement that if the playing field were truly open, then only the most useful and valuable would survive. Could the proprietary and Open Source sectors stand up to that sort of test?

Author's Note:

I'd appreciate comments on this, but particularly on the second issue - as I'm aware that this post could be link bait for those who want to post re: SALTIS/SIF/SIFA - so if you have a post on that subject, please ensure that it's original thinking rather than just a re-statement of arguments which have been or are already being thrashed out elsewhere. Many thanks in advance for your understanding!

[Comments below, or on Ian's blog - Miles.]