Tuesday, April 01, 2008

What Becta / DCSF / someone should do about ePortfolios

[this is quite a long post, but please bear with me, I think it's worth it, if only as a vehicle for reflecting on things]
A few days weeks ago I was sat in a session on ePortfolios at the Naace conference - there were many parallel sessions on at the time and I had quickly and quietly ducked out of the Learning Platforms one (call me chicken if you like). The reason for this is that the whole area of ePortfolios is looming on my mental radar at the moment, for a number of reasons:

  1. there are vague but seemingly definite targets on it (from Government);
  2. there are many definitions about what an ePortfolio is, and not one of them appears to concur with any other;
  3. I don't want us to make a decision which hitches our wagons to whatever's in the news around the time we make that decision.

So, the session is going on and I'm tapping away wondering if being in this session was the correct decision. Then, at some point when the discussion was centred around what should happen next in the area of ePortfolios (which begs the question - what's happened already?), a thought wandered into my mind and wouldn't leave.

Here's the problem: the specification for Learning Platforms is, in parts, one of the most vague things you could ever read. By "Spring 2008" (when is/was Spring?):

every pupil should have access to a personalised online learning space with the potential to support an e-portfolio

Apart from the obvious questions:

  • What exactly do we mean by personalised?
  • Is potential enough, or should it really support an ePortfolio?
  • If (as is indicated above) a "personalised online learning space" is different to an ePortfolio, does a VLE count as one?
  • If that's the case, what's the relationship between a VLE and an ePortfolio?

At the moment, as far as I can tell, the "market" in ePortfolios for school-age children is packed with vendors of all shapes and sizes claiming that whatever they're selling ticks the above box. In some ways, they're right - in that the target above is so vague that you could almost give every child a 8GB USB memory drive and be done with it, so anything with the label "ePortfolio" on it would tick the above box.

Now, some of the tools on offer are substantial, have learning (rather than simply file storage) at the core, and allow reflection, publication, and many of the other tools which crop up once you start to dig, even at a shallow depth, into the literature on ePortfolios. Some are simply file storage with a customisable front-end (no thanks).

The major problem (and one that mirrors the situation with Learning Platforms) is that there's no true interoperability - a pupil who might create a stunning portfolio of work at a primary school risks losing this on transition to secondary school - and the only option might be "getting a copy of it on CD" or, at best, as a structured .zip file with files in specific folders (assuming someone at primary age remembers to place their work in meanginful folders). Also, we have the problem that the "stuff" - whatever's in the portfolio, the associated learning reflections and other information would need to be moved between portfolio hosts if different providers are providing the primary and secondary schools' ePortfolio systems. Now that (so the "ePortfolio" provider would have you believe) is a reason for everyone (across a Local Authority, an RBC, a nation?) to use the same system - which doesn't really sound like true interoperability to me, more like so-called "lock in" on a grander, and possible more frightening, scale. From the market's point of view, it gives vendors a chance to clean up - anyone who gets (for example) the whole of London, or the whole of an RBC, has a licence to print money, as theirs is the only option.

One of the problems is, that the guidance on this is so vague about what an ePortfolio might be that the concept of "interoperability" between different systems is clearly a pipe dream. Of course, any system should be able to read files, but it's the nuances, the processes, the learning associated with those simple files which is as, if not more, important. Coupled with the fact that the Becta guidance just throws that phrase "potential to support an ePortfolio" in there.

So here's what I think Becta should do. It involves:

  • funding;
  • guidance;
  • standards;
  • above all it takes the guts to do something which is quite directive, but at the same time gives schools and vendors the freedom to be creative.

Put simply, in my ideal world / parallel universe Becta would commission what might be the world's largest amount of online storage space (including hosting, backup, redundancy, whatever it takes), using some of the former Grant 121 money designed to help schools meet the vague targets mentioned above. Whatever size they decided it should be, it would be. Wherever it was sited, it would be sited.

What Becta would specify would be the ways in which this space could be accessed - i.e. the ways in which any tool would pass data to and from this space, using common (open) standards based methods. If you've got a flickr account, you'll know it's possible to allow other applications to access the data stored in flickr - well that's what I'm thinking of for this space. Vendors would have to use these standards to be able to write to and read from the space. What exactly they "wrote" there would be up to them - and schools could pick from any vendor knowing that their tool or product could write to this national ePortfolio space.
Schools would have administrative control over their own space and staff and pupils would simply be "hooked" onto the online "peg" of their school and then moved on as appropriate. LAs would have access to their schools - etc. etc., with a bit of thought you can work out how this would work. Of course, users would ultimately own their space (and who knows, the Becta space might be configured to export to currently popular and free online storage in case a user wanted their own backup).

The advantage for users, schools, LAs, etc would be that because any tool could write to and read from this space, that an individual moving from one institution to another could use whatever tool that those institutions used (the same or different) and still access and manage their information - in the same way that a whole load of applications can access my space on flickr (if I let them). This removes the fear of "what if I move schools?" or (horror!) move across an LA or RBC boundary, in which case you're almost certainly back to having your work on a CD.

I think that this approach would increase creativity in the area of ePortfolios as it would be obvious to a vendor that any school, LA, or RBC would be able to move to a better product if the current one wasn't up to scratch - an ease of movement which has obviously been missing from the dominated-by-one-vendor MIS market for years. It also gets around the cumbersome and (in my opinion) wrong view that an ePortfolio should live inside a Learning Platform. At a stroke it would at remove the issue about interoperability - since ePortfolio tools would be interacting with a single central data store, rather than each other.

Is this achievable? I think I know the answer to that one - but I think it's desirable, and as the definitions around ePortfolios are so vague & multifarious that Becta will offend most people if it comes up with a definitive one, could it / should it provide the tools to get all sorts of portfolios working in harmony, rather than simply sowing a few seeds, turning its back and then having to manage all sorts of different, competing crops and their accompanying weeds?


  1. As one who noticed Ian tapping away at the back of the Naace session I am delighted to see he made such a productive use of the time! ;-) An excellent summary of the real challenge we are all facing in this area - and an imaginative solution.

    An eportfolio MUST belong to a learner, not a system, organisation or technology. Other organisations/individuals need to be able to access it to achieve different purposes, such as reading, mentoring, verifying or accrediting. These may be coming via different technologies. An eportfolio may even be drawing content FROM these other systems and technologies. Sound anything like a combination of LinkedIn and Facebook plus some secure central repositories?

    Keep the thinking coming Ian!


  2. This is a fantastic idea and one I'd love to see happen (even if I know in my heart it's unlikely to). It also opens up the possibilities of schools (and learners) being able to write their own applications that interact with the - can't remember what you called it - big repository thing. They can do it for Bebo and Facebook, why not with this.

    Is there any merit in getting some kind of petition or other going to get this idea into Becta? Online, of course :0D

  3. Ian, great idea, but does Becta have a big enough vision for this?

    An alternative approach would be to develop a business model which would mean this could be done as a commercial/non-profit venture.

    Companies that wished to develop interfaces to the data could, perhaps, be asked for a license fee.

    Or there could be additional services available to people that pay for the "professional" subscription (taking the Flickr analogy further)

    Not sure how that would work yet.

    However, I have one issue with these big centralised services. Who owns my data? What happens to my data if I want to move it (let's say I move to another EU country, for example). How do I control who can get to my data?

    This is exactly why I set up my own portfolio (blog), on hosting space that I paid for. I probably would never keep it on a company system or a free service (even Blogger).

  4. Yes you are right with the Flickr analogy and I wonder how data can be moved around and more importantly stored for decades. I can't see Becta doing anything as sensible as this so no doubt we will continue to see confusion and more than a few dodgy solutions. Spot on article, really struck a chord with me.

  5. Interesting post. I feel that with all the talk of SCORM compliance, all other interop questions were left unanswered.

    Unfortunately it seems that it's still going to be VERY difficult to migrate a school from one VLE to another - the resources might migrate individually but none of the VLEs I've come across have either:

    - A standardised method for storing/exporting site structure (that could be cracked very easily if they wanted to, but there is no incentive)

    - A standardised format for exporting the entirety of a pupil's VLE data (i.e. work submitted/marks/portfolio/blog/message inbox/whatever)

    Your suggestion would negate the need for the second part, but the same issues would need to be answered to make it a reality. Primarily it would seem to require a standard schema for the types of information we would like to store. Given that scenario, the software to host the data could be released publicly and hosted anywhere - perhaps at LEA level. If a pupil changes LEA, the API could be used to handle migration between LEAs.

    The alternative (low-tech) approach of having a standard export format is possibly more likely, but doesn't have the advantage of exposing the data to a wide array of external services which could bring massive benefits as you suggest.

  6. Hi Ian,

    I very much agree with your analysis of the current problem (no-one has defined the terms and there is a lack of interoperability) but I am not sure the huge central super-repository is required. All you need is a new Secretary of State who decides to pull the funding and the whole thing goes up in smoke - look at the COL portal. There is also the problem of the band-width that would be required: much portfolio data (especially video and graphics) is much better held locally.

    If you have good standards for data exchange and kitemark products which support the exchange of that data (maybe, in the long term, over SIF hubs) then you will have a resilient, decentralised system, yet the data will still be able to follow the student between institutions, onto different platforms and into different geographical locations. If you have interoperable systems then you don’t need central provision of anything other than the technical specs and certification procedures.

    The question is, who defines the data standards? However broad Becta's vision, I think there are dangers in doing it centrally - particularly of imposing a particular pedagogical model on everyone else. I do not agree with Tony Parkin when he says "An e-portfolio MUST belong to a learner". It depends what you mean by "belong" (it may cover a range of rights, e.g. to delete, to add, to show to others, to view yourself) but if you are dealing with portfolios for assessment, then there is a sense in which exam certificates which may go onto that portfolio belong to the awarding body as much as the candidate - and can certainly not be added at the behest of the student. There is another issue over whether it should be possible for others (e.g. a school) to add material without the permission of the student - teacher notes or negative assessments. I'm not saying what the answers to all the these questions should be – what I question is the “MUST” – I think rather than being dictated by a central authority, the pedagogical answers should be evolve in response to demand from students, schools, colleges, universities and the consumers of such data (employers etc). I suspect that the answers that emerge will be nuanced and will be different in different contexts.

    A unpublished Becta report in 2006 suggested that there were four types of portfolio: for learning (probably involving reflection and self targeting), for assessment, for presentation (or show-casing), and for transition (e.g. UCAS). It is a fruitless debate to say that portfolio should be about one of these and not the others. All that the experts need to do is to develop the terminology to make the distinctions so that teachers know what they are getting and what to ask for – then let the market, driven by professionals, decide.

    Sam raises the question of SCORM. There are two different interoperability issues - interoperability of learning content, normally produced by teachers, commercial publishers and public sector bodies; and interoperability of learning product, produced by students (possibly using productivity tools packaged as learning content). SALTIS' ( www.saltis.org ) first priority is to work through the problems Sam mentions with learning content. But I am keen to get work going in the autumn on portfolio standards. TAG is already working with BECTA and BSI on a standard for the automatic submission of coursework to awarding bodies – a kind of portfolio for assessment. There is also MIAP, which is substantially providing a high-level portfolio for assessment. But SALTIS also includes companies which are involved with portfolios for learning, which would require a very different kind of standard.

    The advantage of SALTIS playing a role in this process is that data standards can be driven by commercial companies (small ones just as much as large ones) so that any data standards to emerge from it do not become redundant academic or bureaucratic exercises, but are tied to real products which are responding to real demand from teachers et al. It can also draw on the technical expertise to be found in the commercial companies. At the same time, SALTIS is working closely with BSI and BECTA, so that any standards are formally accredited, mandated by government, and will respond to and inform the government agenda.

    I see the portfolio space, at the moment rather loosely defined, being filled in gradually by a mosaic of different but compatible pieces, the TAG standard being perhaps the first.

    If anyone else would be interested in contributing ideas to an interoperability standard for portfolios for learning, do drop me a line at crispin.weston@alphalearning.co.uk . As I say, I hope that SALTIS will be in a position to start substantive work in the autumn.

    Best wishes,

  7. Many Local Authorities as you suggest, have gone down the route of buying one e-portfolio solution for all the LEA schools in an area, this is understandable given the belief that this will make way for interoperability, students moving between schools. It seems an obvious thing to write, but more students are going to be moving from secondary school to FE/sixthform College or secondary school to HE Institution, than are going to be moving from secondary school to secondary school. Decisions made by FE and HE institutions seem to be made without taking into account systems within secondary schools in their area. The need for interoperability at this juncture seems to be ignored.

    Is "having portfolios working in harmony" really what we should be striving for? Looking at the wider context, many of us, not just young people, have invested considerable amounts of creative energy producing work which is best accessed using a computer. How many bits of work do we have back ups for, how much of this work can be taken from its framework, application and accessed somewhere else. Take for example this blog, how much work would be involved in moving it to some other provider, could BECTA really be that other provider. Have most bloggers got a back up somewhere else? Could they display it to someone without that person having access to the internet? How many bloggers would know where to start ensuring their work was portable and independent of a supplier?
    Many of us still think first in terms of CDs or USB drives if we were asked to gather together our work into a portfolio. If this is how we think then surely this is how students will think as well. Isn't the requirement to capture the work/portfolio and put it in your own hands is stronger than facility to transfer it from one institution to another. Or am I speaking for an older generation?