Access to OER discussion
From Bjoern Hassler
Contents |
In preparation for a email-based discussion on a UNESCO-IIEP mailing list about access to OER, I was reflecting on the issues that might come up in the discussion.
It was suggested that we might consider the problem in two parts:
- Firstly, the identification and description of the main problems associated with access and infrastructure
- Secondly, reflection upon and exploration of potential solutions.
[edit] 1 I. Issues
Regarding identification of issues, I like to consider two cases:
- Case (1): There is no internet access at all, perhaps not even computers, or even no electricity / no reliable electricty.
- Case (2): There are computers, and there is some internet access.
For (1), there are many issues to be considered, such as cost-effective computing: Considering the whole life-time of the equipment, and the substantial on-costs of any ICT installation, one needs to consider cost-effective maintenance, cost-effective power, appropriate internet access, etc. It's certainly a scenario that cannot be solved with good will alone, but needs substantial investment of one form or other. Hence, to get OER materials into that scenario, requires broad lateral thinking, considering local circumstances, weighing paper against ICT, etc.
Let's now consider situation (2). In my view, with regard to OER, situation (2) is quite tragic. Tragic in the sense that there could easily be more access to OER, but there isn't because access issues aren't understood. In many places, there is increasing access to the internet (and thus potentially access to OER), but this internet access is very slow and/or expensive. For instance, access might be through mobile internet (say 20kb/s on a GPRS modem) or expensive (e.g. through a BGAN satellite receiver, at $7 per MB, which means about $2 per click on an average web site / OER site, sometimes up to $10 per click!).
Now, if OER materials needed to be so large, then there would be nothing that could be done in terms of online access, and we’d need find different ways of getting OER materials to the users. However, the tragedy is that the information we provide is needlessly large: Plain text is dressed up with huge images, huge amounts of javascript/css, stuck inside pdf files, audio files are encoded at 128kb/s stereo, when 32kb/s mono would be well sufficient, ... All this leads to information/OER being inaccessible. At the same time, if the OER were presented well (and appropriately formatted), they could be accessible.
To contrast cases (1) and (2) again: I would argue that in case (1) there are huge infrastructural barriers, and there are simply costs associated with this. Change needs to happen, but there will be costs. However, for case (2), I would argue that it's mainly awareness that's needed (or at least awareness would get you a good way towards your goals). Of course raising this awareness also has a cost associated with it, but raising awareness could bring about a step change.
Let’s consider the issues around case (2). I like to draw an analogy with licensing issues around access to OER. Suppose you went to a major foundation to get your OER project funded, and you were asked about the license proposed for your OER materials. Suppose you said: "Well, we'd like to release the materials as 'all rights reserved', because really people can always ask us if they want to use something. And when we get asked, we can always give extra permissions."
Because CC is paradigmatic in this area, the funder would think very carefully whether to fund your project. That is to say: There has been considerable discourse around licensing ("accessibility in terms of licensing"), and a good model has been found ("CC"). So there's now a kite-mark for legal accessibility (the CC badge), and existing OER resources can be measured against this.
With low-bandwidth accessibility, there are no such standards. Not all OER projects aspire to provide truly global resources, at least some OER projects do aspire to provide global and accessible resources, and perhaps obtain funding on that premise. Yet, when you look at the OER materials, or the websites hosting these materials, they are huge. By huge I mean: Taking minutes to load, or costing (with some sites) $5 per click.
In more techical language: The typical web page these days is about 300kB. In terms of accessibility over low bandwidth, this is huge ($2 per click over BGAN). The recommendation for low-bandwidth accessibility is page sizes of about 25kB.
[edit] 2 II. Thinking about solutions for scenario (1)
The second question to reflect on concerned potential solutions. I'll first address scenario (2), where there are some computers and some internet.
[edit] 2.1 (A) Raising awareness
As alluded to above, I believe the quickest win is to raise awareness among content providers for providing resources in an accessible way. Similarly, I believe that one should raise awareness among content-users for ways in which OER can be obtained faster and most cost-effectively, for instance through using bandwidth optimisation on their local network, by using a service like loband, etc.
Some of this could be done through discussions (like the Access2OER discussion about to happen on a UNESCO IIEP mailing list). One should also approach the relevant bodies, such as wikimedia foundation, OCWC, etc.
[edit] 2.2 (B) An "Access anywhere badge"
In my view, just like a "CC" and "WAI" badge, there needs to be a "low bandwidth accessible" badge, perhaps better called an "access anywhere badge", that sets standards against which web sites / OERs can be measured. Things to covered by such as "low bandwidth accessible" certification would be
- Having a light-weight website (or making sure that your website works with a service like loband, and explicity linking out to the low bandwidth version thus created)
- Providing resources in multiple formats, including text-only, low-bandwidth images/graphics, low-bandwidth audio, and low-bandwidth video
- Providing ways of downloading resources in bulk (for instance by providing machine readable resource indices, such as rss/podcast feeds, or more generic xml feeds)
- Providing ways of off-lining more complex resources (such as downloading the whole or parts of wikieducator/wikipedia)
For each of these requirements that your site complies with then gives you a "low bandwidth / access anywhere" star on your site, as part of an “access anywhere certification".
There are some resources available, that cover these issues (such as the Aptivate low-bandwidth guidelines), and it might be possible to do some of what is suggested in (A) and (B) without any additional resources/funding. However, to launch a proper certification, some sort of programme with associated funding would be needed. Likewise, further suggestions below are likely to need funding of some kind.
[edit] 2.3 (C) A training programme / OER materials on low bandwidth
I believe that providing some OER training modules covering low bandwidth issues around OER would have a good impact, and could provide very valuable in terms of raising awareness and instilling good practice.
It should also be said that while OER users who find themselves in scenario (2) (with computers/internet) will gradually get cheaper internet, many users in scenario (1) (without computers/internet) will gradually move into scenario (2). Thus the low-bandwidth issues are here to stay for quite some time. Hence making an investment in training would really be worthwhile.
[edit] 2.4 (D) A content delivery network
In various earlier discussions, a content delivery network (CDN) for OER has already been mentioned. Institutions with good connectivity use a CDN when the demand for resources exceeds their available bandwidth. Institutions with less bandwidth are exactly the same situation, it's just that their bandwidth is exhausted sooner, and they may need to use a CDN for more or different types of resources.
A CDN (whether it's a global network with a lot of infrastructure, or just some more informal caching arrangements) could provide a very good solution. It would not just be a solution for North to South distribution, but particularly distributing resources from the South to the North, as well as to other places in the South would be helped, and would thus contributing to overcoming the important issue of lack of internet exchange points in the South).
In many ways all of these ideas would require some sort of dedicated project, and some level of funding as well as commitment from existing bodies. It would be interesting to see in the forthcoming discussion on Access2OER whether there is a sufficiently well defined set of problems, and corresponsing agreement on what needs to be done, in order to formulate a proposal for a more far-reaching “access 2 OER” project.
[edit] 3 III. Thinking about solutions to scenario (1)
With regard to scenario (1) (lack of infrastructure, power, computers) solutions certainly requires investment, and progress is difficult with good will alone.
My overall interest for this article is access to OER, and addressing power issues ("We need more power stations.") or global networking issues ("Our country needs more ISPs / more lit fibre.") is beyond the present discussion.
However, consider this question instead: "We have a certain amount of power, and we have such-and-such access to the internet, what should we do?". In other words, suppose we look a the existing infrasturcture, and ask ourselves what kind of equipment/training/etc would need to happen to provide access to OER.
To put this another way, consider this statement: We cannot run computers, because our electricty is expensive.
In my view is it wrong to just look for solutions in the 2nd part of that sentence. Of course cheaper electricity is useful, but it is also likely to be beyond local control. However, computers that consume less power may well be under local control, and thus an achievable improvement.
To stay with this issue for a moment: I also think it is important to address the issue of power consumptions of full blown desktop computers, as well as 2nd hand computers: There are circumstances where a donated 2nd hand computer could be quite useful. But particularly where power is scarce or expensive, it becomes less clear that the 2nd hand computer is really more cost-effective than purchasing a new low-power computer instead. A low-power computer could have 1/10th of the power consumption of a full blown desktop computer, and may be more than adequate for the intended uses.
So overall, such issues around ‘appropriate ICT’ are worth discussing, and there is some really valuable work that could be done in terms of determining best practice, for instance in terms of promoting easy-to-maintain low-power computing, sharing connections to reduce costs etc. More concretely, suggestions for actions could revolve around providing guidance and training materials (similar to (C) above). What equipment, what operating systems work in which circumstances?
Also, a content delivery network (as in (D) above) could be very valuable here. In general, content delivery options should probably be able to work fully off-line, but should scale to up when online access becomes available.
Just like the suggestions for scenario (2), a concerted project, with associated funding would be needed to address some of these issues in detail.
