By Dan Jellinek.
With some assistive technology software such as specialist text-to-speech screenreaders being relatively expensive to buy, moves have been underway for some time to widen the role that free and open source software can play in supporting the IT needs of people with disabilities.
However, while free software sounds like a great idea – and an even better price – its development is not always so straightforward, delegates heard at the recent eAccess 13 conference in London.
A discussion group chaired by University of Southampton internet research fellow EA Draffan focused on “the exciting expansion in the number of strategies involving free and open source software that can support access to technology”. It set out to examine “the robustness of this type of software; ongoing support and maintenance issues and whether this is a feasible route for government and other organisations to take in the long term.”
Free and open source software are notoriously hard to define, but in general terms, free software can mean both that which has no cost to the user, and software which people are free to adapt for their own ends. Open source software (which is sometimes, but not always, also free in either or both senses) is software whose source code is published so everyone can see it. There are various overlaps between these categories, and many varieties of copyright licence under which software can be distributed so as to remain “free”, but on the whole both free and open source software are either no cost or much lower cost to use than proprietary, specialist assistive software which only the developers or owners can sell or licence.
Free and open source software has advantages other than price: the ability for others to adapt it means, for example, that software produced in one language can be taken by another group and translated into another language. And its functionality can be tweaked to suit new groups of users by volunteer developer communities, often working together using online code exchange and development platforms, such as GitHub and SourceForge.
Sounds good, but the fact is there has been a surprisingly low level of free and open source software to emerge in the assistive technology arena, Draffan said. “There is lots of good free software in other sectors – what is stopping us in assistive technology?”
One of the main problems is funding, she said. Although developing software can be relatively affordable, open source software needs to have a community of users built around it to help maintain and strengthen the code, Draffan said. “You have to build your community, to blog, tweet – you have got to have your community stay with you. How do you do that? It is terribly time consuming, and without money, you can’t sustain it.
“The actual capital expense of developing something is 33% or even less of total cost. The real money goes on promotion and marketing.”
Some projects are supported by companies who allow their paid developers time to support voluntary open source projects. Google, for example, allows its engineers a certain percentage of their time to pursue creative projects such as writing open source software; and many other giant IT companies such as IBM support social projects.
For individual projects, however, support can still be a problem. And there are other barriers, too: free software licences sound simple but there are so many variations such as BSD and GNU that the field can be hard to understand, Draffan said. This matters: getting each licence exactly right to suit the aims of each project is “horrendously important”, she said.
Other issues include inevitable limitations to projects supported by volunteer communities: commercial tools, while they tend to be more expensive, often have access to better development funding and hence are generally more fully-featured and support more standard platforms, Draffan said. Open source software, on the other hand, often does part of a job very well, or supports one operating system or web browser well, but does not work so well across the board, she said.
But isn’t the world big enough for all kinds of software? Or, as Draffan put it, “Can we achieve a nirvana where we have a two-tier system?”
To do that, we need more encouragement for open innovation, she said. This could partly be achieved by promoting existing success stories, to fire people’s imaginations, and also by wider distribution of the free and open source assistive technology that already exists. “I give a pen drive with free assistive technology to new students – why is someone not doing it for people in care homes?”
The government may have a useful role to play here, and does support some open source software because it wants to avoid supplier lock-in and control, Draffan said. However, this support has so far mainly concentrated on well-established platforms and tools such as web content management software, she said, “whereas with assistive technology, we are asking for someone to create something new.”
One answer might be to look for software projects that have already been started but need support to reach the next level, she said. “We need to look for something already sitting there that is waiting to tip, that needs a bit of a nudge, so you can validate it, you can say here is what we think is the most promising one.”
The results will definitely be worth it, she said, in an ageing society. Some might think assistive technology is a small area of software development aimed at a small user community, she said, “but is it? If I am thinking of 25% of population that is getting old, it is not so small.”
NOTE: EA Draffan’s slides and other speaker slides and notes from eAccess 13 are available to view online at:
Entries (RSS)
Comments
I think it is inaccurate to say that open source solutions tend to work only on one platform, as opposed to proprietary systems which are more widely available. In both cases there are examples of portability, but I believe there are way more examples that are open source such as Dasher and eSpeak, which work in everything from cell phones to tablets and computers.
Licensing is also not such a big issue. It is only a problem for companies that wish to get the public relations credit of free software without paying the GPL price of decentralized control.
Finally, projects such as NVDA and F123 (the latter I am responsible for), need support to scale up, but do not need any validation beyond the feedback they already receive from happy users. The proprietary software model requires expenditures that are unsustainable, so the government needs the free and open source community as much or more than this community needs the government.
This is a global problem. When I was IT Trainer at the Kenya Society for the Blind in Nairobi during 2009-10 a large part of my work was writing guides, training materials and tutorials for free and open source assistive technology programs.
Why did I do this? Firstly the primary application that KSB was using was the time limited evaluation copy of JAWS, which was frustrating and unsatisfactory for both clients and staff. Second the quantity and quality of the guides, training materials and tutorials for the many good free and open source assistive technology programs was extremely poor. This has been a long standing problem with many if not most free and open source programs of all kinds.
I think we should push the governments to support projects like NVDA by NV Access http://www.nvaccess.org/
If we could have a united treaty between all developed countries to support and fund projects like NVDA commercial screen readers would die a sudden death. Technology should be accessible to all and I firmly believe that technology does not discriminate but people do. I really hope that the right people are able to lobby this to the right stake holders.
Post a comment