Image: Torley Linden is a well-known Second Life resident with Asperger’s Syndrome. Here he’s speaking in Second Life at the Bounce for Autism event in 2008.

Last night I went to an event that combines several of my interests: gaming and virtual worlds with teaching and accessibility. The new Animation and Games Development Specialist Group of the British Computing Society was holding a talk in London with four speakers on the topic of virtual world and gaming for health professionals. Of particular interest were the following two talks:

Nigel Newbutt will present a brief overview of assistive technologies: virtual environments for people with autism. This will include an analysis of VEs that have been developed with the specific aim for users with autism. Within this talk a very brief overview of what autism is will be presented, and some of the advantages and issues that surround the development and use of VEs for users with autism.

Mick Donegan will present on the use of computer technology for people with ‘locked in syndrome’. This will include an overview of assistive technology for people with communication difficulties and how existing and new technologies can help people with physical disabilities.

Unfortunately, two of the four planned speakers were not there, including Mick Donegan. Emmanuel Pourreuix, one of my current H810 students attended. He recorded the session with the permission of the presenters, so a transcript might be available at a later date.

Nigel Newbutt’s presentation was more of a literature review in the area of virtual worlds and autistic spectrum disorders. The main tenet was that people would be immersed in virtual versions of real-life situations to help them become accustomed and comfortable with those situations. Newbutt is hoping to do studies that show social skills gained in virtual worlds can be generalized to real-world situations. That sounds really useful and he said that, while there had been some work in using virtual reality or virtual environments with autistic children, very little of it was obviously generalizable. Newbutt’s following up the hypothesis of Fusar-Poli et al (2008) that Second Life could be an environment that levels the playing field for those with autism by offering an environment for social skills rehearsal. If you are interested in virtual worlds, especially Second Life, as an environment for those with autism, Fusar-Poli looked to have several other relevant journal publications on work in this area.

I put assistive technology in quotation marks in the above paragraph, because using virtual worlds in this way does not fit my conceptions about what assistive technology is. What Newbutt was proposing was more a methodology for teaching skills than using technology to actively assist in real-time, which is how most assistive technology works. I am not belittling the idea. I think anything that allows us to bridge the communicative gap that exists is great. I am just questioning whether it should be identified as assistive technology. Perhaps that is open to debate and I should broaden my definition, although my definitions seems to fit in well with the literature.

The most interesting part of the presentation was the list of questions that could be investigated. In particular, how would students with autistic spectrum disorders represent themselves with avatars and why? Someone in the audience was wondering at what age people at different points on the autistic spectrum were able to differentiate between Second Life as a virtual environment and themselves. To me both questions hint at the “theory of mind” issue that characterizes much of the literature relating to autism. Basically, the theory of mind is the ability to recognize that you have beliefs, ideas, and emotions that are different from those of others (Wikipedia 2010). If you are interested in that, a good starting place is the Wikipedia article for an overview and then explore the references provided. How people perceive themselves and others is a really fascinating area, regardless of the people involved. I would love to see more work on that.

Nigel Newbutt (a very Second Life kind of name) will be conducting studies in Second Life as part of his Ph.D. work at University College Dublin under the direction of Goodman, Parsons, and Morie. One of the things Newbutt mentioned in his presentation was work done by Fabri and Moore (2005) in collaborative virtual environments by embedding a emotionally expressive 3D face avatars in conversations. With the exception of disgust, the Ekman universal six emotions could readily be represented and recognized by people with autism even with very crude graphical depictions of facial expressions. Newbutt’s thesis topic builds on this: “How can facial motion capture be used effectively in Virtual Environments to help improve communication in people with Autistic Spectrum Disorders?” SMARTlab, the group Newbutt belongs to, seems to be investigating other facets of virtual reality technologies and autistic spectrum disorders, so watch that space in the future.

References

Fabri, M. & Moore, D. (2005) ‘The Use of Emotionally Expressive Avatars in Collaborative Virtual Environments’, in Proceedings of the Joint Symposium on Virtual Social Agents at Social Intelligence and Interaction in Animals, Robots and Agents (AISB’05), Hatfield, United Kingdom, April 12-15, The Society for the Study of Artificial Intelligence and the Simulation of Behaviour. pp:88-94. Also available from: http://www.aisb.org.uk/publications/proceedings/aisb05/10_Virt_Final.pdf.

Fusar-Poli, P. et al. (2008) ‘Second Life Virtual World: A Heaven for Autistic People?’, Medical Hypotheses, 71 (6), pp:980-981. Also available from: http://dx.doi.org/10.1016/j.mehy.2008.07.024 (Accessed November 9, 2010).

Wikipedia. (2010) Theory of Mind, [online] web page, Wikipedia. Available from: http://en.wikipedia.org/wiki/Theory_of_mind (Accessed November 9, 2010).

This article on how to approach report or essay planning is based on a posting I made in one of my tutor group forums recently. I know that several of my students are new to the Open University and possibly new to postgraduate work. That probably applies to students in the other groups as well, so I thought I’d make a public version of my advice.

Most people don’t think logically enough to just sit down and write a really good piece of work off the top of their head. I know I don’t. Even when writing a blog post earlier today, I omitted things I’d intended to include because I didn’t have a plan and check things off as I went along. A plan helps you structure how you’re going to tell the story you want to tell and then tell it in a compelling, focussed way.

I’ve used the following method for postgraduate work in The Open University’s IET faculty and I use a variant for writing my other academic papers. I don’t guarantee it will work for you, but perhaps if you try it, you’ll get some ideas about how to adapt the practice for yourself to good effect. I use dedicated outliner software, but a word processor will do, even if it does not have an outliner mode. In a pinch, you can use a piece of paper and a pen, but it’s more difficult to move things around that way. Some people, by the way, use mind mapping or concept mapping tools to do this sort of planning. Post-It™ notes, like in the post’s header picture, will also work great, although most of your planning shouldn’t end up looking so complicated!

I start by making a list of topics the paper needs to cover. This is usually gathered from the assignment booklet section that lists what you need to do. I use those as headings and beneath each one I list as many things as I can think of that would fit into that section. I also list any relevant quotations or references that I happen to remember and would fit in that section.

The first assignment has a 1500 word limit. That is not a lot. No matter what the word limit is, it will never be enough to fit in everything I have listed so far in my outline. I next need to calculate how many words I can afford to use for each section. Not all of the 1500 words will be available to allocate to the areas. That’s because you will also need an introduction, where you usually describe your context and the overall focus of the work, and a concluding section. Keep aside some word count for those too.

Image by Michelle A. Hoyle under a Attribution-ShareAlike 3.0 License
Image: Example of Michelle’s finished outline for an assignment. Note the checkmarks indicating she used that point in her final writing.

Once you’ve decided how much you can afford for each area, go through and pick out the strongest points and references you have. Try to do this with an eye to creating an overall harmonized symphony. Because you’re working with bullet points here, the big picture should be relatively visible to most people still. What we want to do is tell a single story and every point we pick out should help contribute to making that story stronger, so we want things that are complementary and go together well. It’s even better if the same general point can be somehow applied to multiple areas. Depending on the word count, the number of points will usually be fewer than 3 points—maybe only one or two!

Now that you’ve picked out your major points, expand your outline to include relevant references that go with each point or would support the points. Then take a look at the everything within a given area now. Should your major points be rearranged? How will it flow into the next section of your report? The best reports flow nicely from one area to the next without an abrupt break, even if you’re using headings to separate different sections. That’s again because we’re trying to tell a coherent story. If yours is a bit abrupt, don’t worry; it’s just something to aspire to.

Sit down and write a section using your plan as a guide. It’s likely that you’ll come up with something else that fits into it while you’re writing it. That can be OK, but maybe consider how it will alter your plan and keep a ruthless eye on your word count! The idea isn’t to fit as many different ideas into a section as possible but to include enough and in enough detail to tell your story. Check of the points you’ve used out of your outline as you go along to ensure you’ve covered what you meant to cover. Repeat until done with the other sections.

I’ve included a screenshot of an excerpt from a plan I did. This excerpt is a well-advanced plan. In fact, I’ve finished the essay and sent it off. As a result, you can see tick marks where I included a point in my work and some of the ones that I didn’t, because I didn’t have the space. I’ve tried to include full references where possible and quote things when using someone else’s words to help avoid accidental plagiarism. Finally, just because you have a plan, doesn’t mean your end result will end up looking exactly like it. Chances are good your writing will be better for the process though.

Do you use a different process for planning and writing? Share what it is and how well it has worked for you. I’m sure others will be interested too.

It’s hard to believe it’s already week 5 in H810. I commented at the end of week 2 that I thought my group was off to a quiet start, with people working on H800 ECAs, getting married, and being on vacation. My impression was that, since then, things have been even quieter, so I spent part of Sunday afternoon doing looking to see where people were and doing some analysis.

The result of my analysis is the above colour-coded image of people’s participation. The seven green blocks represent people from our tutor group who have posted in Week 3 forum activities. The eight red blocks represent people who haven’t posted in any of the week 3 activities. As it turns out, the image looks the same for week 4 participation.  Week 5 had no postings at the time of my analysis.  So, out of a group of 15, we just under half who who are more or less up to date and a slightly larger group who look to be about two weeks behind based on their forum postings. That’s a lot of behind people and explains why it feels so quiet.

If you’re one of the red blocks, you need to be concerned. Two weeks equates to one third of the block and the first TMA is due in 10 days. That’s not very much time to get caught up and it means you might be missing some major topics to prepare for your first TMA. Week 3, for example, features a really important topic in the course: legislation. While it would be nice to believe that all institutions are acting from a social model naturally, the sad fact is that legislation is probably what compels many, regardless of the country. One of the main threads of TMA 1 is the influence of legislation, so this is a key topic to have a good understanding of. Another key topic for the TMA that arises in week 3 is the roles of the stakeholders in providing and acquiring support, with week 4 providing some background information on how assistive technologies can be used to address challenges that we identified in week 2. Week 5 provides a more in-depth look at adjustments and reasonable accommodations, as well as a variety of online learning tools and how they work (or don’t).

Once you get to grips with the general details, you still need to be able to apply the ideas to your own context. Your first TMA, most likely written in the first person, needs to take what you have learned and reflect on how it is relevant in your own institution and in your own role. Most people will need some time to reflect on that and the blogging and forum posting activities were designed to help you start integrating the pieces together. If you’re behind, you probably won’t have time to write reflective pieces in preparation, but try to think about how things would apply within your own role and institution as you go through the material. That may help in preparing.

Some TMA Tips

Simon, one of the other associate lecturers, posted a useful list of tips for the first TMA based on the previous presentations. I’d like to highlight a few he has mentioned plus include some additional ones:

• You need to be using references. While the report is in your own context, you need to use references for your background information and, if possible, for contrasting or reflecting on what is actually happening in your own context.
• References should be using name-date style. I’m not picky that it conforms exactly to the OU’s Harvard Style but it should be a Harvard name-date style and you should be consistent within your use of it. What I mean by consistent is that one journal article reference shouldn’t be italicized while another one is just surrounded by quotation marks. The OU Library has a Harvard Style help guide (PDF file) available from their cite references page.
• 30% of the marks are for your choice of online resources. So I’d expect to see Seale (and note the spelling of her name) and some of the many online resources used in the course. For the very best marks, I’d expect to see evidence that you have done some research on your own and turned up relevant material, because this is a postgraduate course.
• 10% of the marks are for making effective contributions in the forums. If you haven’t already contributed to discussions up to and including week 4, it’s likely too late. It’s not “effective” if it’s posted a few days before the TMA cut-off date, because you and others are denied the opportunity to engage in dialogue in a timely fashion. However, the process of writing is useful, so please don’t let that stop you! I’m just trying to point out that this isn’t just a counting exercise. Also, note that referencing discussions is useful, but it’s not as valuable as contributing yourself.
• Although you will likely be writing in the first person, because you will be reflecting on your own context, remember that this is an academic piece of work, so the tone of the writing still needs to be scholarly. That means a more formal way of writing: no slang, few contractions (if any), proper grammar, and proper spelling. As a Canadian, I’m fairly flexible on “proper spelling”, as Canadian spelling is a combination of American and British. I do expect you to be consistent, so don’t use colour with “u” in one sentence and then color without a “u” later on. Do proofread and do use your spellchecker.

Extensions and Advice

Simon concludes his list by pointing out this is an assignment and not a test. I agree. While I won’t read draft assignments, I am happy to consider assignment plans and provide other kinds of advice. For example, you’re wondering abut your forum contributions, or you want feedback on a particular blog post, or perhaps you have questions about the finer details of referencing. Don’t be afraid to ask me either via e-mail or in the TMA 1 forum thread I’ve created. You can also have a peek at the postings in the other tutor groups.

I’m happy to consider extensions, but you need to secure permission from me before the assignment’s due date. Due to the tight timing of this course (and general OU regulations about extensions), I may not grant an extension or perhaps may not be able to give you as much time as you would like because it will have a knock-on effect on the next part of the course.

Getting in Touch

Just a reminder that you can get in touch with me in a variety of ways. I check my e-mail frequently and you should shortly receive an e-mail with my new OU contact details. You can telephone me most weekdays at home during the day (except Fridays when I’m at the University of Sussex). If possible, please avoid Sunday, Tuesday, and Thursday evenings after 19:30. You can post in our online forums or I’m on Twitter as OUEin.

Being a Web 2.0 kind of person, I’m always keen to try out new things and use them in an educational context. I’ve been aware of of the Paper.li service for awhile now and I’ve seen it used for other courses I’m involved in around the world. I’ve therefore created an H810 paper for us at http://paper.li/tag/H810 (and case is important. That’s a capital aitch in H810).

Here’s how it works. Every 24 hours, Paper.li scans Twitter and finds any uses of the #H810 hashtag. If there is any sort of link associated with a tweet, it makes those into “stories” for the paper. A story is a brief summary excerpt from the link associated with the Tweet and you can then go off and read the rest if it’s interesting to you.

If you’re not using Twitter, you can still use this to browse related and interesting things. If you are using Twitter, just remember to tweet about interesting articles or sites and include the #H810 hashtag. You do not need to sign up for either Twitter or Paper.li to enjoy this. Do note that Paper.li is free, so it is ad-supported. I can’t take any responsibility for the ad content or the fact that there are ads.

This is a great way to share your own blog postings too! There is, however, a caveat with that. If you’re using the OU’s built-in blogging service and you’ve set your blog to be private or only visible to people at the OU or in the course, the “paper” won’t be able to pick up the story, because it can’t access it. If your blog is public, self-hosted, or hosted at somewhere like WordPress, it’s probably fine.

I did a quick check for accessibility using VoiceOver on the Mac. It wasn’t too bad, although there’s a pile of stuff at the top you can’t skip over. It looked like using VoiceOver’s “auto webspots” feature (VO-U to bring up the web rotor and then VO-right VO-right) would be a decent way to quickly access content.

There is a “paper” for today, but it’s very thin. I should have pre-populated it with some stuff. Nevertheless, do bookmark it and keep an eye on it every day. I may occassionally tweet to remind you that a new one is out, but probably not, as I find that annoying.

Enjoy!

We’re just finishing off the first two weeks of this year’s H810 presentation.   Our group, including me, has 15 people, but it’s been pretty quiet.  Well, that’s not completely true.  Our introductory thread, taking place in the Canadian Rockies, was fairly active and the “context” thread, where people described their context, saw a fair bit of traffic.  However, the topics related to the course content itself, especially for Week 2, have been slow.

If you’ve been looking at other forums in the course and wondering why ours is so quiet in comparison, I think it’s because a number of people are actually essentially away for the first two weeks of the course.  Some are on vacation.  Some are travelling.  Some are working on their ECAs for other courses that end this month.

If you’re one of the people who has been busy and especially if you’re new to the OU, I think it’s important to get a good handle on some of the topics for the first two weeks, as they’ll form the basis of your thoughts when it comes to TMA 1.

In particular, try to participate in the topics around:

• language
• accessibility definition
• challenges for students

In the wiki activity around defining accessibility so far, some key themes have emerged, including the idea of a “level playing field” and the responsibility for accessibility lies with all of us. Is what with all of us as a society (as in the social model) or does it mean all the stakeholders involved in the interaction, including the student themselves? I think this is a very important question to consider. Models, discussed in week 1, don’t just describe how students with disabilities or impairments are treated. Models can also describe how the students view themselves: equal participant, victim, dependent, etc.

I’m a bit surprised that so far nobody’s started a discussion on whether or not dyslexia is a disability or not, although I did see a comment about ADHD and ASD not being considered disabilities mentioned.  I’m quite interested in people’s thoughts on that myself. How do we define a disability? Is a disability different than an impairment? How far should we go? Is, for example, speaking a different language an accessibility issue or an impairment? If you have time, perhaps you’d care to blog about that.

In closing, if you are new, don’t let yourself get too far behind.  Each week is set at about 15 hours, so getting behind adds up quick!  Do ask me for some advice if you need help catching up.

Now that people should be coming back and we’re heading into Week 3, I hope to see more people participating.  It’s a friendly world out here.  Try us.  You’ll see!

What are your thoughts on the first two weeks?

One of the early activities in H810: Accessible Online Learning: Supporting Disabled Students asks students to reflect on their role and context in education and how they relate to accessibility and online learning. In the 2008 pilot, I wrote a little about my own context in My Role and Context in Education. I’m still an associate lecturer and course chair at the Open University, but I’ve since left the world of web development to focus more on my Ph.D.

In my previous article, I mentioned that I’m on the front lines of interacting with students as a result of my role as a course chair and associate lecturer.  I recounted a story of how course design decisions have led to some interesting problems when I have students and I’m unaware of impairments, such as having many, many images to illustrate what a finished student website should look like in terms of functionality and layout. I have another story along similar lines.  This one too involves a blind student using assistive technology to read the multiple choice questions in the PDF assignment document.  The question and answer options look something like this:

Based on the following statements, classify the items shown in bold as being best represented  as objects, properties, or methods.  Note that the statements refer to the same items, so, for example, ‘Doug’ in the first statement is the same ‘Doug’ as in the fourth and fifth statements, and so on.

 

1. Doug has a bike he likes to ride. 2. Freddy is Nick’s large green parrot. 3. Freddy sits on Nick’s high window. 4. The window has a height of 150 cm. 5. Doug, Freddy and Nick like to sing in the pretty park.

Choose one of the following to represent the correct classification:

A object, object, property, property, object, property, method B object, method, property, object, method, property, property C object, object, object, property, method, method, property D object, method, property, object, method, property, object E object, method, object, property, property, method, object F object, object, property, property, object, object, method

Do you see—so to speak—the problem?  A screen reader would read “Doug has a bike he likes to ride.” exactly as “Doug has a bike he likes to ride.”  The screen reader version gave no indication at all of which words were bolded, so the student could not answer the question without assistance.  I solved this by sending an e-mail to the student with the specific words labelled in an audio-friendly way.  In future presentations, I changed the question style.

On the other side of the desk, in my role as a Ph.D. student, I recently filled out an application for a disabled student allowance (DSA) to support me in my studies.  I found it a bit humiliating.  I am sure it’s not intended to be a humiliating experience, but when you find yourself having to justify why you need extra help, it does something to you.

I had a similar experience in dealing with my department.  For the past two years, I have been attempting to convince my department to allow its students to submit a collection of papers as their thesis instead of a traditional thesis.  There are a few departments on campus that already allow this.  I felt that this approach would be a lot easier for me to do with my disability, because the scope of writing a paper is more self-contained and manageable.  I thought that everyone should be afforded this opportunity, so I was canvassing for everyone, not as a personal accommodation.  When I discussed it with the responsible person, they came across very negatively when I mentioned that I thought I would be able to get it as an accommodation. They wanted me to explain why I felt that was the case.  In other words, I was being asked to justify my need for an accommodation.

Even though some of these experiences are negative, there is a positive outcome.  They all serve to better inform me of the obstacles, both intentional and unintentional, that affect students with disabilities and impairments in higher education.  This, in turn, makes me a better mentor, educator, and course designer in my turn.  How will your experiences and interactions change you?

Yesterday marked the official kick-off date for The Open University’s postgraduate course on accessibility and e-learning:H810: Accessible Online Learning: Supporting Disabled Students. I saw in the forums and on Twitter that some students haven’t been allocated a tutor yet. I heard late last week that students were still registering and I know I was asked if I’d consider taking more students. So I suspect if you weren’t allocated a tutor by Saturday’s start date, then it will be sorted out later this week. In the meantime, there is plenty to get started with. You can blog your thoughts on the language of disabilities as well as your ideas about what exactly accessibility is and then introduce yourself later.

My own group currently consists of 12 students and I have to say they’re all over the place. While most had some connection at least initially with the United Kingdom, at least one is in Colombia, another is working in the United Arab Emirates, and another one is in France. It’s fantastic, because how accessibility and disabilities are viewed varies so drastically from country to country. We’re lucky to have such a wide variety of viewpoints available to us.

A big part of this course is context. Context is not just where you live, but also your experience and activities too. In your assignments you’ll be asked to write about accessibility and e-learning with respect to your own context. That will be a lot easier if your job involves working with students or designing courses, but there are other possibilities too. Be sure to check the course guidance on context in Section 2.6 If You’re Not a Practitioner in the Course Guide early on.

I’ve just been dusting off this blog as I’ve officially rejoined The Open University’s presentation team for H810: Accessible Online Learning: Supporting Disabled Students, a postgraduate course on accessible e-learning that covers design philosophies, technologies, and legislation.  The course officially starts September 4th and it’s still possible to sign up if you’re interested.

I started this blog on the pilot presentation of the course in September 2008.  I worked through most of the course material along with the students and blogged about relevant bits and pieces and my own thoughts.  In addition, I used this blog to aggregate blog postings from students in my group and help disseminate course news and other materials of interest around accessibility, disabilities, and online learning.  I’ll likely be reprising its role for all of those things and anything new I happen to think of.

If you’re a student in H810 or a student, what would you want to see here?

I just realized that, in the holiday madness, I finished marking TMA 2 and made the nice chart above, but I did not remember to post the chart or comment generally on TMA 2′s results. Here, belatedly, are some comments to remedy that.

The above chart shows the distribution of marks. The actual range was 44% to 87%—very similar to the range in TMA #1. The chunking, however, was more distinct, with the majority of people scoring in the 65% to 69% range. Similar to TMA #1, the big issue for most people was being able to justify what they had done to specific issues and research covered in the course. For the ECA it will be very important to write from your own context but you must specifically relate to content in the course. For example, “I provided both a PDF and an RTF version of my learning resource because, according to Taylor (2008), many assistive technology devices cannot read Word documents but can read RTFs.” This phrasing accomplishes three things. It explains what you did; it explains why you did it; and it supports your action and justification with a course-based reference. Just stating that you did something, while that something may demonstrate good practice, does not illustrate that you understand why it is good practice. Likewise, being very vague about what specifically you did may not be helpful. Adding alternative text descriptions to graphics may be good practice, but it only enhances accessibility when the descriptions are both appropriately added and appropriate in content. The classic example of that is people using blank “spacer” images to position content on web pages and blindly adding text descriptions of “spacer image”. Imagine how tedious, annoying, and useless it is to hear that over and over again via your screen reading software.

On the whole, though, everyone is doing satisfactory and is well on the way to passing the course. I did get to see some very neat resources. I was particularly impressed with Dwayne’s enthusiastic attempt at converting a PowerPoint presentation into a Flash-based application for use on a Moodle site. That was quite brave, even if the result wasn’t completely what would be desired. I’m sure he and everyone else learned a lot in the process of creating their learning resources. I know that I certainly had a very interesting experience in trying out the various resources using the built-in screen reading software for my computer.

That leaves only the end of course assessment (ECA), due January 23rd. I will be posting some separate advice and comments about that later.

You probably have signed up at a web site where you were presented with a graphic showing you some combination of letters, numbers, or words in a graphic to prove that you are a real human being and not some kind of spam bot. The “following/finding” image above is an example of a word-based version of that task. These images are known as CAPTCHAs (Completely Automated Public Turing Test to tell Computers and Humans Apart). My earliest recollection of seeing them in wide use was on blog sites with open commenting. Automated programs would submit “comments” consisting of links to pornographic web sites or pharmacy sites. For popular bloggers, even if they had a system to moderate comments before making the comments publicly visible, the overhead in managing their blog could quickly become unreasonable. For similar reasons, sites like Yahoo Directory, Google Mail, and HotMail were also fairly quick to adopt CAPTCHAs.

For most people, the main issue about CAPTCHAs was whether they were effective or not. As with anti-virus efforts, it is an ongoing fight between the guys in the white hats to protect their systems against the guys in the black hats who want to pervert the protected systems to their own ends. From an accessibility point of view, though, that issue was minor potatoes. Even users with perfect vision often have trouble with CAPTCHAs because of the level of distortion involved in obscuring the letters or words. The solution to that was to add a “refresh” or “recycle” button to the CAPTCHA so it would give you a new CAPTCHA.

However, if you were blind or had poor vision, it was pretty much well impossible to work past the graphic. What the initial CAPTCHA developers had failed to consider was how users relying on assistive technology to surf the web were going to be able to use a CAPTCHA graphic.   Why was that? Consider the usual way of making graphical content accessible: add a description to the image. If the task for our sample CAPTCHA above is to type out the words in the picture, putting “CAPTCHA image with the words ‘following’ and ‘finding’” as the description is going to help those not using images, yes, but it is also one hundred percent accessible to automated programs. While we obviously like to endorse accessibility for all, there is a tension between accessibility and security;it is completely undesirable for automated programs to be able to circumnavigate a security system so easily.

One solution to the accessibility issue was to add an audio component to the CAPTCHA. The “overlooks/inquiry” image shows a reCAPTCHA example that incorporates both the refresh button (the recycle-like symbol at the top of the column of icons), a help icons (at the bottom of the icon column), and the audio CAPTCHA icon (middle of the icon column). When you click the audio icon, the large word area of the CAPTCHA is replaced with a mini audio player and you are instructed to type what you hear. The audio in most examples I have tried is not the words in the graphical version. The audio quality is usually poor and may, on purpose, be distorted with additional people speaking or background noise in order to make it difficult for automated speech recognition programs to function. I often have trouble with the audio because of my own neurological hearing problems and the interference caused by background noise and lack of context. Try it yourself on a few examples at the reCAPTCHA site.

You might be thinking that the audio reCAPTCHA is a good compromise at trying to ensure accessibility for human beings while denying it to automated programs. Unfortunately, recent research studies have revealed that all of the common audio CAPTCHAs in use were vulnerable to automated speech processing techniques, with anywhere from roughly 50 percent to 70 percent accuracy. This excerpt from the December 8, 2008 Ars Technica article Computer scientists find audio CAPTCHAs easy to crack summarizes the important results:

The work involved gathering 1,000 audio CAPTCHAs from Google, Digg, and the reCAPTCHA service. 900 of these were used as a training set and the remaining 100 were set aside to test the system when done. The software first did a rough audio analysis, dividing each item into equal-sized chunks, each sufficiently long to fit any spoken character. Those segments with the highest energy peaks, which are considered most likely to contain actual letters, were set aside for analysis.

The authors tested a number of methods used to extract features from recordings of speech (for the curious, these are mel-frequency cepstral coefficients and two forms each of perceptual linear prediction and relative spectral transform-PLP). These features were then subjected to analysis using machine learning programs, which were trained on the identification of individual characters. Three methods—AdaBoost, support vector machines (SVM), and k-nearest neighbor (k-NN)—were trained using the 900 audio CAPTCHAs that had been processed manually. The result of this pairing of processing and analysis methods was a total of 15 different attempts at cracking each of the 100 test audio CAPTCHAs.

Google’s audio CAPTCHAs consist of a series of the digits 0 through 9 recited over background noise of speech played backwards. That was nowhere close to enough to consistently fool the researchers’ software; the SVM technique got the CAPTCHA right about two-thirds of the time, and AdaBoost wasn’t far behind (k-NN performed badly in this test). Digg uses both digits and letters, but plays them over a less complex background that sounds like flowing water. AdaBoost failed this test entirely, but SVM was able to clear 70 percent accuracy with several of the processing techniques; k-NN trailed it by a significant margin.

reCAPTCHA’s own audio version was similar to Google’s but used different speakers for different digits. This proved to be a significant barrier to the learning algorithms, which, at best, got it right a bit less than half the time (again, SVM was the star). As the authors point out, however, getting it right half the time would be more than worth the effort for spammers that may have hundreds or thousands of computers at their disposal. Some sites also allow the answer to be off by one digit, which would significantly increase the success rate.

[From Computer scientists find audio CAPTCHAs easy to crack]

We again have that tension between accessibility for people but inaccessibility for automated programs. A 50 percent success rate is not low enough to deter the bad guys. What can be done? The researchers, however, did conclude that “more of just about everything is better: more speakers, more characters, more distortion, and longer strings of tokens all seem to make a difference. As a result, they have expanded their own service to include all numbers from 0 to 99.” Time will tell how that pans out. I still wish we did not have to rely on different speakers, distortion, and entire sentences for audio CAPTCHAs as that too poses its own accessibility issues for those with physical or neurological hearing problems.

Perhaps there is mileage in some of the lesser-used systems that ask people to do simple mathematics or ask common-sense questions like “What colour is grass?” I suspect those too will be quite vulnerable to automated systems as the number of questions will be limited. Unsatisfactorily, we may have to settle with the situation as it currently stands until someone cleverer than me has a bright idea. If you had to solve the problem of making CAPTCHA technology accessible but secure, how would you do it? Or is there a better way to separate the people from the programs?

Further Reading:

• What Is reCAPTCHA? (http://recaptcha.net/learnmore.html)
• CAPTCHA – Wikipedia (http://en.wikipedia.org/wiki/Captcha)

I’m afraid I’m only just starting the TMA 2 marking as my laptop had a horrible fan death a few days after I picked up everyone’s TMA 2. I had to back everything up and send the laptop away, which sucked up a week. The good news is that I am starting today and I hope <crosses fingers> to do at least one per day, returning all submitted on time by Saturday. I’d already hoped to be done, but these things happen.

In the meantime, I’ve noticed it’s been pretty quiet. Threads for Weeks 13 and 14 were created previously, so why not stop in and give those a whirl while you’re waiting for TMA 2 to come back? I’ll probably work ahead a little and create Week 15′s before TMA 2 is submitted.

Presenter: Sarah Lewthwaite (University of Nottingham)

Title: Disability 2.0: Facebook, the Academy and Student (dis)Connections.

Abstract: For many young people, online social networks such as Facebook are an essential part of their student experience. Other social web-based services like Wikipedia and YouTube are also an important facet of everyday student life. New technologies have always been scrutinized for their capacity to support education and, as these social technologies become more pervasive, universities are increasingly seeking to appropriate them for teaching and learning.

However, the educational impact of applying these Web 2.0 technologies for all users is unclear.

The experiences of disabled students crystallize many of the issues raised by the movement of the academy into the digital domain, disputing the notion of social networks as universally popular, transparent and inclusive. This presentation is based upon ongoing qualitative PhD research. Discussion will focus on data collected during 14 interviews with disabled students at different stages in their University studies. Interviews utilise screen capture, participatory and accessible methods to explore how the societal elements of disability transpire and transform online.

The seminar will be in the Interact Lab (Arundel 223), starting at 13:30 and lasting for an hour, with tea/coffee & cake afterwards. All are welcome to attend. More information can be found at http://www.informatics.sussex.ac.uk/events/HCTSeminars/. I shall be there and I’ve heard a rumour that Chris Douce will also be attending.