1 February 2006
In the first of a series of articles on the accessibility of art on the internet, Joe McConnell looks at the wide gap that separates self-declared accessible websites from websites which are genuinely trying to offer disabled people a comparable experience to that of the non-disabled.
After a trawl through the internet art world, many disabled people soon end up gaping across the so-called digital divide at a beautiful promised land from which they are still excluded. In terms of accessibility, the sites offering visual static or time-based art fall into three broad categories. Firstly, there are the sites that are not accessible either through incompetence or through not giving a flying fig. Then there are sites that appear to comply with recognised accessibility standards. Finally there is the tiny cluster of sites that genuinely provide a satisfying experience to disabled visitors.
The formal standards used to measure accessibility are set by the Web Accessibility Initiative WAI of the World Wide Web Consortium W3C. UK legislation uses these standards as a benchmark. There are three levels to these standards. Passing the first level indicates that certain features have been included in the construction of the site that help to improve access. Compliance at all three levels indicates that more rigorous preparation has been undertaken. Unfortunately, compliance is self-testing using validation software . The sites which believe themselves to conform at all three levels, slap an AAA banner on their home page. For full details of these standards see w3.org/WAI/about.html.
In a helpful article on the Disability Discrimination Act and the web, Trenton Moss points out that there is a lack of clarity from the Disability Rights Commission as to legal requirements governing accessibility. Possibly because of this vagueness, there appears to be a general perception that the section of the DDA referring to websites came into force recently, which explains the current scramble of many sites to seek AAA-status. Moss reminds us that the DDA has required service providers to ensure equality of access to their websites since 1999 when Section III of the act came into force. Although there is yet to be a test case in the UK, large organisations have been successfully prosecuted in Australia and the US using similar legislation to the DDA.
A recent investigation conducted by the DRC into the accessibility of 1000 public service websites revealed that a shocking 80% were fraught with access shortcomings. However, at the end of the day, a website proprietor facing prosecution will probably only have to prove compliance with Level 1 of the W3C Accessibility Guidelines to demonstrate that the sad old reasonable adjustments have been made.
The guidelines themselves are a set of extremely useful rules towards achieving a certain level of accessibility. Any cynicism in this article does not reflect disrespect for these standards. In fact achieving compliance at all three levels is extremely hard work. But there is no legal requirement whatsoever to make a website into a positive experience for disabled visitors. And standards-compliance is only one small step towards that goal.
Bearing the legal dimension in mind, I decided to mosey around the UK internet art world in search of good practice. I limited my focus, on this occasion, to access for visually impaired people. This is not out of any belief that this was the only group that mattered, but rather from the conviction that attention to this area, or the lack of it, would probably be indicative of the general will of a particular project to be accessible. I began with the national collections, feeling that these at least would have better resources and awareness of legislation than private individuals.
The National Portrait Gallery uses the BBC Education's Text to Speech Internet Enhancer (Betsie) service to dynamically render a text-only version of each page on the site. Blind visitors are bombarded with oodles of information about the gallery including how to buy prints. But because there is not the slightest element of visual description attached to the images which can be accessed through the database, which gives a miserable little reproduction of each work, all the screen reader will send back to its user are sterile details about the piece without giving any idea of what it actually looks like. Here, the text only facility gives a false illusion of interest in making the site accessible. The site actually fails on all three levels of the W3C guidelines.
Making a Start
The home page of the Tate Gallery also fails all three levels of the same test. On 16th January it had 23 instances of not providing alternative text for images. This is all the more surprising in view of the excellent work that the Tate undertook through the I-Map project which showed great innovation in facilitating the access of blind people to the Picasso-Matisse exhibition of 2002. Admittedly, the 23 instances did not involve reproductions of art works, but referred to images used to structure the page. However, this still indicates insufficient attention to the requirements of people using screen reading software. These elements should still have an alt tag albeit filled with one single space; otherwise, the file name of the image will be repeated every time the screenreader encounters it. The Online Collection, which is being compiled by the Tate's Insight Project does have descriptive elements included in some of the texts attached to individual works. But so far, this provision is rather hit and miss and there seems to be no consistent approach to the provision of visual description.
The National Gallery manages to pass Level 1 of the accessibility guidelines. The site has recently begun to include clips from its audio tour on its website. But the clips are more a discussion of each painting. Admittedly some detail is given but a blind visitor is certainly not being offered anything comparable to the experience of a sighted one. Some of the collections have a certain level of description - again these tend to be more discursive than descriptive. But at least they are making a start. And this is surely where the path towards true accessibility lies.
Days of trawling later, I had still to find an example of exemplary practice anywhere among the art collections in the public domain.
Inadequacy of Standards
Despite the proliferation of AAA banners and similar claims, it soon becomes clear that there is a tiny minority of art-based sites, which reflect a true desire to be accessible and many others which go no further than seeking compliance with standards from fear of prosecution. Surely at this stage compliance to standards should be a bottom line - a starting point?
The standards themselves are truly useful guidelines but totally inadequate as measures of quality. For example, the standards say you must provide alternative text for images. So in order to comply, web designers ensure that some text is attached to each image on the site. However, validation software cannot measure the quality of the information provided in an alt tag. It does not oblige page authors to provide something that gives users an understanding of what a given image looks like. A tag with a name, such as the title of the picture which is probably already included in the caption beneath it does not tell a blind user anything about the content of the image. However, if you include a fuller visual description, each time a blind visitor enters the page they would hear the description. It must be pointed out here that the guidelines have automatic checkpoints which can be detected by validation software (e.g. the presence or absence of a piece of coding on a page.) There is also a set of manual checkpoints which are a series of questions on aspects that the validating software cannot answer. The first one of these asks whether
an image conveys important information beyond what is in its alternative text (1.1) and recommends the inclusion of a longer description. It is clear that many web developers claim level 1 compliance after satisfying the automatic checks, but it is rare that you find that the manual checkpoints have been followed though.
The Inadequacy of Alt
It is generally acknowledged that the alt tag is
inadequate for images that are too complex to be described in a short phrase*. Reproductions of visual art require richer and more expressive description. For this HTML offers the longdesc tag. This allows the page author to virtually embed an extended piece of descriptive text in every image. And here lies one of the devilish shortcomings of the current state of the internet: although the longdesc element has been available since 1995, hardly any of the state-of-the-art browsers or screenreaders yet supports the feature. To overcome this, authors have for some time been using d-links, which is simply the adjacent insertion of the letter d linked to the long description text. This is not always easy to achieve. For example, if you have a row of thumbnails on a page, where do you actually place each related d-link? However, these implementational problems do not exempt authors from providing a visual description while waiting for the day that the browser developers finally catch up.
The validation tools approve the page if alt tags are provided but give no heed as to whether there is true visual description on offer. This is a clear example of the gap between compliance to standards and true accessibility.
* Adam Alonzo: A picture is worth 300 words: writing visual descriptions for an art museum web site. csun.edu
Better Practice and Innovation
As with audio-description, the best visual descriptions of images are based on a true aesthetic engagement with the material. This is not necessarily an easy ride. Finding the right person to do the job and giving them the time to do it is only realisable if included in the production schedule and budget of a website.
Good Practice - an example
When it came to developing her own website, visual artist Sally Booth found that
off the peg solutions were not accessible enough. The site, which will be launched shortly, has prioritised several different layers of access. Each image in her gallery is accompanied with embedded media players through which visitors can access both a visual description and a BSL description. Sally is providing the BSL version as her descriptions (which also appear as text) are also contextual and she is therefore ensuring that no one misses out.
The work so far is impressive. But, in terms of the audio element, we are not talking about prohibitively expensive new innovations here. The artist and designer are using technologies which are accessible to everyone. The big difference is that these elements have been factored into both the brief and the budget.
The technical provision of an audio description facility is not costly. There is very cheap downloadable software which allows you to use your computer microphone to record MP3s which can be uploaded to your website. The provision of BSL video clips is obviously more expensive.
Sally wanted to prioritise access from the offset and refused the approach of getting something out there and then thinking about access issues afterwards. This attitude throws a sobering light on the emerging culture of doing everything on the cheap which is one of the big selling points of many of the services - such as web design - on offer in the fly-by-night world of the internet. It is clear that the public collections have not taken on board the logistics of putting accessibility in place.
Coming across the prototype of Sally's site was like arriving at an oasis in a parched desert. Days of searching the net for inclusive design innovations involving audio description yielded very little. I was not expecting to find abundance but was shocked to find so damn little.
Accessibility - a creative opportunity?
Marcus Weissen, Disability Adviser at the Museums, Libraries and Archives Council, points out that accessibility should be seen as a creative opportunity. Access does not compromise artistic vision and it can enhance it. As is often the case with facilities for people with disabilities, visual descriptions can also benefit the general public. If artists themselves learn how to visually describe their own work a valuable new layer of curation is added.
So the story so far in terms of the accessibility of art on the net for visually impaired people seems to indicate that there is a morass of bad practice and that examples of good practice are very thin on the ground.
Many would agree that the vision of artists themselves should be unrestricted in a true democracy. For example, if an artist is producing pitch black canvasses or moving images in order to depict a dark night of the soul, then it would not seem appropriate to point out that this particular vision is inaccessible. However, more and more visual artists are independently putting their work about on the web. In doing so, they become curators and publishers of their own work. In this capacity they should consider the added value, in terms of usability and also out of respect for human rights, of making their work accessible, even if it is highly unlikely that they, as private individuals, will ever be prosecuted.