A braille-using friend who is a computer programmer and RBD user once told me that he'd been temporarily stymied in his attempt to access some important information on the Internet because his screenreader stopped working. However, he was soon able to obtain the information by using computer braille to read the HTML source directly on his RBD. In fact, he not only read what he needed, he figured out by examining the HTML why the screenreader had stopped and reported the cause of the problem to the screenreader's developers.
If braille is to remain a useful tool, the specification of braille systems should be consistent with the practices of sophisticated braille users who access electronic information via their refreshable braille display (RBD) hardware. These practices include use of their RBD computer braille tables for direct access to the ASCII characters. Braille users have the option of choosing any computer braille table supported by their RBD. However, a braille user's braille transcribing system has the potential to impact their subsequent facility with their chosen computer braille table. Users of systems which, like the UEB, employ the same cells for certain letters and for the decimal digits and/or employ multi-cell symbols for standard characters such as plus and minus obviously have no option but to choose a computer braille table which is harder to learn because it is less similar to their transcribing system.
A "computer braille" table is not a transcribing code or braille system; it is simply a table presenting one of many possible one-for-one mappings between print characters and braille cells. In other words, a computer braille table is a two-way braille-to-print and print-to-braille transliteration. [Cf. Transcribing Codes versus Computer Braille.]
There are a number of different computer braille tables currently in use. These include U.S. eight(seven)-dot computer braille, EuroBraille* computer braille, six-dot ASCII Braille, and Grade 0 braille. Eight-dot computer braille systems can be preferable to six-dot ones since the ASCII character set is comprised of 94 characters. Eight-dot computer braille allows for single-cell symbols for all 94 characters by adding a seventh dot to 31 cells in order to represent the capital letters and five special characters. Six-dot computer braille can, of course, handle all of the ASCII characters by using two-cell symbols as necessary. [This latter method is similar to the approach used in BANA's Computer Braille Code (CBC).]
The use of RBDs has been growing and is expected to continue to grow as a result of new initiatives. Kevin Carey, chair of RNIB, said in January 2012 that "there are already currently as many as 34 technical ideas in outline or prototype format at universities worldwide, any of which might lead to the desired goal of a cheap Braille display...." The RNIB is leading a new project to further an "international plan for disability organisations and others to invest in producing a refreshable Braille device hugely cheaper than current systems on sale [and which plan] has obtained initial approval from the international DAISY consortium for information standards ...."
Carey tied the importance of RBDs to the importance of proper braille systems by stating, “If Braille is to survive into the 21st Century, it will have to re-invent itself as a mass medium, simpler, cheaper and easier to render... unless we face up to these challenges, Braille will die”.
Evidence that RBDs can support the use of braille as a mass medium is that many RBD users already move at will between braille and print. They write braille and print with equal ease via braille and standard keyboards; they read transcribed braille and transliterated print with equal ease on their braille displays. Many braille display users find it convenient to use a standard keyboard to compose text in print while proofing it in contracted braille on their braille display.
Given the usefulness of computer braille, all braille users would benefit from having a version of computer braille in their toolbox. The epigraph to this article gives an example of the usefulness of computer braille to solve an unexpected problem by allowing for direct examination of an electronic source file.
A braille user has the option to choose any computer braille table supported by their RBD. However, a braille user's braille transcribing system has the potential to impact their subsequent facility with their chosen computer braille table. Users of the Nemeth transcribing system have an advantage because of the considerable overlap between the Nemeth symbols and those of the standard computer braille tables used in the United States. Users of technical codes which, like the UEB, employ upper numbers and/or multi-cell symbols for standard characters such as plus and minus obviously have no option but to choose a computer braille which is harder to learn because it is less similar to their transcribing system.
It should go without saying that braille-using computer programmers typically read and write programs in computer braille. In fact BANA's CBC was developed to address this need. Here is how the late Tim Cranmer characterized the situation prior to the adoption of the CBC:
Most of the symbols a student would read in a braille book embossed on paper were entirely changed when presented on a braille computer terminal. Besides having different dot patterns in the paper versions of computer learning texts, many symbols might be made up of two or three characters, although the actual computer terminal symbols would be composed of only one character.
However, all braille users, whether interested in computer programming or not, should be able to use computer braille to read an electronic source file directly. The epigraph to this article gives one example of the value of this skill. The ability to directly read electronic source files reduces dependence on the developers of screenreaders and of transcribing software. This is significant because the relatively small market for these applications negatively impacts both quality and availability.
Another example of the valuable nature of this skill is that computer braille allows braille users to read and write mathematics in LATeX source or other plain text math entry systems. Many braille users have learned LATeX and use it to communicate with sighted teachers and peers. Unfortunately, although LATeX was designed to be fairly easy to write, it is a typesetting markup language and the source was not designed for convenience of direct reading. It is, however, preferable to nothing.
As far as acquiring the general skill of reading electronic source files, braille readers should find them conceptually familiar because of their similarity to braille documents in being linear, marked-up plain text and in their use of a limited (generally ASCII) character set. The main difference is that source files use markup for structural items as well as for character formatting whereas braille documents use direct whitespace formatting for structural items such as headings and paragraphs. Of course electronic source files are typically much more verbose than braille documents since source files aren't designed to be read efficiently by humans.
Even if they end up using a unified braille system, many sophisticated braille users are likely to want to use a computer braille table as well as the braille system since they support different functions. Braille users of a unified system like NUBS, which uses separate cells for digits and letters and one-cell symbols for common mathematical symbols, would thus have an advantage over braille users of a system like UEB, which uses the same cells for digits and certain letters and multi-cell symbols for common mathematical symbols. The advantage comes because the former users would be able to minimize the amount of new learning by choosing a computer braille table with considerable overlap with their braille system.
A DRAFT version of this article was first posted April 19, 2012. A second draft with minor corrections posted June 5, 2012. Please send corrects and feedback to the author: info at dotlessbraille dot org