Jon Pretty
In 2001, as a university undergraduate, a friend and I were lucky enough to get the opportunity to meet the physicist Stephen Hawking at a small private reception after he gave a public lecture in Cambridge.
Whilst the lecture was itself deeply interesting, two things struck me more than anything else from that evening: the first was the warmth of Prof Hawking, who went from that lecture theatre having spoken to an audience awed by his explanations of black holes and branes in eleven-dimensional spacetime, to join us in a quiet side-room where he talked to my friend, also a wheelchair-user, about the availability of ramps and general accessibility in different buildings in the university; some of the most mundane practicalities of his daily life.
The other thing I observed was the ergonomics of Prof Hawking’s computer, as I stood alongside him and watched him slowly compose sentences, before completing them, and having them spoken. Presented with a monochrome screen with alternating selections of words, he had no greater physical capability than to synchronize several clicks of the computer’s single button to select one word at a time and append it to a sentence he gradually built up. It felt frustratingly slow for the rest of us, all too accustomed to the ease of speech, but for him it was an enabler through which he was able to make huge contributions to scientific research, and through which the rest of us were able to see through a small window into his great—and now sadly lost—intellect.
It is a reminder how easy it is for many of us to take our experience of programming for granted: we can quickly glance over source code or compiler errors on our screens to spot bugs or understand functionality. For blind programmers, the screen is usually replaced by a text-to-speech system, which linearly renders source code and error messages as spoken text.
Unfortunately, for a language like Scala, that audible representation is presented very verbosely, and can be painful to use. An engine such as “espeak” may render a definition like,
def foo[A: Bar](s: String): Int = ...
as,
def space foo space open square bracket capital a colon space bar close square
bracket open bracket s colon space string close bracket colon space int equals
whereas that same information could be conveyed more expediently as,
def foo
type a context bar
from s string to int
During the Scala Center Advisory Board last week, a proposal authored and strongly advocated by Sam Halliday, with help from Rui Batista, was accepted as a recommendation by the board, with a unanimously positive vote from all the members, including 47 Degrees, IBM, Goldman Sachs, Lightbend, Morgan Stanley, SAP, Twitter and Verizon.
The Scala Center will work towards making Scala more accessible. This effort has already received some open-source contributions, and the Scala Center will work hard to ensure that these contributions are seen through to completion, as we experiment with effective ways to improve the output from the compiler through a screenreader. This work will be not be tied to any particular choice of screenreader technology or editor.
This also presents an interesting area of research for improving the ergonomics of programming in an advanced statically-typed language for blind and partially-sighted users, which is by no means a “solved problem”, and offers much room for improvement.
Most of the work the Scala Center does contributes small differences to the lives of many developers in the Scala community. This proposal is different in that it provides a potentially enormous difference to the experience of programming in Scala to a small number of users.
But more than that, it furthers our goal of inclusivity; that Scala is a language which rightly knows the importance of supporting all its users, regardless of physical disability; to make it easier, in particular, for blind and partially-sighted people to become Scala developers, working productively alongside sighted developers, and having the greatest opportunity possible to share their valuable contributions in the field of software development.
Read this proposal here: “SCP-016: Accessible Scala“.