Date: Mon, 29 Mar 2021 11:40:04 -0700
From: Bret Victor
Subject: Re: Concerns about accessibility for people who are vision-impaired (from the little that I’ve seen and read about)
Hi Tracey,

A primary purpose of Dynamicland's form of computing is to restore tactility and tangibility to daily work -- working with your hands, with physical objects, in physical space.  I believe that this approach will prove to be far more accessible to the visually-impaired than current computing environments which represent all work as a virtual world on a flat tiny screen, and then attempt to layer so-called accessibility features on top of this fundamentally vision-oriented foundation.

Many years ago, I was doing a project to build an assistive tool for the visually-impaired, and among the people I interviewed was a fully blind electrical engineer.  (Watching a blind man fluently solder was one of the most remarkable things I've seen in my life.)  He used off-the-shelf electronics equipment (oscilloscopes, function generators, etc) that he had lightly modified (some braille labels, audio output) and served him fine.  They worked for him because they were all physical objects that he could carry around and manipulate with his hands, with physical knobs and buttons and such, and that he could physically modify as needed.

Today, such equipment (e.g. the classic Tektronix scopes) is gone, replaced by screen-and-mouse-based apps (e.g. Saleae).  The apps have no physical features, they cannot be touched, they cannot be modified.  What has happened to oscilloscopes has happened or is happening to almost all tools in all fields, and I believe this to be devastating to a wide range of humanity, not just the visually-impaired  With conventional computing, adding computation to a tool means virtualizing it into an intangible, vision-oriented world.  Screen-based accessibility features are not remotely an adequate compensation for removing tangibility in the first place.

Another thing I noticed while visiting visually-impaired people was that they enhanced their homes with assistive cues -- labels on jars and shelves, arrangements of objects and furniture, etc -- and these cues tended to be custom and idiosyncratic.  Everyone figured out a system that worked for them, and they had the freedom to modify their environment to evolve and implement their system.  This may sound obvious, but conventional operating systems and apps do not work like this, at all.  OSs and apps are one-size-fits-all virtual worlds, designed by designers at tech companies and imposed identically on everyone, regardless of who they are.  Manufacturer-provided customization features are, again, not remotely an adequate compensation for the ability to build and modify one's own physical environment.

Central to computing at Dynamicland is the idea that a computing environment is not a product, but a set of practices that you make for yourself, for the needs of yourself and your community, using physical objects in physical space, with the minimum amount of virtuality.  Of course, whatever virtuality remains (e.g. text projected on a piece of paper) could use standard accessibility techniques (e.g. magnification, audio readers) as needed.  But the foundation is manipulating real things in the real world, and I believe this will prove to be the most enabling foundation for most people who have use of their hands and bodies, especially the blind.  (The physically-disabled and physically-disinclined may be better off with a higher degree of virtuality, and that's fine too.  They can do things their way.)

The Dynamicland organization is currently a two-person non-profit with minimal funding and a lot to do.  We are inventing an alternative to a screen-based status quo backed by a multi-trillion-dollar industry of hardware manufacturers, app developers, and investors, all of whom implicitly promote the assumption that all computing is screen-based computing, and thus progress towards accessibility means developing features that provide alternative interpretations of visual virtual worlds.  Alternatives to the assumption itself are almost unthinkable.

We hope to get the foundations right.

-Bret


On Mar 20, 2021, at 6:04 PM, ************************