Reflecting on reflective computing.

[David Manifold]

On Wed, 21 Oct 1998, Christopher Barry wrote:

> We have language compilers for everything under the sun, and people
> continue to develop these things and similar things so that they can
> develop more development tools and it goes on... most development effort
> in the free software community isn't focused on making the applications
> that computers could actually allow us to do useful work with to
> accomplish goals in the real world, but only to further the development
> process itself, which is pointless.

People are probably writing new dev tools because it still is very
difficult to write software.  This is a main issue adressed by TUNES.
Look at how long software takes to be developed.  Then look at how much
work/time was: 

* Duplicated, done before by someone else in some similar, or even very
different product, because people were unwilling or unable to share code,
or didn't know about the other project that they could have used code
from.  We can't do anything about the unwilling, but we sure can about
unable and ignorant.

* Realling making use of the programmer's skills, or was it just time
spent trying to think like a computer (linearly)?

* Spent debugging.

* For the purpose of working around existing software or hardware
limitations.

* Actually went toward programming, instead of going back and forth
between applications/windows/menus/directories/sections of the same
document.

* Spent rebooting. (kernel crashed, or we were developing the kernel)

* Spent waiting on the computer to do something while it ignored input.

* Went toward figuring out how to optimize special cases to get decent
performance.

David Manifold <dem@pacificrim.net>