[Wylug-discuss] Linux GUI design [Was CUPS]

[Dave Fisher]

On Fri, Feb 27, 2004 at 12:50:27PM +0000, John Hodrien wrote:
> On Fri, 27 Feb 2004, Dave Fisher wrote:
> > I suspect that some WYLUG members will find such views quite abhorent,
> > but I would like to set off a _serious_ debate about this issue.
>
> Why abhorent?

It is a sad fact of human behaviour that we often treat other people's
failure to identify with our own values and commitments as threats to
our personal identity, assumptions and worldviews.

When combined with the practical impossibility of value-free judgement,
this fact often turns what might otherwise be considered no more than a
different opinion into a heresy that must be denounced in order to
justify one's own personal preference.

Personally, I see no intrinsic reason why those who wish to treat Linux
as a hackers-only paradise shouldn't continue to do so.

On the other hand, I have yet to meet anyone who managed to practice
that perspective with anything like consistency, i.e. to show complete
disinterest in the doings of non-Linux systems, non-hackers and their
effects on the computing ecosphere.


> > Ideally, I'd like to assume that Raymond's perspective is correct
> > and discuss the positive things that can be done to rectify the
> > situation, but if there are insuffient numbers of people willing to
> > make that assumption, I am more than happy to make the case for it.
>
> Hackers' wives should be more willing to test their creations...

The statistically accurate assumption about Hacker gender does raise at
least one question about how positive a suggestion this might be, i.e.
how many WYLUG members are hackers wives? ;-)

When debating the issues of what to do and why to do it, rather than
discussing how to achieve a goal that has already been
assumed/determined, it's useful to keep the agenda relatively open.

Having said that, I'm happy to start the ball rolling by expressing some
provisional ideas about the elements of current hacker which prevent the
development of good GUIs, and some of the experiments we might pursue to
overcome them.

I don't have the time to compose a detailed analysis right now, but the
tendency to treat collective project work as a zero-sum game is clearly
one handicap that must be overcome.

Forking is an an ecologically normal process which can have incredibly
beneficial effects in terms of experimentation and creativity, but it
can also be incredibly wasteful.

Biologically speaking, evolutionary forks don't produce better
organisms, they merely produce variations which may or may not be better
suited to particular conditions.

The more interesting contributors to contemporary debates in genetics
recognise that human evolution transcends many 'natural' evolutionary
constraints, precisely because humans  modify and even invent the
conditions for their future development.

I'm inclined to think that Hackers need to do something similar if they
want to free themselves from the sociological and economic constraints
of the current environment, e.g. find an itch to scratch which irritates
them and which irritates non-hackers, and make a _joint_ effort to
develop solutions which satisfy their different goals through a common
toolset.

The list of projects which could satisfy such criteria is endless, and
in most cases there is plenty of basically sound and pre-existing OSS to
build upon.

I am sure that other people can come up with more interesting and/or
representative ideas than I can, but here are just two that I've
pondered recently:

1. A fully-featured set of bibliographic tools for OpenOffice

   Bibliographies aren't that interesting, but education/research is
   probably the easiest market for Linux to conquer, and the absence of
   uasable bibliographic tools is an absolute block on its widespread
   adoption therein.

   Latex/BibTeX/SQL has much of the functionality, but decisively lost
   the evolutionary battle with MS-Word years ago.

2. An integrated video/audio editing and authoring GUI

   This is an area of huge consumer/business interest and, most
   importantly, a complete obsesssion for the media organisations which
   dominate our perception of the world.  Change the media view of Linux
   and you change almost everyone's view of it.

   Virtually all the basic tools exist on the command line, and there
   are plenty of partial/esoteric GUIs around, but nothing that would
   let a dumb end-user capture, edit and re-edit multimedia content with
   all of the commonly used formats and devices.

Lunch break over, so back to work ...

Dave