[Linux-aus] Computer users and linux: moving upstream to software producers.

[Janeene Beeforth]

While the debates about advertising/docos/etc have been interesting,
there's one critical thing that I've always felt is being missed when
discussing the potential migration of people from Windows to Linux - and
that is whether the system is actually capable of running the particular
software that the end user actually wants to run.

At this stage, there's alternatives (and in many cases *very* good ones)
for general run-of-the-mill tasks such as web, email, work processing,
etc.

A lot of people however don't get a computer *just* for general
run-of-the-mill stuff - they get it 'cause they want to be able to do
something more specific with it. A classic area in point would be computer
gaming.

This is where (through no fault of it's own) Linux falls down. In the
majority of cases, games are produced for Windows only.

Look further into the games development, and you find that the *tools*
used to produce the games (stuff like havok, gamebryo, etc) will only
produce games for Windows or Windows consoles. This means that the
developers of the game get tied to a specific OS by the tools that they
feel are required to be able to develop the game.

While we've got things such as Wine to try and get this stuff working
under linux it's still very hit and miss (again, through no fault of the
wine guys). The perception amongst the users however is that 'linux is not
able to run direct X' and therefore inferior (somehow the fact that direct
X is owned, designed and undocumented by MS never seems to occur to them).

If a person knows, or even *suspects* that by choosing Linux they will not
be able to run the programs that they want to run - they're not going to
seriously consider it as a choice. For most people, dual-booting is simply
too complex.

If we want people to feel that linux *is* a viable choice, we need the
upstream software providers to seriously consider it a choice. They need
the tools to produce the software, and to *know* that those tools are
available, of production quality, and at least as easy to use as what they
currently use.

Getting together information about the tools that *can* be used to produce
cross-platform (Linux/MAC/Win) games *of at least* the same standard as
recent releases (shaders, physics, etc). For example, if you were to
produce a game like Oblivion as a cross-platform game rather then
Windows-only, what  tools are available to duplicate this development? Are
there any tools that are currently missing?

Educating the development companies in the use and potential benefits of
these cross-platform tools - getting them to move *away* from lock-in
libraries such as direct X - can flow the benefits down the chain to give
end users a *real* choice.

Cheers,
Janeene.