Desktop Linux (fwd)

[bmcculley at rcn.com]

(nb, I use FOSS and Linux interchangably herein)

Derek Martin wrote:
>
>Often, the existence of an alternative is inadequate.  Many 
>people, believe it or not, live in MS-Office all day long.  If 
>the documents their co-workers produce don't import 100% 
>correctly into whatever alternative they use, then it's no 
>alternative at all.

I think this is more pervasive and subtle than we realize.

This thread started out talking about the family desktop, as
benefitting from the business desktop adoption of FOSS.  But
the desktop marketplace is more fragmented than this
discussion model has recognized.  

For one thing, the home desktop is everything from the
Leave-it-to-Beaver non-technical family that will use what the
breadwinner workplace uses, to the Linux penguinistas that
will re-invent the wheel just because they'd rather do that
than buy someone else's.  Neither is much of a market to
support software economics justifying Linux products.

I think there's another big class of users that may be more
important, the so-called SOHO users.  The small office/home
office population includes a lot of freelancers, who need to
be able to exchange work product and work collaboratively with
clients and virtual teams.  AND THEY DON'T HAVE TIME TO WASTE
FIGHTING WITH TECHNOLOGY INCOMPATIBILITIES.  So if they are
not Linux geeks they won't adopt Linux until their client base
does, or until the compatibility problems are solved (which
will be shortly after the Stanley Cup is won by Hades, since
the vendor maintaining the de facto standard has a vested
interest in breaking compatibility with their monopolistic
products - viz Kerberos).

>
>But in order for things to progress for /US/, we need /THEM/.  
>IOW, we need a market for high-quality professional 
>applications, so that they will be produced.  Why doesn't 
>Adobe port Photoshop to Linux?  Simple: there's not enough 
>market for it to make it worthwhile.  GIMP is nice,
>but anyone who does professional image work will tell you that
>Photoshop does a lot more.
>

That's the crux of the issue.

Those for whom technology is a means not an end want tools
that best serve their needs.  With all due respect, FOSS
applications are generally going to be at a disadvantage to
best-in-class proprietary apps, because of the nature of the
development process and the developers, and free market
dynamics.  As Derek's comment implies, we need to look to
proprietary vendors for the high quality professional apps,
and they are inevitably subject to the constraints of market
economics.

So the platform may be FOSS, but the apps will not be, at
least not for the niche apps.  And to get any niche apps onto
the Linux platform it has to be sufficiently widely deployed
to make the market size significant enough for the app vendors
to justify the port.  That will be driven ultimately by the
general purpose apps, the Office suite equivalents, enabling
enough businesses to adopt Linux to provide an installed base
that supports some of those niche apps.  Technology will help
too, as porting becomes easier.

Ultimately the SOHO market will be the last desktop
battleground for Linux to penetrate, and the professional
niche apps will be the last holdouts on the desktop.  But
eventually those frontiers will be tamed and there will be
competitive available choices there too.  I hope maddog is
correct about the timescale, although I tend to be more
pessimistic.

>
>Nope, but if they can't look at the schedule that their 
>manager has made for them in Outlook/Exchange, and their 
>co-workers have trouble importing their documents properly 
>into MS-Office apps, then it doesn't matter if they can do 
>it another way.  If they can't work with the rest of their 
>team, they're totally screwed.
>
>> Is it *really* that bad?
>
>Yes.
>

No, it's worse.

First hand example.  A few months ago I was working in a
corporate environment using MS Project.  I wanted to find an
open source equivalent.  Nada.  Nothing.  Not even close.

I did finally find an open source project planning tool, which
was far inferior to Project.  Not just inability to import and
export files, the functionality was mediocre in comparision.

To top it off, it took me significant effort to even find that
app.  There may be others that I didn't find, but if I don't
find them they don't exist as far as I am concerned.

Issue here is, I don't have time to waste searching for an
equivalent application.  I need to get a job done.  This is a
tool I need to use, if it's not in the toolset then I'll use
what is there.  Even if it's Micro$oft.

Point is, there needs to be vendor marketing of apps to make
things easy for me, and other users.  That again presupposes
there are proprietary vendors to make that marketing
investment, after the apps exist because they make the
development investment.  (Another reason FOSS apps will always
be at a disadvantage, btw)  Of course vendors who have a
proprietary interest in not porting apps will be another
factor, I don't expect to see MS Project on Linux anytime soon!

To summarize, I'd say that what it will take for Linux to win
the desktop is a strong proprietary applications vendor
community with products running on a FOSS platform
distribution that's easy for non-technical resource-challenged
organizations to install and administer.  Oracle et al are
making a start in that direction, the next big step would be
some desktop app (Lotus Notes, anyone?).  The distribution
will be another issue, remains to be seen.

But that still won't be on the home desktop, I'm not sure
Linux will win there before the nature of that market evolves
into something completely different.  Too soon to even speculate.

-Bruce McCulley