Looking at KDE, waiting for the other guy

By Jack Bryar
NewsForge Columnist
Open Source business
Which would you rather do, wait for a gee-whiz desktop interface and
office suite that can do everything you want (and more) — or would you
rather get something that actually works? Or would you still do
nothing?

The folks over at the KDE team hope you’ll use something that works.
I think their newest offering does just that. I’ve been working my way
through KDE’s
newest offering and recommend you check out their newest slideshow
presentation concerning Kopernicus, aka KDE 2.0. If you really
really like it, maybe even develop on
it, and especially start to use it, that could make a big
difference
to a whole collection of Linux and Unix vendors.

While everyone in the Linux community has been in denial about it,
there is a division of opinion among Linux vendors concerning what
desktop platform to support. While most vendors say they support both
GNOME and KDE, they usually make it clear where their heart is. Firms
like Caldera, Corel, Mandrakesoft and SuSE Linux in particular have
promoted the KDE desktop.

There’s a lot to promote. Kopernicus includes the core libraries, an
improved desktop, the Konqueror
Web browser, a prototype office suite (KOffice), plus utility and
administration software, plus stuff for games, graphics, network
utilities, and more.

Is the desktop important? For months, many Linux vendors have
suggested that the workgroup and small Internet server market (where
Linux has developed huge momentum) was more than sufficient to carry
the
OS and its primary vendors. The sinking stock prices of many Linux
distributors suggest otherwise. Caldera has dropped nearly 40% of
its market value in the last month. Red Hat has lost
better than 30% of its value in the same period. And while vendors
like Microsoft and Apple have had their own stock troubles lately, the
low valuations of many Linux software vendors reduce their ability to
acquire other firms, or raise capital, and make them potential takeover
targets. Capturing desktops and lots of them, could become critical to
maintaining corporate independence over the longer term.

Can that be done?

For the longest time it was an article of faith among Linux
developers that the corporate desktop marketplace was nearly
impenetrable. The installed base was too embedded; losing Windows meant
losing customer’s key apps and the data that resided on them. The
technical competency of many corporate IT managers was too limited;
some
kid with a Microsoft certification wasn’t likely to be able to do much
with the array of tools that Linux apps provided. The cost of support
was hard to calculate; therefore relative cost of ownership of Linux
vs.
Windows remained an unknown. The biggest problem was that other firms
had Linux on their desktops. As every IT manager knows, there’s real
risk associated with being first.

Increasingly that reluctance is beginning to break down.

I was asked by a client last week to help him build a case for
replacing his Windows installed base with Linux. The firm has hit rough
seas recently, and the company decided it couldn’t afford to drop a
couple of million dollars on spiffy new hardware and software. It was
also looking at maintaining a prohibitively expensive customer support
system that it had bought two years ago and had never fully
implemented.

If there was a Linux or Open Source alternative to his dilemma, this
tech officer wanted to know about it. If I could surface evidence that
other companies were looking at an alternative to Windows, he wanted to
know about that, too.

I’m still looking for that evidence. I spoke to about 60 mid-sized
firms and found that 20 of them were going through a similar exercise,
but only three had actually begun to port desktops to Linux in any
significant way. Those are interesting numbers. They suggest that the
potential market for a viable Linux desktop is far larger than any
study
I’ve seen to date. But it also suggests that the barriers to adoption
remain high, whether they are critical custom Windows apps, technical
competency of the IT staff, or plain ol’ Fear, Uncertainty &
Doubt.

These issues are important to my client. The tech officer I worked
with expressed considerable fear that the company’s IT staffers would
have difficulty managing any conversion. In addition, many employees
have all sorts of downloaded software programs and browser plug-ins
loaded on their systems. Oddball software is another worry. One
department runs an old custom software program with hardly any
documentation, a constant headache now, but a potential nightmare if
the
company chose to convert. Mid-conversion catastrophes are another
concern. It’s no big thing to convert a few dozen NT stations to
Win2000
each day. Nothing breaks down during the conversion. Can the same be
said if the company began to convert to Linux, the tech officer wanted
to know. Seeing what someone else had done, learning from someone
else’s
experience, was critical.

So is the only barrier to widespread adoption a lack of early
pioneers?

Maybe. So the next question is, if those pioneers try KDE or GNOME
on
for size, will their experience be good enough to help Linux, and many
Linux vendors start to really break out into the larger corporate
marketplace?