On Sat, 16 Jun 2001, Karl Hergenrother wrote:
> This is interesting reading on a subject which is a common thread on this
> list. Be sure to read some of the TalkBack messages included at the bottom
> of the reference. I don't expect many to agree with the conclusions, but I
> think that there are many valid points.
>
> http://cgi.zdnet.com/slink?/adeskb/adt0614/2774791:10151455
Here is the (lengthy) response I sent to my father when he pointed out the
article:
Well, there're several different perspectives, here. First and foremost,
while the Linux crew certainly would like to see Linux on the desktop, it
is *not* currently the primary force behind the Linux movement -- not even
terribly close. Server, server, server. Linux will only get to the
desktop when a) it's fairly easy to install (it's almost there), and
b) there are enough applications for end-users (including games!) that
they're willing to consider it as a viable alternative. This, too, is
happening, to varying extents.
There's one interesting point, however: Microsoft may actually be slitting
their own throat. Specifically, for "Joe Average Home Computer User",
they "pay" almost nothing for software: the OS itself is hidden in the
cost of the computer system, and Microsoft Office, which they naturally
have, is almost definitely pirated. However, starting in the *very* near
future (within a year or so) MS is going to start charging people on a
"subscription" basis, with myriad safeguards to "customize" (read: prevent
pirating) each copy of software for the system it's installed on. What
will happen? If folks want to continue to use MS, they'll have to start
paying money to do so -- real, honest-to-goodness money. If/when this
comes to pass, unless MS really stubs their toes on the copy protection
front, I could definitely see folks becoming more tempted by the free
software alternatives.
Is any of this going to make any difference whatsoever in gross
desktop-based OS/application sales for the next two years? I don't think
so. Is it possible that the latest manifestation of MS's greed, the
subscription-based paradigm, will start to turn off people even more than
they already are, to the point where they'll potentially pursue other
alternatives? I believe so. Linux is here to stay -- it's both nebuluous
and pervasive, simultaneously, and is always standing on the sidelines,
waiting for an opportunity. The opportunity really won't come unless MS
screws up... but I think they may be walking down that road right now.
Just for kicks, though, here're what I consider to be reasonable answers
to his "theses":
1) "Linux is too complex." This is like saying a car is "too complex" to
drive, just because I have no idea how fuel injection works. It's the
interface that matters, and several different options are up and coming.
(Check out www.kde.org or www.gnome.org for more info...)
2) Client-side Java. Ummm... maybe he hasn't noticed, but a) Java really
is pretty darn close to "write once, run anywhere," but, b) frankly, I
don't think it matters. Sun's screwed up Java's chances of becoming a
major factor, and now it's just a nifty way to do some web-based
programming.
3) Microsoft is on defensive. True, but I don't see how this is a strike
against Linux; maybe I'm missing something.
4) Linux as threat to UNIX flavors. Ibid.
5) Balkanization: wrong. People have been saying this for over five
years; if balkanization were going to occur, it *would* have.
6) OS is small component of TCO on a server. Okay, so... what point does
this make again? Apple claimed a lower TCO over Windows -- on the
desktop, not the server. (Actually, in the timeframe he's talking about,
*nobody* would have used *either* Apple or Windows on a server.)
Furthermore, it's clear he's never had to do remote administration of a
server: Linux, and Unix in general, is a *DREAM* to troubleshoot remotely,
whereas Windows has the whole GUI garbage, which almost requires on-site
staff to attend to issues.
7) Big hardware may co-opt Linux. Huh? IBM, Sun, HP, Intel, etc., all
have massive support for Linux -- IBM for their *entire* product line,
from a freakin' wrist watch, all the way up to their top-of-the-line S/390
mainframes. None of them have in any way co-opted Linux. This one's just
silly.
8) "Don't let [Linux] define you, [or] your IS shop." Why not? What's
inherently worse about it than Windows -- at least, from a server
perspective? It supports open standards (something MS doesn't), it costs
*DRAMATICALLY* less as an enterprise server (~$10K for a fully-configured
Windows 2000 server with e-mail back-end, SQL database, file/printer
sharing etc., vs. FREE for a functionally equivilent Linux box). Once
again, I completely fail to see what point he's making.
9) Linux will be a common operating system in [embedded appliances]. This
is a gimme: you can change the source, and you don't pay royalties. For
embedded applications, frankly, you'd be stupid to go with anything else,
and the embedded market is already reflecting this. I don't see what this
has to do with the desktop, though.
Bottom line: he makes many points, some valid, some, in my opinion, not.
While there *are* some valid arguments (as I outlined above) against Linux
on the desktop, I think he's done a wonderful job of failing to address
most any of them.
-Ken
**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************
