[snips]
On Tue, 17 Apr 2007 21:35:39 -0500, Linonut wrote:
> After takin' a swig o' grog, Hadron Quark belched out this bit o' wisdom:
>
>> Kelsey Bjarnason <kbjarnason@xxxxxxxxxxx> writes:
>>
>>> Give it a week. Even if it sells a million copies, we'll have DFS or some
>>> other twit harping on about the microscopic market share of Linux,
>>> completely oblivious to the existence of other distros, free downloads,
>>> allowed copying and the like.
>
> Right on cue....
As predicted. I wonder why they enjoy being so completely predictable -
and so predictably and consistently wrong?
>> The market share is easy to define for anyone looking for real numbers
>> and who can think a little - the share of possible installs. Its
>> really not that difficult no matter how you COLA dolts try and wriggle
>> out of it.
>
> Then teach us, O Hadron, O Fellow COLA Dolt, we beseech thee.
We do? Personally, I'd rather he and a couple others - DFS, flattie, etc
- simply vanished. They don't even add comic relief anymore.
That said... "the share of possible installs". That's an interesting
notion. So now he's going to take into account every single 386 or later
machine ever built, is he?
If we assume nothing prior to, oh, Win95 is remotely relevant and want to
constrast "Windows" with "Linux", we'll need to include all the 386s.
Let's not forget, though, that by today's standards even Win95 on a 386
isn't particularly usable - yet it can be eminently usable with Linux -
turn it into a print server, for example. A mailer for your LAN. A
caching news server for your local network. All sorts of things. Hell,
it's even usable as a *desktop* system with enough memory. Yeah, you
could also get Win95 working in such a state, but with Linux you'd be
using _current_ software, not something better than a decade out of date.
So... every 386 or later machine is on the table, and of course, Hadron
has all the sales figures for every such machine. He also has the figures
showing exactly how many of those machines have been destroyed - not
simply replaced with newer machines, but actually destroyed so they
*cannot* be used for something else.
I'm quite certain he has no such figures, yet he'd have to in order to
make this concept of "possible installs", since the only limit on how many
installs are possible is how many viable machines are in existence.
Assuming he can ever support that claim, then he's right back to the
existing problem, that there's no way to count the actual usage of even
Windows, never mind Linux, so his "possible installs", difficult enough in
itself to determine, still offers nothing.
How these people can heap unknown atop unknown atop unknown and convince
themselves they know something as a result is not quite clear.
|
|
