On Tue, 8 Oct 2024 23:12:05 -0000 (UTC)
Post by Lawrence D'Oliveiro... the reasons why freenix still struggles to gain mainstream
acceptance ...
I don’t know what you mean by “mainstream acceptance”, given that
Open Source dominates the entire computing world.
(Meant to get to this last week, but life got busy.)
So, this is a thing that a lot of freenix evangelists like to claim,
because it's kinda-sorta correct, in some contexts, but it also sounds
like it means much more than it does, which is convenient when you're
trying to build an argument off it.
FOSS Unices dominate certain specific segments, unquestionably: server
infrastructure and HPC, f'rexample. They're also making substantial
inroads in the embedded space, now that the typical microcontroller is
beefy enough to run them. They achieve this dominance for several
reasons: *A.* they're free, *B.* they're generally technically sound
and reliable, *C.* they're free, *D.* they're Unix, which is a mostly
standard environment with 50+ years of familiarity, *E.* they're free,
*F.* these spaces rarely need UI more complex than the command-line
anyway, and *G.* they're free.
Which is great, congratulations, golf-clap. But it's not really
"mainstream" in any sense except that most IT departments, given their
'druthers, would opt to build their tech infrastructure off something
that - while a bit clunky to use - doesn't cost them anything but is
nevertheless pretty reliable. Which is not exactly a cosmic revelation.
Outside of the "underpinning stuff people need to not have to think
about as much as possible" department, things are *very* different.
Techies generally prize their own particular definitions of technical
excellence (and their own personal ideological hobby-horses) over user
experience, but for most people it's the other way around - they prefer
something that is pleasant and intuitive to use over something that may
be better "under the hood," but has a weird or clunky user interface.
FOSS stuff still sees acceptance in this space, but not to nearly the
same extent. To take graphic-design work as an example, people don't
use GIMP because it's better than Photoshop, they use it because it's
cheaper than buying Photoshop and easier than pirating it - and the
majority of working professionals...use Photoshop, because its workflow
is greatly superior and that makes a real difference in their ability
to work, so much so that they're willing to put up with all the heinous
licensing bullshit Adobe foists on them.
(And this is the case *even though* GIMP's backend functionally is
largely comparable to Photoshop's - which handily illustrates my point.)
And in terms of personal-computer operating systems (and the associated
desktop environments?) It's not even *close.* Taking for an example the
data at https://gs.statcounter.com/os-market-share/desktop/worldwide
(dunno from these people myself, but they're evidently legit-ish enough
for Ars Technica,) OS market share over the last 13 mos. has Windows
sitting at an average of almost 72%, a figure that is technically known
as "an absolute curb-stomping."
And Linux? It's sitting at somewhere south of 4% - in other words, less
than one in twenty-five PCs is running it. If you count Chrome OS in
with that, it still doesn't crack one in fifteen, despite the fact that
piles of schools in the US hand out Chromebooks like breath mints. (But
then, they *last* about as long...) Even if you're generous and count
the "Unknown" and "Other" columns as probably some freenix flavor, OSX
(which is Unix, somewhere down under the layers and layers of NeXTSTEP/
Macazoid stuff, but isn't "free" in any sense) outpaces it seven to
five, and that's with "being able to afford a Mac" as a handicap!
Of course, there's a number of reasons why that's so - but not nearly
as many as there used to be. Grandma probably *could* do all her book-
facing and instant-gramming and whatnot in Chromium on LXDE...but she
isn't. Even gaming is much less limited than it used to be (thank you,
Gabe Newell,) but the vast majority of gamers have stuck with Windows
despite that, and it mostly serves to make the Steam Deck a viable
product (congratulations, Gabe Newell.) If most of the reasons people
*couldn't* jump to freenix are gone or nearly so, and yet only a tiny
fraction of them have done so (Linux market share has more than doubled
in the last five years, but in context that's like saying Grand Fenwick
substantially expanded its territory by annexing a neighboring farm,)
that strongly suggests that there are reasons why they don't *want* to.
And - to circle back 'round to the actual point we were discussing -
where FOSS alternatives genuinely do dominate in user-oriented spaces,
it's pretty inevitably the ones that actually *do* put some real effort
into presenting a consistent, intuitive user interface. Mozilla's had a
bad habit of (badly) copying Chrome's (dumb) ideas the last decade or
so, but they attained the following they did by making a legitimately
huge improvement in usability over IE. LibreOffice has learned lessons
from MS Office that *Microsoft* still hasn't learned (like "everyone in
the world hates your stupid Playskool activity center, please just stop
it already.") Android's QC on third-party applications may be spotty,
but the core user interface is a perfectly reasonable copy of iOS.
User experience may not be priority #1 for programmers - but it really
does matter. And usability and technical excellence are not mututally
exclusive, at least not to remotely the extent that programmers like to
think they are. And the fact that many, many freenix advocates like to
brush off usability concerns is certainly not the *only* reason freenix
struggles to gain more than a small fraction of the personal-computer
market, but it sure doesn't *help.*