Comments embedded with the quotes.
On Tue, Dec 8, 2009 at 5:43 PM, Duncan <1i5t5.duncan@xxxxxxx> wrote:
> Boyd Stephen Smith Jr. posted on Tue, 08 Dec 2009 06:36:16 -0600 as
>> On Monday 07 December 2009 23:44:25 Thierry de Coulon wrote:
>>>
>>> May I ask what changes with free programs as long as you are no
>>> programer? I love this affirmation but it's very theoretical for the
>>> average user....
Actually it's the love of C++ for application level programming that
is really limiting the growth of Linux. C++ is a language that has
never really appealed to the masses. Full of archaic syntax necessary
back in the days when people programmed using editors like VI and did
so from really primitive terminals. One of the biggest things that
really expanded the abilities of DOS users was the legion of people
programming in languages like Pascal, Basic, Fortran, etc. Languages
that tended to have stronger mass appeal and took far less time to
master. It also takes far less time to code as your spending a tenth
as much time dealing with syntax and spending that saved time on
program logic. Your seeing kinda the same thing with the huge
outgrowth in PERL & Python programs used in default distros today.
C++ makes perfect sense writing a sound system, drivers, the kernel,
and lots of the goo that makes the OS work and play well together. It
however locks away the unwashed masses from ever contributing to
Linux. There are millions of eager volunteers who are never going to
master C++ and to be honest you don't want the unwashed masses writing
in a language so prone to memory leaks, buffer overflows and other
serious security issues. There is no need for such fine control for
app level programs. If these waves of people started contributing in
C++ it'd become a M$ like security nightmare. Instead if they code in
languages less prone to such issues there would be a legion of people
contributing to the functionality of Linux and other FOSS OS's. Folks
happy to spend their free hours and donate that work to the common
good.
Even people like myself who once made a living writing C code, many of
us would rather chew our legs off with a dull hampster than write a
large app in C++ any more. I can write the same app in Gambas in at
least a 10th the time. I could do it in Pascal once I brushed the rust
off me Pascal in a third or less. There are a 100 languages out there
they could do this in. That is if APIs were exposed too those
languages. It's a chicken and egg thing.
Me I say build it and they will come. When they do, it'll leave the
more advanced programmers free to work on more advanced topics. We all
benefit and we gain not only more apps but the load of FOSS proramming
is shared out among a greater group of people. Think about how many
times you've heard a comment like this one about wanting to help but
not being a programmer. Most of these "non" programmers could have
code up in running in a month or two in something like Gambas. That's
the whole point of the visual languages. To concentrate on program
logic and flow and minimize syntax. If just 1 in 100 got serious
about it and mastered these languages that's tens of thousands helping
expand the boundries and capabilities for all of us.
>> In practice, that singular entity rarely dismisses a modification
>> entirely, but does exert monopoly control.
It is common to fork a project when such "control" conflicts with a
large group of people's wishes. Sometimes the forked version dies off,
sometimes the two merge back and sometimes the origional branch dies.
Sometimes both keep on going and morph into very different
applications. Happens all the time and I see a fork coming in KDE.
There is enough revulsion for KDE 4 in the hard core KDE base too
warrant a fork. Whether the fork can attract enough coders to keep it
viable will be the question. Another question is which way existing
KDE developers will go. Not all of them can be happy with this
abomination called KDE 4. Starting to make XFCE look good.
> Would you buy a car with the hood welded shut, that had to be taken to
> the dealer even to fill the window washer fluid?
Ah but how much different is software using this analogy when the
developer doesn't weld the hood shut but holds one ransom with various
dependencies and using libs/code that make it extremely complex to get
under the hood. Attempting to modify KDE for example is a daunting
task for any developer. Yes it can be done, even by an individual who
is determined enough. However then comes the next bug fix or
dependency upgrade that wipes all your work out.
Being FOSS doesn't mean it's really free. Nor can FOSS developers be
expected to exert the effort to make things accessable to everybody.
Some software is just too complex and the effort to give users greater
access too the code just not feasible given how little time it's
possible to devote to just functionality. Few FOSS coders are paid
for the work they do. They can often sneak it in at work to an extent
but only so far as they can prove it's value too their employer who is
not going to buy making it easier for others to tweak their code.
> One real practical effect of this on the FLOSS community in real life, is
[nvidia rant snipped]
> fewer users who chose to play nVidia lackeys, the entire xorg ecosystem
> would evolve faster. Here, nVidia isn't holding only nVidia lickspittles
> in servitude, but the entire community is held back.
Nvidia is one of many such entities but this is a chicken and egg
problem as well as one of practicalities.
Lets start with Nvidias foot dragging on drivers. Nvidia of course
being one of the few companies to actually provide reliable Linux
drivers so it's hard to criticize them for being slow when they feel
and by the market stats can prove they have no viable interest in
providing Linux drivers at all.
The market stats are in error. Mostly because so many Linux users are
invisible. We just stand around and take it. We write work arounds, we
run stuff in virtual machines, we use emulators to get things to work.
Mostly we don't utter as much as a peep. A large percentage of us
built our machines from scratch or bought a machine with windoze on it
and installed Linux on top of it. Those few companies who've offered
Linux have had difficulty because instead of buying a machine offering
Linux we save a buck or two buying a windoze box and thus are counted
as Windoze users when it comes to Market share. We don't send in
complaints, we either find a work around or do without.
So how can we expect Nvidia or Adobe or Apple to take us Linux users
serious? Until we make our presence felt, until we appear on the
market stats we won't get those drivers. I've been trying for 3 years
to get a very common Tascam break out box to work under Linux. There
are thousands of these things out there but Tascam's reply boiled down
to "We don't support Linux, we do not see Linux as having any real
market share and until we do we will not support Linux". Ironic since
it has Apple drivers yet Linux has equaled or topped Apple in desktop
share. It is because we are invisible that we are treated this way.
On the practical side. I have a MB with Nvidia drivers. My options
were order the MB on the web, wait for who knows how long for it too
arrive, hope I didn't have to cross ship it because it was damaged,
option 2 was drive to Houston (800 miles away) as that is probably the
nearest storefront I could have purchased a MB that didn't use an
NVidia card or use Nvidia. I chose to use Nvidia. I needed this
machine up imediately. My last cobbled together system had died a
horrible death and I was without computer. I'd rather do without food
:)
Nvidia is easily the top selling Vid card out there. It comes on many
of the propriatory machines, it is on the vast majority of low and mid
level MBs you can buy. It is the top dog when it comes to Vid cards
today. An effort by Linux users to boycott Nvidia would not only be
futile it'd be counter productive and since when has Linux users
banded together about anything? I can't even get help with
saving/introducing features ideas to help me convert windoze users.
There is a deep arrogance at the heart of the FOSS development
community. You have to respect the effort they put out and the high
quality of the work, but at the same time resent the often dictatorial
way they intereact with the community as a group. One on one almost
every FOSS developer I've talked too has been eager to help. Several
have sent me new builds to see if the bug/feature I've brought up to
them worked. It is as a community that the watch gaurds who rarely
actually develop Linux but who have taken up the role of isolating the
development community which I see most of the arrogance.
Some of it is necessary. Not every feature request makes sense and the
sheer noise level of even the good ideas would leave them no time to
actually code. Needs to be a middle path in my opinion. One where more
feedback reaches the developers but where it's condensed into levels
they can actually get a pulse on what's going on without having to
read each and every comment.
The KDE developers I'm sure are blissfully unaware of the outrage
generated by KDE 4. I've seen it on dozens of lists since KDE 4 came
out. I've never seen a software release spark this much of an outcry.
Only 2 things can happen. Features are re-introduced allowing fans of
both to have it their way, such as has happened with the menu. You can
have old style or new style menu or both. Or there will be a fork. The
dissatisfaction isn't going away. People loved KDE. It has THE most
fanatical following of all the desktop managers, or at least did. I've
seen lots of people turn away from KDE already. While Gnome and KDE
users wound up about even in most polls or a slight edge to KDE, the
most vocal and adament were always KDE users. Many of which are now
lost, the title of this thread says it all. We have lost KDE. What is
called KDE 4 is not KDE. It is a pale ghost of what once was.
___________________________________________________
This message is from the kde mailing list.
Account management: https://mail.kde.org/mailman/listinfo/kde.
Archives: http://lists.kde.org/.
More info: http://www.kde.org/faq.html.
