HAL must die!

Chad Perrin perrin at apotheon.com
Thu Mar 17 22:48:20 UTC 2011

On Thu, Mar 17, 2011 at 05:35:57PM -0400, Jerry wrote:
> On Thu, 17 Mar 2011 13:48:52 -0600
> Chad Perrin <perrin at apotheon.com> articulated:
> > I blame Microsoft, GNU, and Canonical for this trend, mostly.
> Chad, I believe I stand on firm ground when I state that you would
> blame Microsoft if the sun didn't come up tomorrow. You obviously must
> have a life time membership to "Slashdot".

I would not blame Microsoft if the sun doesn't come up tomorrow.  I
probably should have included Apple in my list, and in fact in this
particular case I'm skeptical that Microsoft is even as much to blame for
feeding this trend as Apple, at least in the cases of iOS and MacOS
"classic" (though MacOS X does mitigate this corporate impulse a little

Every membership at Slashdot is lifetime, as far as I'm aware.  If you
create an account there, I don't think they delete it unless you do
something really egregious to annoy the people who admin the site.  That
said, I've made comments to maybe two or three discussions at Slashdot in
my life, I've only submitted one story there ever (and it wasn't
accepted), and I have long since forgotten my login credentials for the
site because I haven't visited it in years.

You don't seem to stand on such firm ground, after all.  Are you sure
you're not thinking of some other Chad?  Maybe you saw something aobut
Rubyist and author Chad Fowler encouraging people to join a robots.txt
fueled boycott of Microsoft's search engine.  There's some crossover
there, given I'm a Rubyist and an author (of a somewhat different, but
related, variety) as well, though I never joined the robots.txt boycott.

> Microsoft creates programs, utilities, etc. for Microsoft. With very
> few exceptions, it does not actively create programming for a
> non-windows theater. It certainly never created any programming for
> *nix systems that deal with hardware discovery & configuration.
> Microsoft is not your problem. The people writing software, drivers,
> what have you for *.nix & *.BSD are your problem.

Microsoft encourages, employs, and supports a software design philosophy
that prioritizes user obsequious operation over user enabling
functionality, regardless of whether any of Microsoft's offerings are
compatible with any computing platforms outside of what Microsoft itself
produces.  By producing and marketing software that proceeds from that
philosophy, organizations like Apple, Canonical, GNU, and Microsoft help
push it into the mainstream within their particular market niches, where
each of them is significantly influential.

The people writing software, drivers, "what have you" for Unix-like OSes
(including me) are to some extent influenced by these things.  Despite
the fact that Microsoft is not really in the Unixy markets very much, it
has a disproportionately large effect on those markets because its
flagship platform is so widely used by people who develop software, even
when they are developing it for other platforms.

> Contrary to whatever you may think, the over whelming majority of users
> want a system that just works. They don't want to spend hours/days or
> more just tying to get a printer or sound card to work. They would
> rather leave that to people like you who have time to waste. Those of
> us on the clock don't have the luxury of fiddling with a piece of
> hardware when we could be doing something productive.

I never said that offering automatic configuration is bad.  I think it's
great, as long as it doesn't hinder, override, obviate, or even prohibit
user attempts to specify custom configuration.  If you think I was saying
nothing should ever be done automatically, you were not reading very
closely.  What I was saying was that trying to do things automatically in
such a way that it *breaks* the ability to engage in direct customization
when such is desired is a bad way to automate such things.

It's great if sound "just works".  It's not so great if some software
tries to make sound "just work" and fails, then prevents the user from
fixing it, when fixing it was trivially easy before that software was
integrated into the standard distribution of the system.

> Two statements in your post stand out as being totally bizarre.
> 1) <quote>
> People design software meant to eliminate the configuration and
> management hassle from the end user, but it doesn't always work
> perfectly.
> </quote>
> OK, what software always works correctly? That never fails, hangs, or
> just blows up. Deal with it, it is a fact of life. Ask any of a number
> of users who have tried to get OpenOffice to work as designed the first
> time; or even ever.

I never said any software "always works correctly".  There actually is
some software that does so, I'm sure -- but I was not saying that it was
reasonable to excoriate the developers for writing the software in
question in this case because there are bugs or limitations in the
software's operation.  That statement was not meant to be taken in a
vacuum; it is the set-up for the following statement, and I believe that
the fact you have taken it as a stand-alone comment conveying some
critical judgment that was not in fact evident in my phrasing is an
indication that you really are not interested in having a reasonable
conversation.  Indications, conversely, seem to suggest that you will
take anything I say in the worst possible light, even inventing
unreasonable interpretations of my words to support your belief that I'm
a bad person for saying something unflattering about Microsoft.

> 2) <quote>
> Unfortunately, it so zealously attempts to guess what the
> user wants that it effectively *disallows* easy fixes when the user
> discovers that something needs to be "fixed".
> <quote>
> Of course your statement is sans any documented proof, which in itself
> is not news worthy; however, who's specific configuration should the
> designers of said software use for a template? I know, yours, right? It
> is obvious that the designers are attempting to guess what the user
> whats. Obviously, they are not going to guess right 100% of the time.
> It is just the nature of the beast. Are you trying to say that you
> cannot manually change a configuration file? By the way, it will only
> get worse as no one can come to an agreement on one unified replacement
> for HAL. Nothing like fragmentation to make things work better. I
> believe that XFCE has dropped all support for HAL too. Is that
> Microsoft's fault too?

This is not the six o'clock news.  My statement does not need to be
"newsworthy".  It is a simple statement of my perspective and experience.
If you have some kind of evidence or reasoned argument to the contrary,
please share it; I'll be happy to consider it and, if I find it worthy,
change my perspective accordingly.  Simply accusing me of being some kind
of anti-Microsoft zealot based on the turning of the tides and astrology
(or whatever motivation you falsely apply to my statements) does not do
much to convince me I'm wrong, though -- nor does pointing out that I
have not cited scientific studies when making a simple observation based
on my own experience.

In any case, you seem to have completely missed my point here.  My point
is not that anyone should design software to enforce the configuration
that works best for me, specifically.  Rather, my point is that software
developers should probably consider how easy it is to make things right
when their software breaks.

Of course I can manually change a configuration file, but I think you
completely overlooked my statements about how things should really be
easier than a lot of current autoconfig software makes it when the
autoconfig needs to be overridden.

> I have two Linksys Wireless-N PCI cards in front of me that work fine
> on a Windows platform. FreeBSD doesn't even have a driver for them,
> thereby rendering them useless. I suppose that is Microsoft's fault
> too.

No, not really.  It's more the fault of the hardware manufacturer.

I don't know why you have such a problem with me that you are unwilling
to read my words as written, and just make up your own unreasonable
interpretations and misrepresentations instead, but it isn't very

Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ]
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-questions/attachments/20110317/711745d9/attachment.pgp

More information about the freebsd-questions mailing list