Date: Tue, 31 Aug 1999 23:35:27 -0500 (CDT) From: Jay Nelson <jdn@acp.qiv.com> To: chat@FreeBSD.ORG Subject: Re: Why? (Was: Re: FreeBSD, the follower of Linux ?) Message-ID: <Pine.BSF.4.05.9908312034570.1356-100000@acp.qiv.com> In-Reply-To: <4.2.0.58.19990831184900.00a3ed40@localhost>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 31 Aug 1999, Brett Glass wrote:
>At 06:57 PM 8/31/99 -0500, Jay Nelson wrote:
>
>>Excuse me for butting in, but I have some serious questions about the
>>whole tenor of the "advocacy" issue and some of the complaints Brett
>>is raising.
>
>That's fine and healthy.
I'm not so sure;)
>>Why should we evangelize? What value is there in selling any Unix OS
>>to the Maudie Fricks of the world?
>
>The main value -- both socially and to users in particular -- is that
>their computers will be reliable and secure. Daily, I assist users
>whose Windows and NT systems have crashed, lost data, been penetrated
>and corrupted, etc. We are *all* hurt by this, as it raises the cost of
>goods and services and prevents businesses from serving us as well as
>possible. On a more individual level, it hurts people by destroying
>their work, eating their valuable time, and frustrating instead of
>assisting them. UNIX isn't an ideal operating system by any means,
>but it can certainly help this situation.
My experience is that users don't give a damn or care -- as long as
they don't have to invest any effort. Nor do they want to deal wth the
necessity of plugging in a network cable to make their system work.
Their perception is more important than ther reality.
OK -- this is admittedly a jaded view, but one I have confirmed more
frequently than any other. From an admin point of view, I would prefer
they use _some_ real OS, so I wouldn't have to walk all the way over
to their machine to return configurations back to sane values, but
that ain't gonna happen.
[snip]
>Even Windows really isn't "about" the point-and-click user; it just seems
It's not? My experience asserts otherwise. Sorry.
>to be initially. (Ultimately, the user needs to know a bit about what he or
>she is doing, just as you need to know where the gas cap is on a car. If you
Ahh... therein lies the rub. You are in the publishing industry. I've
spent many years in nearly the same industry. From what I've seen
produced, most users don't have a clue about what they're doing. "Good
enough" was the phrase I heard over and over. How would any OS help
them produce better quality? As a M$ promotional years ago promised,
they'll be "professionals in an hour". Any Unix isn't likely to change
their attitudes.
>don't, the car will eventually stop.)
>
>But I am more optimistic about the development of GUIs for UNIX. I think
>we will soon see not one but several options which are usable by naive
>users.
There's no question that the GUIS are getting fatter and fancier. But
if you are to retain the power of X, how can you shield the users from
the complexity of X? Conversely, how can you prevent the user from
opening one of the potential vulnerabilities of X? Have you not heard,
"... I didn't change anything, but it doesn't work."? Hell, users
won't be happy until an OS can understand their intentions. Unix will
_not_ be on the desktop until users are capable of understanding Unix.
Sorry.
>>Rather, Unix has been about more serious, working systems.
>
>I do not believe that working -- seriously well -- precludes user friendliness.
Bull. If you have 10 choices, you only have 10*10 things to learn. If
you have 1000 choices... Those of us who have come to rely on Unix
have generally had a problem to solve that couldn't be solved
otherwise and had a clear enough vision of the result to wade through
the complexities. Do you seriously believe the average "user" out
there cares enough to do the same?
If you reduce the choices (aka user friendly) available in Unix, you
reduce Unix (which you can certainly do;) But why? Why is it so
essential that the casual user enjoy greater reliability when he surfs
tits.com or turns out more "good enough" grocery store style
publishing? What is gained by this "market" share?
>>I run into
>>a large number of old farts in the Unix world who are still bitter
>>about Sun abandoning SunOS in favor of SYSVR4. Most have been in the
>>trenches and are delighted to know that BSD is still alive and well --
>>most of them aren't too thrilled about Linux. Most of them don't deal
>>with toy systems, rather 24/7 mission critical systems and have a
>>tendency to judge quality systems by how often and when their pagers
>>go off. I agree with them.
>
>So do I, in fact.
Then how can you have it both ways? The development effort for this
mythical user friendly GUI is many times the effort required to make a
stable production system and, in my opinion, causes the same problems
exhibited by M$. If they have failed to create a stable system with
all their millions and snickers, what makes you think volunteers and
enthusiasts can improve on that model?
>>The development model that appears to stiffle "creativity" and the
>>"sociology" -- both of which are null terms, as far as I can tell, is
>>precisely the model under which professional software is developed.
>
>The "sociology" to which ESR refers is not limited to the development
>model but embraces both the developers and the user community. In fact,
>it has MORE to do with marketing and advocacy than with development.
>There is no one model under which "professional software" is
>developed.
Probably true, but I haven't seen the Linux model used in the real
world, yet.
Again, I'm sorry if I'm jaded, but "advocacy" and "marketing" has
meant, in the past, that I enjoy the unique opportunity of following
behind to make the marketing lies come true. Brett, I'm sorry if my
attitudes aren't very appealing, but my attitudes are more prevelant
than you may realize. I don't like having to do the clean up, nor do
many others. The reason that you are seeing resistance can be traced
back to variations on that same attitude.
>>It's the model that makes all of the *BSDs a superior choice to the
>>haphazard phenomenon lumped under the umbrella of "Linux".
>
>One cannot infer cause and effect here. I am inclined to believe that
>it is 20 years of history -- including the development of consistent
>conventions and lots of tried and true, time tested code -- that gives
>BSD UNIX the edge. And the edge may go away if there aren't lots of
>enthusiastic eyeballs poring over the code and making improvements.
Again, I disagree. It may be the difference of the worlds in which we
each work, but from what I see, when someone invests a $1/4M in a
system, the cost of the OS is pocket change. The criteria change. The
basis for decision changes. Slick desktops don't amount to a tinker's
damn.
Think outside the box for a moment. Intel hardware is beginning to
grow up. Major manufacturers are beginning to look at the higher end
market, mission critical servers and data warehouses. Trust me when I
tell you that they are paying attention to the free OSs and seeing
them as a potential means of offloading development and maintenance
costs. At the same time, they are furiously trying to learn how they
can leverage the free OS hysteria (to which Linux has greatly
contributed) into more profitable hardware sales.
Seasoned admins are disenchanted with Linux, are looking for stable
platforms and will go with which ever one won't get them out of bed at
2AM. They don't care whether it's free or not. The *BSDs are growing
in the quiet corners. From my perspective, I would rather appeal to
the professional admins than the Maudie fricks.
[snip]
>The BSDs' market share is currently shrinking, and its mindshare (while
>I and others have worked to boost it) is not keeping pace with that of
>Linux. Again, it is being squeezed out of its ecological niche. "Patience"
>will only result in a continuation of this trend.
So what? Let Linux go for the desktop. Let them fail as M$ has done;)
Why must we be Linux?
>>Most professionals are turned away by evangelical zeal
>
>I disagree with this opinion. In my experience, the professionals with
>whom I deal will only consider a new operating environment if many people
>with whom they come in contact STRONGLY recommend it. The BSDs' lack of
>strident evangelism makes it less likely that this will occur. On the
>other hand, Linux owes its runaway success to it.
You deal with different people, then. Most of the people with whom I
deal are more motivated by what allows a good night's sleep. They
don't give a damn whether it's Solaris, AIX, HPUX or any other OS as
long as they don't get paged in the middle of the night or come in to
find they've been cracked. _Now_, if they can find that reliability
_with_ source and decent support, they pay attention.
They've been disappointed with the Linux hoopla and are quitely
turning back to the roots. The SunOS admins are beginning to come home
now that the true BSD is back. True, some are going to one of the
other BSD's, because FreeBSD doesn't run on their hardware, but that's
not a bad thing. Please don't screw the pooch by turning BSD into an
evangelical crusade that makes us look like Linux.
>>-- they've seen
>>too much in the past, they won't buy it now. Surprisingly, they are
>>also skeptical of the availability of source -- they perceive it as an
>>invitation to intrusion as has been demonstrated so many times in the
>>past with Linux systems.
>
>Because the availability of source is common to both the BSDs and Linux,
>this is not a factor that would lead one to choose one over the other.
Of course not -- it leads them to reject both because of the bad
experiences they've all heard from Linux.
>However, open source is making a good name for itself in the area of
>security; on balance, administrators are coming to see it as a "pro"
>rather than a "con."
Not in my world. Why not talk to the people who put 16 hours a day
maintaining mission critical systems and then get paged in the middle
of the night? They don't care nearly so much about open source as they
do about reliability and security. If they can get that with open
source, more the merrier, but most of the box office has shown that
the stellar open source system has been the target of the most
successful attacks. They don't put their butts on the line, no matter
how many evangelists there are.
[snip]
>>So the real queston is this: it's a long, slow process converting
>>professonals,
>
>True. (Though that's not a question. ;-)
Yeah, OK, your right -- I've been dealing with users too much;)
>>and zealotry won't do it -- quality will.
>
>Advocacy ("zealotry" is your word, not mine) is also essential. Again,
>in my experience, repeated word-of-mouth recommendation -- STRONG
>recommendation, not a wimpy "oh, yeah, it's the other free UNIX, you might
>consider it too" -- is what sways professionals. So do other factors,
>such as marketing, certification, support after the sale, etc.
Brett, I'm sorry, I don't see that working in my world. 90% of the
decisions boil down to what hardware and applications are
supported. Management goes for price and sizzle, staff goes for
whatever means a warm supper and a good nights sleep. If you really
want to promote BSD, lobby the ISVs to deliver working _mission_
applications for FreeBSD and cede the desktop to whomever is foolish
enough to chase that chimera.
>> To whom would
>>you rather appeal; the professionals or the desktop crowd?
>
>Both. To frame it as "one or the other" is a false dilemma.
Excuse me? Not in my world. Where I work, there is a _strong_
distinction between the two. Most of the large manufacturers with whom
I have contact view Linux as a passing fad and one they can leverage
to generate more hardware sales and ultimately pull the IT
infrastructure into their own OS and support. The real OS dollars are
in support;) They don't really give a damn about the desktop. They're
more interested in infrastructure and big iron.
If we squawk like penguins and other assorted fowl, we will be
considered in the same light and ultimately do more damage than good.
If, on the other hand, we maintain a professional demeanor and
product, we will continue to make inroads in the infrastructure and
ultimately gain the respect and support of not only IT staff and
management, but the manufacturers as well.
IMHO, chasing the Linux hysteria is a very bad thing to do and a waste
of the skill and effort that brought the BSDs to their current
quality. Besides, the "big" names have used BSD code for years and
will continue to do so. It's the BSD copyright that rolls by on
commercial OS installs -- not Linux.
-- Jay
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.9908312034570.1356-100000>