Subject: Re: "Well, I want to switch over to replace EMACS LISP with Guile."
From: Erik Naggum <erik@naggum.no>
Date: 16 Oct 2002 00:12:58 +0000
Newsgroups: comp.lang.lisp
Message-ID: <3243715978755380@naggum.no>

* Nathan Whitehead
| The whole rationale of forking is that the owner (or leader, or manager)
| disagrees that your idea is the way the product should evolve.

  Quite so.

| If I have a new idea for Windows and Microsoft doesn't like the idea,
| what am I to do?

  The same thing you do if you have a better idea for anything else in any
  other industry: Decide if you think the market for your improvement may
  be big enough that it is worth your time to build a competitor.  People
  have actually done this very frequently in every other industry.  I have
  no idea why people consider software so hard that they could not develop
  something better with their own money and source base.  Perhaps the lack
  of working intellectual property mechanisms (e.g., patents, which are
  horribly unfit in their current application) in the software industry
  makes it so much harder to build on the past, but I fear that the real
  reason is that we all know how shoddy most software products are and how
  little /real/ value is produced in software development.  Microsoft's
  crap would never have gotten off the ground in a market where the
  customers were competent to judge the quality of their products.  It was
  all about enticing ignorant people in "future" gains.  The belief in the
  rosier future confused the financial markets for a long time, too.  I
  marvel at the sheer evil of the manipulativeness and propaganda skills of
  those who could actually lie to so many investors about the ability to
  make money with the investments they sought.  Microsoft has been built on
  deception and fraud from the get-go.  Their incompetence and the amazing
  manipulativeness of the marketing nonsense around their inability to get
  a working version of Windows on the market when promised should have been
  enough for anyone with half a brain to realize that this swindling outfit
  was slated for DoJ investigations if not criminal proceedings not too far
  down the line, but, no, people bought MSFT shares and their products.

  What /appears/ to be true, therefore, is that if you have a better idea
  for the software, that is wholly irrelevant with such a fraudulent gang
  of incompetents as "market leader" -- what matters in software is how
  many people you can con for how long, or so people think.  This may not
  even be possible to change with a really much better software idea, but
  mostly because Microsoft has convinced so many people that to produce
  even their crapware, you need /many/ billions of dollars and nothing
  short of hundreds of thousands of man-years.  The Open Source world has
  not disproved this point at all: counting the amount of effort that goes
  into the Linux kernel and the GNU software suite alone is intractable.

  So if you sit there with an idea you think you could realize with a few
  months' worth of effort at most, there is nothing to build on that would
  make it possible to do this without practically infinite amounts of
  money.  But this belief is actually very, very wrong.  It is perhaps the
  single most successful deception that Microsoft has foisted upon us:
  Software is so hard to get right that even they cannot do it.  This is
  the one point where the Common Lisp community knows better, but how can
  we tell people this when they are deluged with viruses and their source
  of trust, the very same Microsoft, says that they cannot fix the problems
  and that malevolent people take advantage of their desire to help users
  work smarter and more efficiently?  (The latter is also wrong.  Very few
  administrative tasks in the running of any business costs less than they
  did two decades ago.)

  The question I ask is therefore: Why do you think you could succeed any
  better if you had the source code?

| If I have a new idea for Linux and Linus doesn't like it, I can fork the
| linux kernel and still do it.  You are right, maintenance, bookkeeping,
| finding users, etc. make this option almost impossible a lot of the time.

  Some of the Linux distributions have used deviant Linux kernels and some
  weird versions of GNU utilities, such as the GNU C compiler.  This is a
  source of nothing but trouble and much hostility in the Linux community.
  It is partly because of the idiot waste of effort and lack of ability to
  cooperate in the Linux community that I consider free software less able
  to succeed in its stated goal than commercial software development.

  The important thing to remember is that there is more commercial software
  than Microsoft.  Being better than Microsoft in software quality is easy.
  Being better than any of the other small companies that are better than
  Microsoft is not that easy.  Microsoft holds the whole industry hostage
  with their own incompetence and the almost religious faith in Microsoft's
  ability to cater to market needs has made it very, very hard to succeed
  for the *huge* number of small companies who can do better than them.
  Had Microsoft not been so incompetent at software development, their
  competitors would have been fewer, stronger, and better, and thus more
  able to succeed in competing with them.  It is not the first time in our
  modern industrial history that low quality and good marketing has been a
  strong winning bet.  The only solution to this problem is good education
  and people who have /learned/ to think long-term.  Since our business
  schools are not exactly churning out that kind of leaders at this time,
  we are not likely to come out of the "IT winter" any time soon, but my
  bet is that some small company that just does /much/ better than the
  idiot behemoth will take the world by storm.

-- 
Erik Naggum, Oslo, Norway

Act from reason, and failure makes you rethink and study harder.
Act from faith, and failure makes you blame someone and push harder.