Re: tlug: Re: Many Faces on Linux

["Stephen J. Turnbull" <turnbull@example.com>]

>>>>> "ash" == Andrew S Howell <andy@example.com> writes:

>>>>> "Jonathan" == Jonathan Byrne <- 3Web <jq@example.com>> writes:
    Jonathan> On Fri, 25 Sep 1998, Andrew S. Howell wrote:
    >>> After your and Jonathan's post, I realized that I have be
    >>> living in a (UNIX) shell too long. The shareware movement
    >>> seems to be mostly PC based, or rather most (all?) of what I
    >>> use is not shareware.

    Jonathan> If people could write open source shareware and make
    Jonathan> money on it while providing source to registered users
    Jonathan> (at least, and maybe to everyone), the ability to make

Uhhhhhm.  What do you guys think Ghostscript is?  Chopped liver?

I don't know that "shareware" is going to make it on Linux.  But I
think we will start seeing a lot of commercially restricted open
source licenses.  I can see a market for shareware to some extent, but 
I think shareware is probably on it's way out.

Shareware is good approach for single-application utilities.  But
Linux makes it just too easy to emulate the basic capabilities, and
the really basic stuff goes free too fast.  Look what happened to
pkzip, for example.  There's never going to be a shareware compressor
on Un*x.

A shareware window manager or spreadsheet?  I doubt it.  I think the
Ghostscript / Cygnus model is the one to look for.  Give away
crippleware or dinoware, and restrict distribution of the latest and
greatest, or consult for new features and support.  You can't support
that kind of development for dribs and drabs of shareware registration
fees; you really need contracts for sizable amounts of money.  Or you
can get people to do it for love.  But they won't work on junky stuff.

    ash> What happens when I come up with shareware and charge x for
    ash> it? Will other developers continue to work for free while I
    ash> rake in the big bucks :)

Yes and no.  Aladdin gets a lot of fixes from beta testers, and of
course almost all the device drivers are ultimately user-contributed.

However the core stuff, such as language standard compliance and the
recent development of a whole new language front end (for PCL/HPGL),
is done by paid staff at Aladdin, funded by contracts from users who
want specific features.  Many of the users are historically open
source supporters (the NSF, for one), at least if you look through the
looking glass at a different field (scientific research).

Shareware isn't big bucks, anyway.  Not even for Phil Katz.

    ash> Though linux is very good, the one thing that I wish the
    ash> linux community would do is not be so linux centric. That is,

Don't we all.

    ash> I would like to be able to run some open source stuff on
    ash> commercial UNIX systems. As I write this, I can't think of a

Emacs, Ghostscript, gcc, Perl, python, Apache, X11, TeX, uhmmm, what's
the problem?

Yeah, there's lots of neat stuff written for Linux that doesn't port.
What's amazing is not that it doesn't port to Solaris, what's amazing
is that even with hardware dependencies, typically you can run it on
any CPU that Linux supports as long as the hardware needed is either
PCI or SCSI.  Wasn't so long ago that half the stuff written for DOS
you had to have a GUS for sound and a Mach32 (no Mach8s or Mach64s,
definitely no Neomagic need apply....) or it fried your CD-ROM as it
took the system down, and certainly didn't work.

Glass is half-full, IMHO.

True, Linux isn't yet generating a lot of port-able stuff.  But I
think we'll see it happening as the momentum shifts from the FSF
"we do FreeSoft right" crowd to the "oh, gomen ne, it never occured to 
me you might want to run this on *BSD" crowd.

    ash> good example, but I know in the past I have found stuff that
    ash> is so full of "linuxisms", that it is difficult to port.

cthugha is one, I still haven't figured out how to get it to talk to
NAS ;-)  Enlightenment is another, more important example.  Or maybe it
was windowmaker.

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences        Tel/fax: +1 (298) 53-5091
---------------------------------------------------------------
Next Meeting: 10 October, 12:30 Tokyo Station Yaesu central gate
Next Nomikai: 20 November, 19:30  Tengu TokyoEkiMae 03-3275-3691
---------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp