Annoucning DragonFly BSD!

[Matthew Dillon]

:>   anyway, not our Latin alphabet ] effort, a dubious idea to divide
:>   the number of shoulders that load sits on.  There's already another
:>   cross platform ports project anyway (Freshports?)
:> - A new distribution mechanism (whatever) ? maybe - but again
:>   if better, that technology should be adopted & merged into other BSDs.
:> 
:> http://www.dragonflybsd.org/ may be a just a troll erection, it's
:> constructed so there's nothing real to see.  A troll site ?  No
:> where to click & sample code inside browser, you'd have to cvsup &
:> extract localy to check real code. No interest until others confirm real.
:
:you missed the entire source tree? look again..
:and I doubt that a troller would have redesigned the entire kernel
:to make a troll and made it work.. if he did we should invite him in..

    Yes, that would be some trick, considering that the unified diff
    between my tree and -stable is over 347,000 lines long!  Sheesh, I
    guess I really *do* have to get cvsweb up and running for people
    to believe it, ftp and cvsup apparently aren't enough!

:
:it's himm.. believe it..

    I don't understand, do some people not believe that I am heavy-weight
    kernel programmer? <GRIN>  I mean, sheesh, this reminds me of my old
    Commodore PET days, when I wrote a centipede game entirely in 6502 machine
    language and submitted it to cursor magazine for publication.
    They declined, I think because they didn't quite believe that a 14
    year old kid could *do* that.  It was a damn fine game, too, the last
    level featured an invisible centipede who only turned visible for a
    few seconds when you hit one of his segments.

:> ( Julian Stacey <jhs at berklix.org> )
:> The logo is useless (& a troll give away ?):

    Useless!  You try staring a three inch long DragonFly in the face for
    half an hour!  It was fate is what it was, that Fred was so photogenic
    because it took about 20 shots before I got him framed and focused 
    properly and he basically refused to budge despite my comings and goings,
    only occassionally startling, flitting around the yard a bit, and then
    landing right smack back on the same frond he had just taken off from.

:> There's too many BSD's already.  More complete BSDs aren't of
:> personal or business benefit.  More kernels, tools, & experiments
:> in ports/packaging etc could be useful though, but to be of most
:> benefit such work should be fully integratable, & not further split
:> the available BSD workforce.
:> Julian Stacey       Freelance Systems Engineer, Unix & Net Consultant, Munich.
:
:Well if you take away his commit bit treeat him unfairly, what other
:choice does he have? 

    Well, I don't really care about that, but this points to an interesting
    dichotomy in the perception of people who use open source and of people
    who write it.  I don't know about other open source programmers but my
    motivation is interest and invention.  It has nothing at all to do with
    towing some imaginary line.  Why should it matter what operating system
    base I choose?  If Linus felt that way he would never have started Linux.
    It is a concept that non-programmers like to banter about on forums like
    slashdot but it is utterly meaningless to most of the people that do
    the actual programming.  There is responsibility, yes, but it is an
    effect rather then a cause. 

    History is filled with underdogs winning against the behemoths against
    all apparent odds, and turning into behemoths themselves only to be
    displaced by the next underdog when their little clique starts believing
    in its own immortality.   As a programmer who has gone through several
    generations of operating environments I don't believe in the immortality
    of anything, least of all FreeBSD or Linux, or my own code.  But it
    doesn't stop me from working my favorite project on my favorite platform,
    whatever that happens to be.  Ultimately the only thing that survives
    history is the invention and the concept, and memory.  If people can see
    that a concept works and go and implement it in their own favorite
    environment then that counts as a success and another notch on my
    sleave regardless of anything else.  If people can make positive use
    from something I've done, that's a nother notch.  It's amazing to me
    how people can belittle the work that Rik has done on the Linux VM
    system, for example, under the misconception that not having outright
    adoption means that it was somehow a failure.  How absurd!  That work
    created a competitive environment which had the direct result of several
    people building upon the concepts and implementating something far better
    then what used to be there.  That's a notch in Rik's sleave, and in mine
    too for having been able to contribute to the discussion.

    It's amazing to me how many old Amiga users have emailed me in the last
    two days about DICE.  DICE is a C compiler I wrote for the Amiga.  In
    modern day terms it is consigned to history now, having outlived the
    platform it was originally designed for (the Amiga, though there are
    people who still use it to support legacy 68000 based hardware).  What
    is amazing to me are the stories from people who got their start in
    programming using that compiler, and have gone on to great jobs and
    interesting work in later years.  *THAT* is what I care about.  I don't
    care a whit about what happens to the actual code because I know it
    won't last.  I've written a huge amount of code in my life and 95% of
    it is no longer being actively used.  That same 95%, however, has
    effected the lives of thousands of people in a positive fashion so the
    idea that code must somehow become immortal or its a wasted effort is
    just absurd.

    Open source is the ultimate expression of Darwinism, and evolution takes
    many forms, but one thing is for certain:  There is no such thing as
    immortality for Linux, FreeBSD, or anything else for that matter.

					-Matt
					Matthew Dillon 
					<dillon at backplane.com>