after Linux, what? in place of Hurd, Eros, Brazil,...?

Toomas Karmo verbum-qazKcTl6WRFWk0Htik3J/w at public.gmane.org
Wed Oct 29 21:46:01 UTC 2003


Thanks to Christopher Browne (http://cbbrowne.com/info/)
for opening up an important topic, making it clear that I had
overstated the merits of Hurd:

((QUOTE)) 
The problem is that Hurd is tied to Mach, which never really got
properly finished, and which is tied to architectures that are rapidly
getting obsolete.  Hurd only supports filesystems up to 1GB in size, to
name the most problematic antifeature.  I can readily have more RAM than
that on hardware that is relatively pedestrian.  And the hardware I
_want_ supports on the order of 16GB.

I can't see Hurd returning to interest without them implementing a fresh
microkernel from scratch, which amounts to going pretty close to
starting over, and at that point, it's likely sensible to call it
something fresh.

Of course, by that point, it might be as sensible for (say) DragonFly
BSD (a project to build a highly threaded, message passing kernel) to be
run with a Debian or BSD Ports "user space" and have the improvements
we'll be looking for...
((/QUOTE)) 

I hereby recant. It's now NOT clear that Hurd is the Next Great Thing. 

At this point
I'd like to ask whether anyone can peer into a crystal ball and see
what the Next Great Thing might be. Rob Brockway has referred in passing
to Eros, Brazil, and Plan9 as interesting operating systems. Do any of
these have passionate backers? Is anyone  on the listserv
keen to wade in and DEFEND Hurd? 

My motive for stirring up this hornet's nest is to improve some writing
that I have been doing on the world as it is likely to be in 2184, when
fossil fuels are exhausted and society is recovering from a
general infrastructure
collapse. I paint a picture of a culture that scrapes together
enough energy from wind turbines to run light railways, and also to
implement
**SOME** of what Jeremy Rifkin has optimistically hailed as the 
impending "hydrogen economy". In
the course of this writing (in that part of my
narrative which depicts  
a stroll through far-future rural Kent, in the
company of Saint Thomas More), I have the following historical
flashback:

((HISTORICAL_FLASHBACK__IN_CHAPTER_2_OF_"UTOPIA 2184"))

Commons and enclosure. 

A grey afternoon in England. The farmer, his goodwife, their four
children, their diminutive pony-cart heaped with such chattels as an
agricultural family may command in 1520. The better chattels - the pots
and fire-irons and bedstead, and of course the pony and cart - they will
sell at auction in the town. The pouch of shillings thus realized will
suffice to buy lodging and food for some few weeks. After? God will
provide, or not. Behind them, workmen from the manor house are already
ripping thatch from gable. My lord's fence shall run here, says the
shire-reeve, and here, and here, 

Commons and enclosure. 

Linux, though rather ad hoc, met a need. In the 1980s, Unix was an
enclosure, propertized. Richard Stallman in Massachusetts thought he saw
limitations in proprietary software. He and likeminded programmers
founded the "Open Source" movement, donating to the emerging
cybercommons first this tool, then that: they developed a text editor,
an ANSI C compiler, a command-line interpreter, another text editor, ...
GNU, they called their initiative, for 'GNUS's not Unix.' Witty, like
that learned footnote ever so meticulously explaining that real number e
is permitted, yet not required, to be the base of the natural
logarithms. Unpack the recursion, and you get 'GNU's not Unix's not
Unix', and then 'GNUS's not Unix's not Unix's not Unix', ad infinitum.

What was lacking (or, rather, what never seemed to get finished) was a
non-propertized kernel. A kernel: core software bringing order and good
government to an entire workstation or server. The kernel is the
software that coordinates. 

At the University of Helsinki, Linus Torvalds built a rudimentary
kernel. Version 1.0 was completed in 1994. He named his creation Linux.
Linux rocked, ruled. The suits liked it, started companies with names
like Red Hat, sought to make money as consultants even while keeping the
source code open, visible in the commons, open to continual peer review.
A step, then, in the right direction. Of course people in the know
sniffed at Red Hat, installed Debian GNU/Linux instead. Debian GNU/Linux
was the Linux distro with the public charter formally forswearing
commercial ambition. And the distro with the best formalism for finding,
installing, monitoring, uninstalling software packages. 

Work on an alternative to Linux continued under the rubric 'Debian
GNU/Hurd'. 'Hurd', it was explained in the commons, was short for 'Hird
of Unix-Replacing Daemons', and 'Hird' short for 'Hurd of Interfaces
Representing Depth.' 'Hird of Unix-Replacing Daemons' thus itself
expanded to 'Hurd of Interfaces Representing Depth of Unix-Replacing
Daemons', which itself expanded to 'Hird of Unix-Replacing Daemons of
Interfaces Representing Depth of Unix-Replacing Daemons', ad infinitum.
Witty. 

The wider significance of the 1990s cybercommons did not escape the
notice of social commentator Jeremy Rifkin. Here, he wrote, was a model
for the creation of a new energy- distribution infrastructure. He
published The Hydrogen Economy: The Creation of the World-Wide Energy
Web and the Redistribution of Power on Earth at J.P. Tarcher/Putnam on
2002 September 12. Rifkin's work was rightly criticized as erring on the
side of optimism. Nevertheless, it pointed a way forward, a tenable line
of development, tenable even for wretched peoples reduced to hammering
out the more basic turbine parts with hand tools. 
((/HISTORICAL_FLASHBACK__IN_CHAPTER_2_OF_"UTOPIA 2184"))

Christopher Browne's comments suggest that this flashback will have to
be edited or rewritten, perhaps by simply cutting out the 
discussion of Hurd. It would be nice to hear from anyone with
suggestions on appropriate points to make in a rewrite. 

If you need wider context, you can find 
"Utopia 2184" in its entirety in the "Literary" section of
http://www.metascientia.com.



Rapidly, 


Tom = Tom Karmo
http://www.metascientia.com 
--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list