Too me, the Singularity stuff is kind of an easy way out. Its like an
excuse not to do anything because "the Singularity will do it". Whether
Singularity happens or not, we should act as though Singularity will never
happen, if only to get things done. I personally am doubtful about
Singularity anyway, so it doesn't really impact my predictions or decisions.
>As for work, I am curious about how many subscribers to this list
>are involved in any of the various free software projects such as
>GNU, Linux, or the Debian distribution packaging project. Lately
>I have come to a realization that the success of Unix over other
>programming environments (and here I'm talking about success with
>programmers, not success with users) is a consequence of the fact
>that it was designed to evolve in a way that constructivist-rationalist
>operating systems such as Windows NT were not. I would not be at
>all surprised if the future evolution of free Unixen is intricately
>intertwined with other future technological developments, even
>though I know all about the shortcomings of Unix (it's not all that
>hot at realtime stuff, the main process forking mechanism is
>inefficient, the security model puts too much demand on skilled
>system administration, etc), but this is where my own interest has
>been taking me lately. I am particularly interested in the various
>attempts to transmogrify Unix into a distributed operating system
>(such as Plan 9, its offspring, and Gnu's HURD).
>
>What Debian is doing for the evolution of computer software seems to
>me to be something roughly analogous to Cairns-Smith's hypothesized
>cutover from clay crystals to DNA molecules. The Debian project is
>coming up with a standardized system of interchange of "genetic"
>(memetic, really) information between computer systems, so that
>operating-system modules can be exchanged and improved fairly easily.
>Unfortunately, intellectual property restrictions impede this process,
>but fortunately, intellectual property restrictions are strictly
>optional.
>
>Opinions? Comments?
>
I've been working on free Linux software off and on. I think the free and
open system encourages the rapid evolution of the software and OS. And
quite frankly, it runs rings around my Sparcstation, in both speed and
flexibility. In my case, I have been working on a free, large-scale
relational database system for Linux. Something equivalent in both power
and flexibility to Oracle Workgroup Server 7.3, while being more closely
integrated with Internet technologies than the Oracle servers. I have
looked at extending the Linux OS to something of a native distributed
request broker architecture, mostly for my own amusement.
I cannot wait for the day when free Unix goes head to head with The Big Guys.
But this is getting a little off-topic...
-James Rogers
jamesr@best.com