Older blog entries for uriel (starting at number 5)

I'm a bit annoyed that after more than a couple of months in advogato and getting certified by a quite a few people, I still can't even post comments to existing articles, which was one of my main reasons to register...

I think I wanted to post some comment to mako's article about CC(to which incidentally I made some very small contributions in the hours prior to it's publication, mostly suggesting bits that could be deleted ;)), but by now, I have forgotten what I wanted to say, except probably that mako toned it down too much ;)

Oh, and neither can I 'register' with various of the projects I'm involved(eg., wmii)... oh well...

Ok, now I should stop whining and go hack wikifs... oh, and I had an idea for a cross between wiki, blog and irc that I plan to write for Inferno. If only days had 32 hours...

27 Aug 2005 (updated 30 Aug 2005 at 20:14 UTC) »

Just found that the 9P/Styx implementation in Java is alive and well, the other day I was looking for it to add the link to the v9fs homepage but could not find it. It recently moved to a new site: http://jstyx.sourceforge.net/

I don't think I have any use for it(I rather don't touch Java if I can help it), but it's nice to see more 9P implementations out there. And now that I think of it, maybe I could use it to interact with some legacy Java code we have at work from our new systems in Python and Limbo.

So far we have 9P implementations in C, Limbo, Python and Java, not bad.

I had been toying with the idea of doing a Ruby implementation, mostly for fun and to learn the language better, but I don't like Ruby, so I'm thinking Erlang would be a much better idea, as caerwyn said the other day in #inferno, what other good concurrent languages are out there? aside from Limbo and Erlang? I know none, of course there is also libthread.

In related news, at last I convinced garbeam to make wmii implement proper 9P(oh, yet another implementation, and this one under a BSD license, yay!), and the conversion is almost done, this will allow to make our filesystem interfaces much saner.

I hacked the wmii configuration/init system and worked out a simplification that got rid of that wmira kludge and lets us use Plan 9 from User Space tools without polluting the PATH. Oh and of course it's also got rid of a few hundred lines of convoluted crap. As God Ken once said:

One of my most productive days was throwing away 1000 lines of code. -- Ken Thompson

When I hack wmii, usually that is the kind of work I do, fortunately garbeam can appreciate it; that can't be said of the developers of many other projects, who are obsessed with adding more and more code rather than step back and do things right.

19 Aug 2005 (updated 19 Aug 2005 at 19:02 UTC) »
Rumors of the Death of Plan 9 Greatly Exaggerated

So slashdot, always bastion of reliable news reporting, had to bring the news: "Bell Labs Unix Group Disbanded", from this others went on to deduce that Plan 9 development has stopped.

Nothing further from the truth, not only is Plan 9 development at Bell Labs ongoing but development outside the Labs is as active as ever(and expect more news on this front soon).

Sape Mullender wrote to 9fans:

1127 ceased to exist some time ago.  The department was renamed and got a
much longer number which, of course, I can no longer remember.  The current
split of what's left of 1127 is not considered to be a big deal by most
members of the department.  There have always been many interdepartmental
collaborations, so 1127 has always been more of a management entity than a
group/club/fraternity.

[...]

Plan 9 is alive and reasonably well as our current vehicle for wireless systems research. Plan 9 has gone embedded (but rest assured, all our wireless software runs in user space, using Plan 9's real-time scheduler to make sure it runs when it needs to.

Using Plan 9, wireless code becomes useful in a fraction of the time that code can be made to run on all those commercial embedded operating systems (which I do not care to mention) and also much quicker than on Linux which I am sorry to say is just too large and illegible.

May the research go on!

And don't let the "Fourth Edition" release date fool you, Plan 9 ISOs are built from the latest code every night, and the current codebase is very far from what Fourth Edition was, it's just that nobody has been bothered to make a new "official" release. Plan 9, as usual, is more about code and quality than about hype and buzzwords.

15 Aug 2005 (updated 15 Aug 2005 at 01:29 UTC) »

Back from WTH, home at last, more about my wandering around Europe later.

Interesting papers from the nice guys at lsub:

And a few more in their papers page.

I'm not sure if Omero will replace Acme as my favorite development environment, but it surely looks very intriguing.

There are rumors that a 9con might be organized this fall in Madrid, it would be great to hang out with all the Spanish 9fans.

2 Aug 2005 (updated 2 Aug 2005 at 10:36 UTC) »

Charles Forsyth from Vita Nuova wrote a great paper: The Ubiquitous File Server in Plan 9.

I'm also trying to get 9grid back on track and setup the Kencc web site.

Still in transit returning from WTH, more when I'm back home.

26 Jun 2005 (updated 1 Aug 2005 at 18:42 UTC) »

Been very busy lately with work, but v9fs is getting good feedback and might go soon into the mainline kernel.

I hope dho starts working on a *BSD implementation of 9P soon, maybe Leimy might want to help him with that?

The latest wmii-2 snaps are starting to work very well, and wmii is close to become the 'least sucking X window manager'!

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!