fozzilinymoo.org

Samba: Integrating Linux and NT Environments

Jeremy Allison of the Samba Team at Silicon Graphics led this session describing Samba v2.0 and its new security=domain configuration option.

This was a really good session as Allison is a core programmer on Samba and explained how the SMB protocol was reverse engineered from Microsoft. He also pointed out many fundamental flaws in Microsoft's designs and how difficult it was to figure out how Microsoft did things because of the lack of documentation.

Simon Horman, while demonstrating that he had a decent understanding of CVS and how to use it in his own environment to do version control of system configuration files, did a TERRIBLE job of putting on a presentation.

Dirk Hohndel of S.u.S.E. and the XFree86 development team had a packed audience for this session. The room was filled to at least twice the posted capacity (156 people). Apparently lots of people are interested in the direction the X Windows System under Linux is going.

Some very interesting news items about XFree86:

• Hohndel predicts XFree86 v4.0 will be out within four months.
• v4.0 will have multi-head support for a virtually unlimited number of monitors. This multi-head support also will not be limited (like Windows 98 is) to using the same video adapters and monitors for each head.
• v4.0 will be modularized. There will be one server binary (unlike v3.x and previous versions) and drivers for specific hardware will be loaded as modules. This will make for small footprint X configurations if needed.
• Server modules will be platform independent. X server binaries will be OS and platform dependent and the loadable modules will not.
• Support for older cards (i.e. Genoa, Paradise, RealTek) will not be supported by v4.0. Why? Because it's not worth the time to convert the drivers over to 4.0. Nobody uses these chipsets anymore!
• While Hohndel refrained from indicating a preferred graphics card, he mentioned Matrox G200 enough that it is obvious it's well liked by the development folks.

Subject-Specific Searching with Harvest

Brian Hancock of Wichita State University gave a presentation of the Harvest system. Hancock admitted right away that he was not a technical guy, but rather a humanities librarian at the university. It showed.

It was rather amusing to watch and listen as Hancock told us how to configure and compile the Harvest system (which consisted of the regular 'run ./configure and then type make' and so forth).

Nevertheless, the Harvest system seems interesting.

Panel Discussion: The Continuing Revolution

Eric Raymond (esr) moderated a panel of famous open sourcers in the big exhibit hall. The panel included Richard Stallman, Larry Wall, Linus Torvalds, Guido van Rossum.

This time, Christian and I were first in line at the doors. I managed to steal a seat in the press/VIP section and got video footage of the entire session from right in front of the stage. I was also sitting right next to Chip Salzenberg when he rose to erupt about Microsoft's briberous ploys to swindle the US education industry into using its products. I got close ups!

Dinner: Ravioli's

For dinner, I walked over to a restaurant near our hotel called Ravioli's which, not surprisingly, serves italian food. It is more of an italian fast-food. I ordered a large combo of spinach ravioli which included breadsticks and a salad. I took it back to the hotel room and ate while I worked on my laptop.

We pretty much just hung out in the room the rest of the evening, installing Linux packages, trying out new window managers for X, watching Star Trek, and talking philosophy.