Robert Pogson

Squeeze is firming up quite nicely. I noticed today that LibreOffice3.3 is now in Debian Squeeze. I used the download from www.LibreOffice.org and encountered some instability. I trust that will be fixed in the release.

The bug count is 82 now. Not bad for 25000 packages on 12 architectures. The release is imminent except that there is a steady flow of bug reports and fixes still. I don’t see any show-stoppers at the moment. I have only one package not working and I can live with it.

UPDATE Release Candidate 1 of the Debian Installer was released on January 13, 2011. Give it a trial.

I downloaded and installed VirtualBox and Netinst-rc-1 and everything worked. The installer assumes more, though. I asked for “install” and the thing installed Virtual-box-ose-guest stuff and X etc. without being further prompted. That gives me less control or wastes time. I usually like to do a minimal installation and add what I want… I still had the option to unselect “graphical desktop”. I kept “standard system utilities” I ended up with 320 packages installed in 720 MB of storage. I apt-get install xfce4 openoffice.org chromium-browser and added another 292 packages and 322 MB more downloading and a further 806MB storage used.

Based on this minimal test, I would say that the Debian Installer is quite usable. Let us see what others find on their hardware. It works for me.

UPDATE Today’s Debian News estimates, ” roughly speaking, about 20 release-critical bugs remain to be solved for the release to happen.”

It should be an interesting weekend.

Everyone makes them. That is why we have syntax checkers, debuggers, and alpha and beta releases. It took me quite a while in the beginning of my career before I could produce a page of good code in the first draft. Every now and then I get lucky again on some little project and a smile results.

eWeek has produced a list of some disasters but conspicuously absent from the list are disasters from M$:

• endless vulnerabilities in that other OS causing $billions of failure and workload for IT
• releasing WP7 without cut-and-paste
• thousands of millions of re-re-reboots because M$ cannot keep the registry or anything else straight
• hundreds of millions of instances of slowing down…
• having little more than vapourware teed up for ARM
• latest release not having drivers for tons of widely used hardware
• needing 1 gB of RAM to do e-mail
• taking more than a few seconds to get a working desktop on machines that GNU/Linux can boot several times faster
• emphasizing quantity over quality of code
• making code complex to mess with competition
• failing to implement ISO standard M$ pushed through
• more because the bosses know little or nothing about computer science

The FLOSS communities had some failures too:

• lots of black screens resulting from changes to Linux and X
• major regressions and vulnerabilities

Of course, FLOSS had some major successes and most of the failures were related to innovation. You don’t cook eggs without breaking a few. Some of the failures have resulted from the growth of FLOSS. It is hard to maintain quality while growing rapidly. Still, things get sorted out much more rapidly with FLOSS than closed source. Where I last worked what was accomplished with IT increased many times with GNU/Linux compared to that other OS, which limited us mostly to word-processing. We added multiple servers and services at no extra cost and obtained valuable and reliable service. With that other OS, it was all we could do to keep machines running (slowly).

It looks as though Android/Linux will continue to roll in 2011 and GNU/Linux will continue to nibble at the XP machines and OEM products. I expect that to continue throughout the year with a flourish of ARMed devices of all kinds in the second half. ARM is ready for multimedia, e-mail and browsing. The more people use the web, the more ARM will be part of their day and GNU/Linux will join. Android has definite advantages for rapid development of software for touch but GNU/Linux already has a ton of software ready for larger machinesscreens that should come on in the second half.

On ARM, GNU/Linux will go head-to-head with that other OS and do well. GNU/Linux is so much more efficient because it does not need anti-malware, phoning home and DRM wasting resources and there is much less bloat. The fact that M$ is bothering with ARM is proof that they see the risk otherwise they would develop WP7 and WCE and leave “8″ out of the ARMed picture.