Published by Robert Pogson February 19th, 2010
in Uncategorized.
Matt Asay writes that he now uses GNU/Linux on a desktop PC.. It’s about time. What kept him? Macs.
Oh well, better late than never. The year of GNU/Linux on the desktop has come and gone, he agrees, and we will go forward to much greater accomplishments.
Some of the comments to TFA are humourous. There are some who still feel Linux drivers are a problem. I have installed GNU/Linux on hundreds of PCs over the years and find drivers less of a problem with GNU/Linux than that other OS. In the past few weeks I have had XP machines reject HP Laserjet4 and USB mice, at least temporarily. I have not had that problem in years for GNU/Linux. I also run a single image for all of our PCs with GNU/Linux but I need a separate one for each type with that other OS or it takes forever and a lot of re-re-reboots to get an image to work. My users have no need for BASH, either unless I am teaching them the details of GNU/Linux. Few except the computer geeks doing Computer Science need that.
This story is an example of why I was uneasy with the naming of Matt Asay by Canonical. Why would one devoted to FLOSS find MacOS better in any way? Perhaps he was preferring non-x86 hardware previously… Nevertheless this is a step forward. Let us hope there are many more. I hope Matt’s new fondness for GNU/Linux extends to ARM and thin/virtual clients as well as x86.
- Robert Pogson
Published by Robert Pogson February 19th, 2010
in Uncategorized.
SCO v World will not die…
Over at Groklaw, we read, “The Court finds that Plaintiff’s slander of title claim, as a claim that was resolved on summary judgment on the sole issue of copyright ownership, was appealed and reversed and is now before this Court for trial.”
Those are the judge’s words in deciding to allow SCO to present evidence about the slander of title claims that were thrown out and not-appealed by SCO. The logic of the decision baffles me. If SCO did not appeal the slander of title ruling, how can the judge state that it was reversed? The slander of title ruling was based on the well-founded observation that SCO was unable to prove ownership of the subject copyrights and so could not claim slander and also that Novell had a reasonable belief that Novell owned the copyrights and had not transferred them to SCO so on two grounds slander of title was out. Kimball wrote:”This court’s conclusion that Novell owns the UNIX and UnixWare copyrights impacts several of the claims asserted by both parties and several pending motions. Novell’s motion on the copyright issue is brought with respect to SCO’s First Claim for Relief for slander of title and Third Claim for Relief for specific performance. Novell is entitled to summary judgment on SCO’s First Claim for Relief for slander of title because SCO cannot demonstrate that Novell’s assertions of copyright ownership were false.” This does not seem to be an error by judge Stewart. He deliberately allowed a motion that Novell had not appealed a matter so could not discuss it now but disallows a motion by Novell that slander of title should not be back on the table, magically resuscitated without having to appeal. That makes no sense. Appealing to the SCOTUS will take longer than the coming to trial of this matter so the trial and justice will be in desrepute.
Presumably Novell will soldier on with this handicap and SCO gets to bring up years of irrelevant evidence as it did before. How the trial can go in two or three weeks with this Pandora’s box now opened is beyond me.
Another strange thing. The judge made his ruling without a hearing… That can happen if the matter is clear but there is obviously a dispute. Why not have it out in a hearing? I fear the fix is in. SCO is getting everything its own way, even having Utah law applied to a California contract. Novell will have to start all over again in SCOTUS. How many years will it take?
- Robert Pogson
Published by Robert Pogson February 19th, 2010
in Uncategorized.
Baseline Magazine has a slide-show of 40 Fast Facts on Linux. Have a look if you can. It’s fun. There are a couple of slides with which I disagree:
- Slide 14: “In 2009, Linux has 33.8% revenue on servers compared to M$’s 7.3%”, and
- Slide 15: “As of January 2010, Linux still had only 1.02% marketshare within desktops.”
Slide 14 makes no sense at all:
“Microsoft Windows server revenue was $4.5 billion in 3Q09 showing a 12.8% year-over-year decline and comprising 43.0% of all server revenue in the quarter. Windows servers account for the single largest segment, by operating system, in the worldwide server market.
Linux server revenue declined 12.6% year over year to $1.5 billion in the quarter. Linux servers now represent 14.8% of all server revenue, up slightly from 14.0% a year ago.”
see IDC
Slide 15 is based on Netcraft’s silly non-representative sample of webspace. Their partners are largely business to business sites as far as we know. Business is stuck on that other OS so it is not surprising their stats show that other OS dominates to that extent. In the real world GNU/Linux is on 7% or more desktops according to Ballmer and real-world weblogs. Brazil has had production figures up to 20%. Other equally non-representative webstats are W3Schools showing 4.6%.
NetApps numbers are about the lowest you will find on the planet. I suppose that is why fans of that other OS are so fond of them.
M$, itself, has numbers which show Brazil was making huge moves to GNU/Linux as early as 2003. See figures. Anyone who clings to 1% is a desperate fool.
- Robert Pogson
Recent Comments