Around the world, folks are reporting that demand by consumers for “PCs” has dropped off. At the same time demand for smart phones and tablets continues strongly. Think what that means… OEMs of PCs are not giving consumers what they want, small cheap computers. Take a hint, fellows. Small cheap computers are selling. Crank them out. No, you don’t need Wintel to do that. Linux on ARM will do just fine. Continue Reading

Enough dancing and hand-waving. Adobe has drunken the 1% KoolAid.

• no more builds of Adobe Air for GNU/Linux except by partners of Adobe
• ditto Acrobat Reader

Commentators pointed out that Adobe does not supply 64bit Flash for GNU/Linux so who needs Air, a Flash generator? 64bitness is pretty standard these days even on Atom processors. How weird.

Adobe quotes 0.5% of Air downloads were for GNU/Linux but with much of the world using 64bit GNU/Linux, that is pretty good. Adobe, you have to supply what the consumer wants. Years ago, I identified 64bitness as one of the key features of GNU/Linux and use it on terminal servers extensively. I was quite annoyed that it took so many years to get a 64bit version of OpenOffice.org.

Apparently, Adobe sees the future as Android/Linux and we can run Android on GNU/Linux so it may not be a huge issue, just a change in programming language. That must be much more expensive than building for GNU/Linux. That means a re-write.

HTML5, here we come.

see Adobe – Focusing on The Next Linux Client

Despite a public roadmap showing future development of Itanium processors past 2011, Oracle has decided to end support of Itanium systems for its database. HP is suing because it makes a ton of money selling Itanium servers intended to run Oracle. Ellison insists that Intel intends to end production of Itanium. Is he mad?

“Intel’s work on Intel Itanium processors and platforms continues unabated with multiple generations of chips currently in development and on schedule. We remain firmly committed to delivering a competitive, multi-generational roadmap for HP-UX and other operating system customers that run the Itanium architecture,” said Paul Otellini, president and chief executive officer of Intel. That, in response to Oracle’s end of support announcement, is pretty definitive, I would say.

HP claims Oracle has contractual requirements to support Itanium and Oracle denies that. Grab the popcorn and watch the collision in the courtroom. They cannot both be right and the judge should be able to decide this one clearly…

Comments from all over:

• 2011-3-28 Art Wittmann of InformationWeek recommends cutting reliance on Oracle. Does that mean PostgreSQL? That can be justified on technology, TCO and support. Here’s a test against Oracle on an existing database.
• Ellison does correctly observe that RedHat no longer supports RHEL 6 on Itanium but will supply security fixes until 2014 on RHEL 5.
• HP also accuses Oracle of dirty tricks to get HP’s customers to switch to Oracle on SUN servers by providing no more fixes for Oracle on HP servers…
• HP’s legal complaint has redactions hiding the juicy parts.
• Oracle promises to reveal Intel’s plans to end the life of Itanium in court. This could seriously damage Intel and HP if true. This could seriously damage Oracle if false…

With such childish arguments in the playground from pillars of IT, more businesses should consider FLOSS and if PostgreSQL is lacking, contribute to beefing it up. That would bring lower costs and greater security than relying on these guys. The whole world needs databases and can make its own. There is no need to have a guy like Ellison stomping through server rooms with his monopoly on enterprise databases.

Unlike Oracle database, PostgreSQL is Free Software and you get the licence with a free download of the software and you can run it, examine it, change it and distribute it. That’s the right way to do IT.

Benchmark with PostgreSQl on one SUN T2000 = 778

Benchmark with Oracle on two SUN T2000s = 733

Is it hard to justify double the price for similar performance?