Archive for October 2nd, 2010

FLOSS on YouTube

YouTube is a wonderful tool for communication. Text does not carry the body-language/facial expressions that video does. On YouTube there are many fine pieces about FLOSS from OpenWorldForum:

  • Roundtable: FLOSS: Successful implementations in the regions of Europe – four informed people including Florian Schiebl of Munich’s Limux project discuss the status and challenges of implementing FLOSS in governments. The consensus seems to be that all technical problems can be solved and everything is easier by converting to open standards first. Exit costs of selecting any technology should be considered. This accounts for lock-in and makes it real.
  • The Brazilian Experience. 50000 developers and users share 30 projects.
  • The State of Play in the UK (“The Sick Man of Europe in FLOSS”) – the government is seriously locked-in and spends $16 billion (UK) annually on IT and has some legacy cruft. The UK government has often sided with big business in IT. In spite of that there is lots of adoption in local governments, businesses and schools. All national political parties support FLOSS.
  • Mark Shuttleworth’s keynote. He emphasizes that FLOSS sells itself to business but there is work to be done to make FLOSS attractive to ordinary people: regular short release schedules, long-term support releases every two years, quality (uniformity, automated testing and formal quality control), and better design for usability.
  • Check out the ad.
  • Guadalinux on hundreds of thousands of computers in schools and offices in Andalusia, Spain. A million downloads so far.
- Robert Pogson

Mobile

The mobile world is getting very excited with smartphones. Interestingly, NetApplications uses mobile share to downplay hits by GNU/Linux systems. Android and JavaME (some of which run on GNU/Linux) are not counted as “LInux”. That would show “Linux” had an increase this month instead of the decrease shown. Further, the USA-centric nature of NetApplications may be reflected in the radical differences between the chart in this report on North America and other regions. Here are Gartner’s recent results for mobile. That other OS is really in the tank globally but it has a decent share in USA. Go figure…

Now, NetApplications mobile hits are small but they proportionally hit GNU/Linux share harder because of its small base. Thus, the sudden drop in GNU/Linux share is nothing to do with GNU/Linux at all but the increasing use of smart-phones to browse. If all the smart-phones that use GNU/Linux were added in there would be a nice pop instead of a drop. It’s as if Vista were suddenly dropped from M$’s share… They would not do that but they do for GNU/Linux. That looks like bias to me.

Let’s do the maths:

  • GNU/Linux desktop share 0.85%
  • JavaME mobile share 0.95%
  • Android mobile share 0.24%
  • Total share for GNU/Linux of all kinds 2.04%

See NetApplications numbers for September 2010.

Compare that with Wikipedia’s numbers for June, 1.88% and W3Schools numbers for August, 4.9%.

When you see the shocking differences in OS between smartphones in USA/World, you can appreciate the deception that NetApplications represents the world in any way. NetApplications’ world certainly is not the world of high performance computing where 91% of the top 500 run GNU/Linux. Is it not interesting when performance counts and competition is wide open that that other OS slinks away?

- Robert Pogson

Aiding and Abetting Theft

There’s news that police have rounded up a bunch using e-mailed viruses to access bank accounts. Why isn’t M$ in the docket for aiding and abetting? Aren’t the people who made a browser/OS in which clicking on or viewing an e-mail installs a virus just as guilty as the thieves? Remember Napster? Why were they shut down and the trash that M$ distributes as an operating system is allowed to wreck the web?

- Robert Pogson

Desktop Applications

InfoWorld has a FUD article questioning the viability of FLOSS desktop applications. The authour and some commentators go on and on about the lack of a viable business model for FLOSS. They ignore the obvious:

  • FLOSS does not need a viable business model to succeed. It just needs people to use and to embrace FLOSS. The world needs software and can create its own. It does not need “business” to do that. Many hands make light work. The world has billions of people and millions of us can create software.
  • Successful business models can be created on FLOSS: OpenOffice.org, FireFox, Google, RedHat, Suse, etc. They can do it on the desktop and server and they have done so.
  • The question of how the programmers get paid has been answered:
    • businesses that make money doing whatever can pay programmers instead of licensing fees
    • youngsters learning the trade can do FLOSS as a part of their learning and to put something good on the resume
    • people can create FLOSS applications for fun or to help them do what they like to do or for work
    • people can contribute to FLOSS what they can and share the work with others while using others’ work – sharing works
  • the GPL and other Free Software licences mean Free Software does not depend on the success or failure of a business. The software remains Free however the business goes

So, TFA and some of the comments are pure ignorance. The presumption is that OpenOffice.org has failed as desktop software because M$’s Office still remains active. What nonsense! That is “winner-takes-all” and not relevant to the question of success or failure. Being driven from the market is only one way to fail and monopolizing the market is only one way to succeed. M$ has failed drastically in that while people may use the software, many hate the company and its software. How is that “success”? Further, OpenOffice.org is doing very well with 100 million users. How is that “failure”?

I instroduce Scribus, Koffice, Lyx and AbiWord to my students. They like them all. All those products are successful. People can do amazing things with them and the programmers did get paid.

I have introduced many teachers and students to GCompris which is a suite of 80 applications to teach young children all kinds of useful skills and facts. The teachers that created the programme produced it to help them do their jobs. They are paid to teach and the sharing of the software is a natural thing they do to support their profession and to help children. GCompris is a wildly successful desktop application.

There are thousands of desktop applications that are FLOSS and many are successful, so the premise of the article is false and the answer to the question is that FLOSS is the tree of life for desktop applications. They all start small and grow or not according to the vagaries of the software cycle, not the wishes of large corporations. Large corporations have done well to embrace FLOSS on the desktop and server but they are not the deciding factor and their particular business models are not particularly relevant to the topic of FLOSS.

- Robert Pogson

Small Chips in Big Servers

Intel, naturally, is still trying to push big chips in big servers because they have Xeons out there that can really move data. However, the energy cost per transaction is very high:

  1. 64bitness may be OK in the data-channel but does the CPU need it? For addressing large memory, perhaps.
  2. we don’t need 12 MB caches on a CPU directing data from disc storage DMA buffers to a NIC. That’s a tight copy/execute loop.
  3. we don’t need a huge instruction set for some servers
  4. if the CPU cannot handle all processes in caches, it needs data-paths and speed much faster than the load requires

It’s just smarter to have a reasonably fast CPU waiting on one or a few processes and having many CPUs like that in a big server. It depends on the application, database, and load to decide what works most efficiently. Xeon is good but it is not the best answer to every question. Atom/ARM may be better in some cases. ARM is pushing. Intel is not.

Intel was caught napping on 64bitness. They may be napping on the ideal cloudy server. If Intel lets ARM march in, ARM and GNU/Linux could take a big chunk out of Wintel in the server/workstation space. What’s wrong with a ton of passively cooled CPUs on a motherboard? Is Intel worried for the CPU cooler makers? Is Intel worried that ARM may be better at this role than Atom?

- Robert Pogson



Archives by Month

My Mission

My observations and opinions about IT are based on 40 years of use in science and technology and lately, in education. I like IT that is fast, cost-effective and reliable. I do not care whether my solution is the same as yours. I like to think for myself.

My first use of GNU/Linux in 2001 was so remarkably better than what I had been using, I feel it is important work to share GNU/Linux with the world. I have been blessed by working in schools where students and school systems have benefited by good, modular software easily installed in most systems.

I have shown GNU/Linux to thousands of students and hundreds of teachers over the years and will continue in some way doing that until I die in spite of the opposition.

Posts

October 2010
S M T W T F S
« Sep   Nov »
 12
3456789
10111213141516
17181920212223
24252627282930
31  

    Writing

    3434 articles
    30664 comments

      Comments

      platforms
      linux 17507
      windows 12796
      macos 206
      sun 3
      wp 2

      browsers
      firefox 23960 
      safari 11878 
      chrome 11730 
      ie 4657 
      iceweasel 4280 
      opera 1643 
      konqueror 198 
      netnewswire 14 
      epiphany 2 
      flock 0 
      bonecho 0 
      lynx 0 

Bad Behavior has blocked 6348 access attempts in the last 7 days.