Robert Pogson

One man, closing all the windows.

Posts Tagged / uptake

  • Apr 17 / 2014
  • 34

FLOSS Is A Winner

No matter how many times we read that FLOSS is junk made by amateurs “In 2013, for the first time, we saw open source quality for the projects in the Scan service surpass that of proprietary projects at all code base sizes. The 2012 Coverity Scan Report looked at a sample analysis of more than 250 proprietary code bases totaling more than 380 million lines of code, with an average codebase of nearly 1.5 million lines of code, and we found that open source code had lower defect density levels up to 1 million lines of code. For the 2013 report, we analyzed approximately 500 million lines of code across almost 500 proprietary C/C++ projects.” the reality is different. FLOSS is made by all kinds of programmers but because it is FLOSS and everyone can run, examine, modify and distribute the software, more eyes make bugs disappear. Coverity is one of the eyes and they tell us that out of hundreds of millions of lines of code scanned, FLOSS has a lower density of defects.

Size of Codebase (Lines of Code) FLOSS Non-Free
Less than 100,000 .35 .38
100,000-499,999 .50 .81
500,000-1 million .70 .84
More than 1 million .65 .71
Average across projects .59 .72

Accept it. No matter whether it’s price, performance or correctness that matters most to you, FLOSS is the right way to do IT.

See 2013-Coverity-Scan-Report.pdf.

  • Apr 16 / 2014
  • 1

Wintel Sinks Further

As expected, Intel has raised prices in an attempt to maintain profits as long as possible rather than trusting the market to yield them a reasonable living.“PC Client Group revenue of $7.9 billion, down 8 percent sequentially and down 1 percent year-over-year.” This will hasten the demise of Wintel as consumers see greater advantages to switching to */Linux on ARM. Without the monopoly on retail shelves for legacy PCs there’s no way Intel would raise prices at all and consumers should vote with their wallets. Expect 2014 to be the greatest year yet for FLOSS on ARM.

See Intel News Release.

See also, CFO Commentary on First-Quarter 2014 Results

  • Apr 15 / 2014
  • 3

April 2014 Project of the Month, Free Pascal

Sourceforge has been and will continue to be one of the world’s great repositories of Free Software. Project of the month is my old favourite, FreePascal. Some of the reasons for my liking it are condensed in a single question/answer between SourceForge and the original authour, Florian Klämpfl:
“SF: Who can benefit the most from your project?
FK: I think there are multiple groups who can benefit from it:
- People who want to learn only one programming language which allows them to use it for almost everything: FPC can be used to do big database applications but it can be also used to program embedded devices. It can used to write numerical applications but also to code for mobile devices.
- People who have a large Pascal/Object Pascal code base
- People who are interested in a programming language which offers a compromise between high productivity and the advantages of native code.”

Amen! I would add that FreePascal also has a very dense/sparse syntax so it is very easy to learn and use to full advantage. This also greatly increases readability.

See April 2014 Project of the Month, Free Pascal | SourceForge Community Blog.

  • Apr 11 / 2014
  • 15

Wintel Balloon Deflates

One of the most fun things in life is the site of a buoyant balloon taking flight, reaching unimaginable heights gracefully and easily. That was the old Wintel monopoly when neither Intel nor M$ had to do anything to dominate all of IT. When the leak started in the middle of XP’s reign, no one was fired for buying Intel and M$’s stuff. Many folks were run out of business simply for providing good products at reasonable price. Not so now. M$ has had to actually build decent products over the last decade but it wasn’t enough to keep the balloon up. Wintel was too expensive, too bulky, and too rigid to do what users wanted done.

“executive VP in charge of operating systems Terry Myerson, told ZDNet’s Mary Jo Foley that he was OK with a services model. Specifically, when she asked about the Nokia X line of phones–those new low-priced Android phones running Microsoft services announced at Mobile World Congress. He was cool with a phone running Android, precisely because he was happy to see Microsoft services running on as many devices as possible, regardless of the operating system”It took a lot of work but ARM and the FLOSS community have bypassed both companies. To remain relevant, Intel is diversifying into ARM and making x86 as efficient as they can with Moore’s Law and every other trick they can find. Still Intel’s processors are more expensive than ARM even if energy consumption is not as much disadvantage. There’s just too much silicon involved. Meanwhile, 8-core CPUs and great graphics modules abound in the ARMed world and ARMed CPUs are outselling Intel by a wide margin. Android/Linux has done a similar number on M$’s stuff, so much that M$ now seeks to leverage its PC and server platform into services to sell, just like Google which Ballmer accused of having no business-plan…

See Turning The Ship: Microsoft Might Have Begun A Subtle Shift From Windows To Services.

  • Mar 20 / 2014
  • 2

Tablets A Bulk Commodity

In 2012 tablets were barely a trickle. In 2013 they took off, at least for the Android/Linux kind. Now shiploads of them are moving from factories in China to every corner of the world.“This is a pivotal year for tablets as the traditional markets leading computing and CE deployments reach a tipping point and give way to growth regions.The most mature markets, including North America, Western Europe, Japan and South Korea, are yielding to other regions, such as Latin America, Eastern Europe, Middle East and Africa, where the connectivity infrastructure and distribution channels are coming of age.Market intelligence firm ABI Research forecasts 200 million branded tablet shipments this year—a 20% gain year-over-year from the 166 million shipments in 2013.”
You just can’t hide 200 million of the small cheap computers. They are the new PC, along with thin clients, ChromeBooks, smartphones and those dusty legacy PCs… I completely missed that. There is only one tablet in the house and I rarely use it. I thought for sure GNU/Linux would take over the legacy PC to replace that other OS. Instead the small cheap computers are replacing the legacy PC and the OS M$ ships. It’s all good.

I expect with this flood we will see more people doing the obvious and attaching keyboards and mice for “productivity” and just never replacing those legacy PCs. Wintel may well survive a bit longer but either as one of a large number of choices in the market or a solution to use until it dies on the current hardware. The last locked-in market for Wintel seems to be businesses who figure they need such and such application which only comes on that other OS. That will change pretty rapidly as there’s even choices for web applications and cloud services. The solid share that Wintel had with consumers just a couple of years ago will evaporate just as rapidly with businesses given a choice. Let’s see, what’s a business to do when the choice is eternal slavery or picking up a truckload of tablets and connecting them to a GNU/Linux server or some cloud application? Everyone knows they have choices now and many will take other choices than Wintel.

Let’s see who can paddle the fastest to escape Wintel.

SEE Branded Tablet Shipments to Reach 200 Million in 2014.

  • Mar 10 / 2014
  • 10

Positive Feedback: M$ Uses XP To Publish The Insecurity Of Using That Other OS

I thought it was some kind of a joke when I read the quotation to the right.“There’s the real possibility that large-scale infections of Windows XP will paint the Windows brand as insecure, fulfilling the implicit prophecy the company made late last year. To most people, Windows is Windows is Windows, with no distinction between XP and the newest, locked-down 8.1. And for those people, Windows is Microsoft because it’s the best known of the company’s software.” After all, I and many others fled to the relative security of GNU/Linux simply because of the insecurity of XP. “What reputation?”, I thought.

The flaw in my thinking, of course, is that others are in the same condition I was in when I used DOS and that other OS back in the day. I just didn’t know any better. I blamed the hardware. I blamed myself, because software I wrote didn’t crash when I clicked on something. I had to be doing something wrong because a big outfit like M$ just wouldn’s ship crapware. Would they?

That’s the point the authour makes. Even if the reputation is false/fluff, it’s what people think they have to deal with using that other OS and they think it’s OK that malware infects a PC. Well, whatever the thoughts, when the ~50K bugs that M$ created in XP and of which M$ has fixed only a tiny percentage, continue to be discovered by motivated malware-artists, knowing none will be fixed… all Heck is going to break loose in the world of XP. M$ is gambling that the Horror will drive folks to buy a new PC with a new version of that other OS but what if folks finally see the light and blame M$ for the crapware that M$ has produced?

I believe a lot will shift to GNU/Linux or Android/Linux or Chrome OS. See, while I had to do a bit of work to discover I had choice back in the day. Current users of XP know there are several choices not from M$. M$ has even been advertising Google’s Chrome OS on prime-time TV… Chuckle. This could be even bigger than I thought. The only way the effect will not appear is if the malware-artists collude to prevent XP bogging down too much, by adding a scheduler for some slice of the free time on XP machines. Nah, that won’t happen. Those guys are greedy and want it all. Besides they are rugged individuals who do their own thing. Some even think they have a backlog of malware they have been waiting to release until after M$ gets out of the picture…

See Perspective: Microsoft risks security reputation ruin by retiring XP.

  • Mar 07 / 2014
  • 44

M$: We Can’t Beat ‘Em, So We’re Going To Snow ‘Em

Ghandi was right. When they fight you, you win…“In an internal memo sent to Microsoft employees Nadella said that the core focus of company is going to be advertising and not software development. He clarified that by advertising he didn’t mean display advertising that Google is into. Under Mark, Nadella is setting up a new department which will get most budget and the core focus of the group will be to create anti-Google ad campaigns.

Microsoft is expected to spend over $5 billion dollars on anti-Android and ChromeOS ads. Microsoft’s previous campaign internally called ‘MTP – Mess The Press’ has already started bearing fruits as even the most noted journalists now ‘recite’ that the company is making billions of dollars from Android, even if there has never been any proof.”
M$’s new plan is corroboration of what I have been writing for years now. M$ can’t beat FLOSS on technology. You just can’t beat Free and $free. It’s too good in price/performance. The obvious thing for a real business to do would be to adopt FLOSS methods and profit but M$ is not a real business, but a monopoly, a fading monopoly. Cornered beasts are very dangerous.

Fortunately, the law is on the side of FLOSS. Deliberately setting out to harm the brand of a competitor is widely seen as a crime in USA and globally, so the cornered M$ reveals its true self, a lying bully extorting the world’s supply chain for decades to remain on top in unit sales. Fortunately, FLOSS has won in the market despite all M$’s machinations. Spending $billions on ads to deny that will be seen by the world as the babblings of a demented old man long past his prime.

As an M$-hater, this is the next best thing to M$’s bankruptcy. I will get to see this train-wreck happening in markets and courts for years to come as I watch my grass grow, flowers bloom, garden bear fruit, hunts feed my family and this blog celebrate the decline of evil in the market for IT.

See Microsoft to cut on software development, focus on ads.

  • Mar 06 / 2014
  • 5

The World According To IDC

I like IDC. They are an invaluable source of information about shipments of personal computers. I love PCs.“PCs include Desktop, Mini Notebook and other Portable PCs which possess non-detachable keyboards, and do not include handhelds or Tablets” One thing that’s pretty silly, however, is IDC clinging to the strange idea that a personal computer can’t be a tablet or smartphone. That warps whatever maths they do to make their estimates. What happens if 100 million people decide a tablet is a PC this year or next? What happens if 300 million people decide a smartphone is all they need?
The world looks quite different then. Clearly that should happen. Why has there been decline in IDC’s PC-worldview in 2012, 2013 and 2014 (forecast), but no further? That’s not realistic. There are reasons PCs of IDC’s kind have declined:

  • size/portability
  • price

The small cheap personal computers are small and cheap. People love them. They are personal computers, IDC.
People are just tired of paying for what they don’t need: wasted material and space and licensing fees for M$. Smartphones can do everything most people need done with a computer, by far: HDMI, USB, browsing, messaging, voice and GPS… Hook up a keyboard, mouse and monitor if you need something bigger. There’s no need for a bigger, more expensive box. Nowadays, almost everything except the package is built into a single chip.

See IDC Expects PC Shipments to Fall by -6% in 2014 and Decline Through 2018.

UPDATE In a fine example of one hand not knowing what the other is doing, IDC reported that the rate of growth of tablets shipments is slowing and shipments will be about 261 million in 2014. Surely, some percentage of tablets are replacing those clunky legacy PCs so there is continuing reason to believe growth of legacy PCs will be low or even negative for years to come and IDC knows it.

  • Mar 05 / 2014
  • 5

Wikipedia: Decline Or Maturity?

My recent migration to Debian Jessie, which required fixing my local tweaks for web-applications ended with some examination of Wikipedia, of which I have a snapshot from 2004 or so. I have edited it a lot and customized it for use in the schools where I worked but lately I have not maintained it well. Through various upgrades some links broke and I must have restored a backup that messed up the archive of images. Even damaged it is a huge asset.“The first few edits of these newcomers indicate that they are trying to contribute productively (i.e. acting in good faith) and, therefore, likely will become valuable contributors if they remain in the community. We show empirically that, while the proportion of desirable newcomers who arrive at Wikipedia has been holding steady in recent years, a decreasing fraction of these newcomers survive past their initial contributions. We demonstrate that the decline has been caused, at least in part, by the Wikipedia community’s reactions to the enormous influx of contributors between 2004 and 2007. In order to maintain quality and efficiency during this period, the community’s views toward the goals of the project changed. These new views were instantiated in a set of policies, and a suite of algorithmic tools were developed for enforcement. Over time, these changes resulted in a new Wikipedia, in which newcomers are rudely greeted by automated quality control systems and are overwhelmed by the complexity of the rule system. Since these changes occurred, newcomers – including the crucial, desirable newcomers – have been leaving Wikipedia in droves.” Currently, Wikipedia is one of the world’s great websites, full of information and very accessible. It is somewhat mature in that just about everything has some coverage but more work remains to be done. The question of decline or maturity is about whether or not Wikipedia is less vigorous because the job has been largely done or because changes made years ago have become a millstone, weighing it down.

I will describe some of my contributions over the years. Where I have particular knowledge and I see a gap in some article that I was reading for my own purposes, I have offered edits from time to time. At first, that was that and I went on with life. Then came a time when whatever edits I made were almost certainly rejected by some nameless creatures in the system, rejecting my work sentence by sentence because I did not provide proof of every assertion, almost every sentence or phrase. I kid you not. A paragraph could not be contributed. It had to be a list of sentences with one or more references to the web for each one. Asserting that the sky was blue was unacceptable. One had to prove it. Stating the obvious and how it related to well-known facts and principles, reasoning, was never enough. To me it was as if binary bits were OK if they were copies of stuff elsewhere, but expressing any idea however modest was unacceptable.

Apparently I was not alone in this depressing phase of Wikipedia. According to TFA linked below, thousands of editors have dropped in, contributed, and fled. Wikipedia just isn’t a great place to live any longer.

Don’t believe me. Look at some examples.

GNU/Linux Adoption – when I added this edit,
“There is another reason that web counters are unreliable. Some are clearly connected with use of operating systems in business. For example, when 10000 users at Google’s headquarters moved at once to GNU/Linux in the summer of 2010, [[Net Applications]]’ web stats showed a swing from a few percent to 88% for the city of [[Mountain View, California]]{{cite web|url=|title = Mountain View, California, Penguin Heaven|accessdate = 14 March 2012|last = Pogson|first = Robert|year = 2012|month=March}}, a city of 74000 people, and a swing from 1.87% to 18.69% for [[California]]{{cite web|url=|title = Penguins Seen Over California|accessdate = 14 March 2012|last = Pogson|first = Robert|year = 2012|month=March}}, a state of 37 million people. Clearly, 10000 people is a small change in usage but was over-counted because it was used in business. The fact that Windows is heavily used in business results in low numbers for GNU/Linux. Web-counters can readily select for business usage by counting during hours of business in a location or counting only clients from business Internet domains.” I ran afoul of the “no original research” rule, and my contribution was tagged with a “conflict of interest” because I provided a link to my website. In the discussion of my contribution, you can see the problem:
“NetApplications does not publish the charts, only values month by month and location by location. My blog collects their data. I am not the source. Do I need to cite the URI for each of dozens of datapoints? The paragraph I added is about the change and how and why it happened clearly showing bias in Net Applications numbers. I have a M.Sc. in Nuclear Physics and know how to analyze data. The wide publication of “1%” is clearly wrong information coming form Net Applications and my collection shows that quantitatively. Assume a world using 90% GNU/Linux and Windows adoption at Google. 10K people showing huge adoption that is not valid. Pogson (talk) 13:29, 14 March 2012 (UTC)
Your own blog is not an acceptable reference as per WP:SPS, regardless of what qualifications in stats you have. You can’t cite data points with your own graphs and interpretations about them either as this would be WP:OR and WP:SYNTHESIS and is specifically not allowed on Wikipedia. You need to find proper reliable independent third party refs to retain this text in the article. – Ahunt (talk) 13:45, 14 March 2012 (UTC)
The refs you added do not support the claims you have made in the text, so I restored the “citation needed” tags. Since you removed those again the only choice remaining is to remove the challenged text as per WP:V, which I have done. Please don’t add it back in without proper, reliable refs that actually support the text this time. – Ahunt (talk) 16:30, 14 March 2012 (UTC)
Sounds good, maybe we should archive these older threads as well to stop people adding to discussions from some time ago (hard to follow). IRWolfie- (talk) 09:48, 15 March 2012 (UTC)”
So, how does a contribution to human knowledge make it into Wikipedia? Politics. Popularity. Stuff like that. It’s not enough to be correct or useful, information has to be acceptable to some elite in the organization. Wikipedia has lost its way.

The words that greet me on Wikipedia, “Hello, Pogson! Welcome to Wikipedia! Thank you for your contributions to this free encyclopedia.” are hollow and a sham. If Wikipedia were really free, the powers that be would be all over my contributions improving them rather than deleting them. What does their page say about the reliability of web stats these days? Well, that’s water under the bridge. The Linux Adoption page is quite different now. Just a line or two on the matter but they still cite other’s original research on the topic.

Really, how vital is an organization of house-builders that insist on bull-dozing each other’s work?

See halfaker13rise-preprint.pdf.

  • Mar 05 / 2014
  • 1
Linux in Education, technology

New ICT Curriculum In Indian Schools

“The requirements of the curricula are not to be hardware or software speci c. Undoing the general trend of limiting software to office applications, which are not only ill suited for educational purposes but also tend to narrow down the view of what computers and ICT can achieve, a wide range of software applications specifi cally designed for education are introduced. Use of proprietary software would become very expensive and make the implementation unviable. Therefore, Free and Open Source software have been suggested throughout the curricula. The use of FOSS applications will also obviate software piracy and enable customisation to suit local needs.”

AMEN! This is a national curriculum for one of the largest countries on Earth. It is professionally done and not just about students. It includes training for teachers. Wow! If implemented widely, this should see increased use of computers in education and FLOSS in a country with 1billion+ people and many millions of students.

Further, the new curriculum does not hold teachers back. Those already skilled in ICT will be able to be certified in short order. The new curriculum does not hold students back. It starts with programming computers in the first year (~10 years of age).

The arrival of small cheap computers on retail shelves and OEMs’ catalogues will actually make this possible to implement widely as every school should be able to afford this and if they can’t the central government should be able to fix that.

See ICT Curriculuma.pdf.

  • Feb 24 / 2014
  • 1

Wow! Tux Machines IS BAACCCKKKK!

I noticed a huge pop in my web stats for today and wondered why. The cause was a link on Tux Machines. There was a recent change of ownership. Now Dr. Roy Schestowitz and Rianne Schestowitz seem to be the main authours. In a couple of days they produced a huge number of informative articles mostly links to diverse sites advocating FLOSS and GNU/Linux. I love it.

TUX MACHINES has been around longer than my site and is ranked 8479 by Netcraft. They’ve changed the location and software on the server going from Apache on Debian GNU/Linux to Apache on CentOS. The passion of the Schestowitz family and the user-base of Tux Machines could mean the start of something huge. Good luck to them.

I notice they’ve changed how the site operates to improve performance: “In another attempt to restore user registrations, this time on the new server which has just been configured for mail, we are enabling anyone to quickly self-register (takes less than a minute and requires no verification), then immediately post comments, forum posts, etc.
For well over a year it has been hard to leave feedback in this site and it’s all the fault of script kiddies. This site is ranked 40126th for traffic in Netcraft (without us using the toolbar, which helps one game the numbers, extrapolated from a relatively tiny sample set), so it’s a shame that people can hardly leave a comment.”

I can relate to that… ;-)

  • Feb 24 / 2014
  • 0

Yes, You Can Run (most of) a Winery On GNU/Linux

The crazies who come to my blog keep saying one must have that other OS everywhere on full thick clients. Here’s another anecdote of doing the obvious, running GNU/Linux thin clients almost everywhere with all kinds of benefits: “Our original GT shipped with an early 2.x Gnome release. This had more to do with my general lack of skills with package management and live image building than by design. Since the distro I was using at the time shipped Gnome by default – I went along with it. Since then, we’ve migrated to KDE 3.5, back to Gnome 2.8 and finally to KDE 4.9 which we’ve just completed the rollout for, and which now makes up approximately three quarters of our 250+ desktop fleet.
The key to all smooth migrations we’ve found is Desktop Environment consistency. Keep the major applications cross-platform where we can (browsers, office suites, assorted tools). Keep the icons where people are expecting them (they’re in the same spot on our Windows desktops too).”

Amen! Give the users a little consideration and better performance at lower cost and you’re well on your way to software Freedom. I recommend Debian GNU/Linux and XFCE4 but whatever you choose will work for you and your organization.

See KDE Software Down Under.