08.04.10

Misleading Stories Based on Misleading Statistics (Microsoft and GNOME)

Posted in GNOME, GNU/Linux, Microsoft, Ubuntu, Vista 7, Windows at 6:41 am by Dr. Roy Schestowitz

From Wikipedia: “Example of biased sample, claiming as of June 2008, that only 54% of web browsers (Internet Explorer) in use do not pass the Acid2 test. The statistics are from visitors to one website comprising mostly web developers.”

Summary: Why Microsoft’s latest Vista 7 ‘usage’ hypnosis is false and why the GNOME census does not tell the whole story

WE ARE NO FANS of statistics (although we quite like Netcraft saying that Techrights is ranked 1060^th on the Web for traffic because it’s skewed towards/in favour of UNIX/Linux users). While statistics are enormously valuable in scientific research, a lot of statistics are being delivered without the raw data or the methods used to arrive at the statistics’ output, e.g. tables and charts.

To give an example, right about now the hardcore Microsoft boosters (e.g. Ina Fried) cite the Microsoft-sponsored “NetApps” to pass a talking point: new Windows is better than old Windows. Wow! How can Vista 7 possibly be used more than Vista, according to a Microsoft-sponsored source? Well, whatever. Net Applications does not share information about its methods or its data (let alone the data itself) and it declined when asked to. Based on the evidence we have, Vista 7 is not selling well, but Microsoft hypes it up with some meaningless numbers, just as it did when Vista was ‘sold’ (these are not real sales for reasons we listed before).

The point we are trying to make here is that there are “lies, damned lies, and statistics,” as the old saying goes. To quote Wikipedia, “A misuse of statistics occurs when a statistical argument asserts a falsehood. In some cases, the misuse may be accidental. In others, it is purposeful and for the gain of the perpetrator. When the statistical reason involved is false or misapplied, this constitutes a statistical fallacy.

“The false statistics trap can be quite damaging to the quest for knowledge. For example, in medical science, correcting a falsehood may take decades and cost lives.”

“Asay needs to learn basic math. So RH existed ~2x as long as a Gnome contributor. Still doesn’t explain 16:1 ratio.”
      –Fab ScherschelYesterday was the last time we wrote about the flack Canonical was getting over the volume of its contributions to GNOME [1, 2, 3, 4]. Some moments ago we found an interesting remark from Matt Asay, Canonical’s COO. He is saying that all-time contributions were measured for the GNOME census, going back well before Ubuntu’s existence, to which Red Hat’s Wildeboer responded to me with: “…which is partly correct, but Neary only looks at stuff in GNOME 2.30, so subset only. [...] It is all components that are in 2.30 with their full history. Not full hist of GNOME.”

There is an ongoing thread about the subject (see the arguments there for more details). Fab Scherschel takes Wildeboer’s side and says: “Asay needs to learn basic math. So RH existed ~2x as long as a Gnome contributor. Still doesn’t explain 16:1 ratio.”

I have actually been attempting to defend Ubuntu here, right from the very start. Here is an article from Bruce Byfield and from several others who wrote about the same subject this week (analyses and opinion we have not mentioned yet):

• Does Ubuntu Contribute its Share to Free Software Projects?

  But, whatever the reason, responses from Canonical were quick in coming. Canonical Chief Operating Officer Matt Asay tweets that Neary’s analysis “tracks *all-time” Gnome contributions. Canonical will never catch up w/ RHT. It’s not helpful data.” Asay is referring to the fact that Red Hat was already founded when GNOME began in 1997, while Canonical did not exist until 2004.

• Commercial Firms–Especially Red Hat–Move GNOME Along

  Dana Blankenhorn hits the nail on the head regarding the Red Hat contributions. “GNOME is far more valuable to Red Hat than a Red Hat UI could ever be, because it’s open source, because it’s a commons,” he writes. Exactly, and efforts to refer to GNOME as “The Red Hat UI” are off-base. It’s an open source interface–open to everyone, benefiting from contributions from a commercial open source company that is doing well.

• GNOME and the long tail of the commons

  Now it’s true, there are opportunities here for people who run GNOME to take some meetings with firms who currently get more in benefits from its code base than they contribute. How can we help you help us, they might say. I suspect, though, that the answer is simply they have other priorities right now beyond a desktop Linux UI.

• Red Hat Is Number One Gnome, Is There A Control Freak Problem?

  Red Hat says that it is committed to the collaborative development of an open alternative to proprietary client operating systems. The company Red Hat serves as a member of the Gnome advisory board, in addition to Canonical, Collabora, Debian, Free Software Foundation, Google, IBM, Igalia, Intel, Motorola, Mozilla Foundation, Nokia, Novell, OLPC, Oracle and the Software Freedom Law Center.

• Courage, Class, and Canonical

  DeKoenigsberg’s story reminds us all that behind the code, the business strategies, and all of the words that are thrown out at each other in the Linux community, there are still real people with real emotions running this crazy, chaotic show.

• Who Really Contributes the Most to Linux?

  Now, this is a dynamic that plays out in individual organizations, and isn’t the case in the anti-Canonical world. Canonical’s “marketers” aren’t making promises that Linux developers must deliver on, so I think the anti-marketing bias is residual. Or maybe it’s resentment that the Linux distro that gets all the glory isn’t an engineering powerhouse, but just a re-packager that’s riding everyone else’s coat tails.

Ignore the flames from longtime freedom basher Andrew Orlowski, who also loves to promote the MPAA|RIAA. He is one of the reason I quit reading The Register. █

Permalink  Send this to a friend