Linux Today: Linux News On Internet Time.

Editor's Note: Get the Facts Yourself, Redmond

Mar 17, 2006, 23:30 (25 Talkback[s])
(Other stories by Brian Proffitt)

By Brian Proffitt
Managing Editor

I went to Vincennes University on Tuesday to give a talk about the technological and business structure of Linux and open source software. There were about 70 students in the audience, and I think the talk went well. (At least no one passed out from sheer boredom.)

It was a mix of business and computer students, with the computer students comprising of about one part programmers and three parts system and network admins. Interestingly, during the Q&A session, it was the alpha male of the programmer gang who seemed to have the most trouble believing that Linux was a viable desktop solution, asking me if I really believed it was more stable than Windows.

I responded that in all my years of using computers, I'd only seen two full-blown system crashes on Linux, and countless blue screens of death on Windows machines. He indicated that the opposite was true--he'd seen plenty of kernel panics and only a few BSODs. This was one of those discussions best taken off-line, since the rest of the audience was getting quickly bored with the technical back and forth. Unfortunately, the kid left before I could ask him just how the heck he got so many panics--though I suspect it was due to his work as a programmer.

Clearly, though, the concept of open source had not impressed him or, it appeared, his cadre. Vincennes University as a whole, however, is very much embracing open source, having implemented it in quite a few IT courses with more plans on the way. Very exciting stuff.

After getting back to the home office, I found an anonymous note in the contrib queue of Linux Today (the page where I see all of the story recommendations sent in via the site). It was titled "Fake Microsoft Story?" and pointed to a URL on Microsoft Malaysia's Web site. I read it, and what was a good day became that much better.

Though you wouldn't think of it when you read the article's title: "Jukebox Eliminates Inefficiencies by dropping OpenOffice For Microsoft Office," this article contains some gross inaccuracies that reflect poorly on either the vendor, the customer, or both. The article is a case study about Malaysian retail shoe company Harian Shoes Sdn Bhd, which is known for its Jukebox chain of stores in that Asian nation.

Apparently, according to the study, Harian switched over to OpenOffice.org a while back and found that they didn't like it: "However, [Harian] got less than it bargained for when it discovered the application had certain weaknesses such as the unreliable conversion of documents to Microsoft Office format. In addition, the company's IT users spurned OpenOffice [sic], preferring to use Microsoft Office which was installed on some of the company's PCs."

Now, right away, you have to wonder: "unreliable conversion of documents"? What's that about? I remember a time when documents saved as .doc or .xls in OpenOffice.org were slightly off in formatting (usually on documents that were heavily formatted to begin with), but that old bugaboo hasn't been trotted out in quite some time. But, even if you buy into this problem, there are some serious contradictions in this story, most of them surrounding the issue of timing.

You see, according to the Harian employee interviewed for the study, Marketing Manager Anson Leow, Harian migrated to OpenOffice.org in the "late 1990's."

Hello? Late 1990s? That would be a neat trick, considering OpenOffice.org didn't even exist back then.

OpenOffice.org, the open source offspring of Sun Microsystem's StarOffice, was not released to the general public until October 13, 2000. (It was announced on July 19, 2000.) In the late 1990s, only StarOffice was available.

While a few year's difference may not seem like much, it pokes a huge hole in the veracity of this case study. Because either the case study is misrepresenting the time factor, the products involved, the problems with the non-MS Office products, or some combination of all of these.

Here are a few ideas about what I think really could have happened:

Scenario #1: Harian adopted StarOffice in the late 1990s, not OpenOffice.org. This seems plausible because StarOffice did had more problems with handling the Office formats back then. Sun had just acquired the technology from Star Division and wasn't putting a huge coding effort into it, other than making everything in the application purple.

But then why not say this was StarOffice in the first place? Well, for one, it blows any chance for the case study to use the "no vendor support" line of FUD. If Harian were using StarOffice, and claimed they couldn't get help, then everyone would ask "what about Sun?" Plus, and I think this very likely, why point fingers at new business buddy Sun when you can just what a poor, "defenseless" open source community?

Scenario #2: Harian adopted StarOffice in the late 1990s, then OpenOffice.org when it became available. That still accounts for the alleged format problems, and the discrepancy of the products' timelines. Still, if that were the case, then Harian's IT staff was so not on the ball. If they were really having trouble with StarOffice, why in the world would they re-deploy OpenOffice.org? Perhaps the problems weren't so bad after all?

Scenario #3: Harian adopted OpenOffice.org, but later than the "late 1990s." This makes a bit of sense, since one of the chief advantages cited for their initial switch was the fact that OpenOffice.org was free--something StarOffice wasn't for commercial deployments. But then, why the wrong date? You could argue a simple misrememberance, but this glaringly wrong fact makes the other facts in the case study that much more suspect. Though if you look at it, you will see a strong absence of facts regarding their reverse migration to Office.

For example, how much did Harian's move back to Office cost? Leow's only comment on this described a "small initial investment." Given that Office retails for around $300-$400 (depending on how you get it) and Harian has 31 stores and who-knows-how many desktops in their company, I question the use of the term "small."

Unless, of course, you count the big price break Harian probably got for participating in this case study.

Another question I have is, regardless of when Harian started using OpenOffice.org, the case study is quite explicit on the fact that they migrated back to Office in 2005. Even if they started with OpenOffice.org in 2000, that's five years of "suffering" with an office suite. I'm sorry, but if an application that was a major part of my business' productivity model was giving me that kind of difficulty, I'm sure I would not wait five years to make a switch back to what I was using. In fact, if Harian had tested OpenOffice.org on a trial basis and really, truly discovered these conversion problems, why the heck would they deploy company wide?

Finally, how accurately portrayed were these problems? Leow states: "Unlike Microsoft Excel, OpenOffice [sic] didn't have full functionalities for pivot tables, standard deviation, indexing and hyperlink functionalities." Really?

OpenOffice.org has had pivot table functionality since 1.1.2 (it's called data pilot), and I know you can do standard deviations. I know there's an INDEX function in Calc and, by golly, there seems to be an Insert|Hyperlink menu command in my version of Calc 2.0.2.

In the interests of fairness, I tried to contact Leow via e-mail, and posed these questions to him. As of press time, he has not replied.

This article is another jab at Microsoft's open source competition, one that is replete with misrepresentations and factual errors. IT customers should take away this article as a reminder of just how far Microsoft will go to hold on to their market share.