The Software Conspiracy: Maybe Minasi Is onto Something
Why is the quality of released software often low, with vendors releasing software that has known defects? Paul Thurrott explores users' acceptance of the situation as a primary factor.
August 7, 2000
Recently, I've written a lot about problems with software and hardware. My experiences with a Xircom USB card, for example, prompted me to write about problems installing hardware even on new systems today. Enough readers have recommended Symantec Ghost as the solution for me to try it. I'm devout about backing up my data, and keeping an up-to-date Ghost image should provide the other half of the restoration picture nicely. I'll let you know how the Symantec Ghost solution works.
Then came a related experience. I was browsing through Borders bookstore because a trip was coming up. I'm a voracious reader, so when I travel, I grab enough reading material to keep my mind off too-small airplane seats and miserable airport accommodations. As I perused the computer section, I came across something unexpected: "The Software Conspiracy," by Windows 2000 Magazine author Mark Minasi. (I write for the same magazine, but that connection has nothing to do with the fact that I highly recommend this book.) I meant to save the book for the trip, but—given what I had been thinking and writing about lately—I had to know what Mark had written.
Mark takes exception to the way software vendors describe what is essentially a mistake or defect. It's a lot easier to say that Windows Millennium Edition (Windows Me) ships with a few "known issues," for example, than it is to simply state that the product has mistakes that could cause data loss or an unresponsive computer. But the following paragraph from Mark's book describes what aggravates me the most about the state in which software is released:
"According to Cem Kenar, software quality and testing expert, 90 percent of the bugs consumers report to software vendors are already known to the vendors. Perhaps more amazingly, the vendors had full knowledge of those defects when they decided to ship the products. Believe it or not, fully 15 percent—one in seven—software firms surveyed about software quality said that they regularly shipped out software that they'd never even tested."
As I read those words, I recalled the numerous times I've discovered that Microsoft, for example, knew about bugs in its products before shipping them. One very recent example is Windows 2000 Service Pack 1 (SP1), which officially shipped last Monday. However, as is often the case, Microsoft quietly seeded the release to download servers, which eager users quickly discovered. I read somewhere that more than 64,000 people downloaded SP1 before Microsoft "officially" released it. Of those 64,000 people, more than 30 wrote to tell me that the download had destroyed their ability to access a local network or the Internet. In every case, the users were using a personal firewall product such as ZoneAlarm or BlackICE Defender.
I published an article about the early release of SP1 on the WinInfo Daily UPDATE Web site and later added information about SP1's personal firewall incompatibilities. That's when the email from testers started coming: By Sunday, I had half a dozen emails from SP1 beta testers. They informed me that Microsoft knew about the personal firewall problem months ago and refused to fix it. The company told testers that ZoneAlarm in particular used some undocumented feature of Windows that wasn't guaranteed not to change--and, no surprise here, Microsoft changed that feature in SP1. Mark Edwards, news editor of Security UPDATE, is currently investigating this issue, and I hope to know the full truth soon. Regardless of blame, however, Microsoft should have been more forthcoming with information so users could have known about the personal firewall problem before they installed SP1. After complaints about this issue, Microsoft finally added text about personal firewalls on the SP1 Web site. It was the least the company could do.
We should demand better quality from software. One reason Microsoft and other companies release badly flawed software is that we purchase it with little complaint. A similar situation developed in the US auto industry in the 1970s—and it took Japan's severe butt-kicking in the 1980s to turn that industry around. Where's Microsoft's "Japan" when we need it?
About the Author
You May Also Like