I ended my recent posting on Ethernet connections in a hotel room not being secure with "You don't read PC magazine for mutual fund advice, and you shouldn't read the Wall Street Journal for computer advice." The Cranky Geeks did a show earlier this month about technology journalism that made some interesting points along the same line.
The show is hosted by John C. Dvorak, the guests were John Markoff, Gregg Zachary and Sebastian Rupley. The show ran 30 minutes, I found the first half more interesting than the second, your mileage may vary.
Some tidbits from the show:
- Most tech journalists are PR people, rather than journalists.
- Tech journalists have very little independence.
- There is a big lack of historical awareness by tech journalists (CNET took a hit here).
- Intel created the technology coverage at the Wall Street Journal.
- Industry sets the news agenda in technology.
- Technology is in the Business rather than Science section of newspapers because that's where the ads are. As a result, tech coverage tends to follow the money rather than the technology.
- Markoff noted that his feature stories are now 800 words whereas they used to be 1,000 to 1,200 words. This creates a dis-incentive to reporting.
Going from the general to the specific:
No one from CNET has ever indicated to me, in any way, that they care about the number of words in a posting. In that respect, us CNET bloggers (and most bloggers I suspect) are lucky, we can use as many or as few words as it takes to make our points, without concern for things like square inches in a dead tree.
This blog is not influenced by PR people. Perhaps there is something to be said for flying under the radar.
CNET offers their bloggers total independence, members of the CNET blog network are not even CNET employees.
Finally, I cover the technology, not the business side of things and have more historical awareness than I care to admit to. :-)
This Defensive Computing blog isn't perfect, heck it's a part-time gig, but at least it doesn't suffer from the above problems.
See a summary of all my Defensive Computing postings.
Like so much else on the Internet, you have to be skeptical about the star ratings of software. Perhaps you suspected this, but now there is proof.
A software developer in the U.K., Andy Brice, was suspicious about the ratings assigned to his software, so he did a test--a lab experiment if you will. He started with a plain text file that said "this program does nothing at all" a few times. Then he renamed the file so that it ended with ".exe" and submitted it to 1,033 download sites. The "program," if you can call it that, won't even run.
Being as obvious as he possibly could, Andy called the program "awardmestars" and included a description of the program that said, "This software does nothing at all." He even included a screenshot that said very plainly that the software does nothing. See his blog for the full details: The software awards scam.
Andy says his nonfunctional software was listed on 218 Web sites, and some even gave him an award. "Approximately 7 percent of the sites that listed the software e-mailed me that it had won an award," he said. His submission was rejected by 421 Web sites, but since he listed it as a utility, many of these rejections were because the site didn't include that type of software. Many submissions are still pending.
Since a picture is worth a thousand words, take a look at a screenshot of awardmestars version 1.0 at Topshareware.com where it was certified as having no spyware, adware or viruses. The user reviews are hilarious. PC World magazine listed it originally, but has since withdrawn their listing. As I write this, however, the listing at PC World as of August 15, 2007 at 17:01:08 GMT is still available in the Google cache.
Trustworthy software downloads
Andy mentioned three Web sites where a human being obviously reviewed the software because they wrote back to him, either appreciating the joke or being annoyed by it. The sites were Filecart.com, Freshmeat.net and Download-tipp.de. He considers the fact that a human responded to him sufficient to recommend these sites. I consider it just the first step.
In his Security Fix column in the Washington Post, Brian Krebs wrote about this today (Beware of Five-Star Vaporware) and concluded with " ... I've never strayed far beyond a handful of sites that I have come to know fairly well, such as CNET's Download.com, SourceForge.net and Tucows.com."
If you want to judge CNET's Download.com Web site, which I trusted for years before having any involvement with the company, then see:
- CNET Download: How we test and rate
- How we test for adware and spyware
Here is a quote from the first page above:
"In addition to screening for common viruses and spyware, we look for other threats that might interfere with our users' security, privacy, and control. When evaluating a submission, we consider publisher Web sites, publisher conduct, and our own experience with a particular product."
It's a cruel world out there.
- prev
- 1
- next
