Category: Journalism

Journalistic integrity

Robert McLaws at LonghornBlogs has a lengthy post this morning entitled Journalistic Integrity Revisited.

It says some complimentary things about me and some very unflattering things about Paul Thurrott.

I’d let it go, except that Paul decided to chime in with a comment to Robert’s post that takes a few cheap shots at me. I’m not a journalist, says Paul, along with a bunch of other name-calling.

Well, I guess I’ll have to give back my awards from the Computer Journalists Press Association [real-time fact-checking!] and the American Business Press Association. That’s OK, it was a pain in the ass finding a box to pack them in every time I move.

Anyway, I don’t care if someone wants to call me a journalist. I’m much more concerned with my reputation for accuracy and honesty. I don’t post as often here as some people would like, precisely because I try to make sure that what I post is truthful and accurate.

I’m pretty picky about whose RSS feeds and Web sites I read regularly. I like people who are passionate and willing to take risks. I have no patience for people who aren’t willing to give credit to others or to check their facts.

In his comment, Paul says, “I worked with Microsoft to uncover what happened here. I had used two sources for the original article I wrote, and so didn’t use terms like ‘according to reports…’ or similar, which generally accompany such things.”

Ahem. That’s not what my sources at Microsoft tell me. Two sources? Please. If that’s true, those people lied to Paul and he ought to delete their names from his address book. I have some pretty good sources at Microsoft too, and when I called, I couldn’t find a single person who had ever heard of anything remotely like this story. Because it’s a complete fabrication.

In fact, one contact at Microsoft yesterday complained that they had to call Paul to complain about the story and that he didn’t want to run a correction because it would be “embarrassing.”

Oh, Paul also says Robert Scoble isn’t a journalist either. “Frankly,” says Paul, “he hasn’t been at this very long.” [Robert McLaws says Paul was probably referring to him, not to Scoble. Probably right. Still laughable, as Robert M. has been building an excellent reputation online for three years now.]

Pardon me. I’ll need a minute to compose myself. And would you people in the blogosphere please stop laughing so loud?

Longhornblogs.com and the Scobleizer are both on my daily must-read list. I read Paul’s stuff whenever someone else links to it.

‘Nuff said.

Highway 61, Visited

There’s a wonderful piece in the Travel section of today’s New York Times. Called Highway 61, Visited, it’s the chronicle of a road trip along Highway 61, “the fabled Blues Highway that runs from the Mississippi Delta through Duluth, where [Bob] Dylan was born, and that Mr. Dylan mythologized in his 1965 masterpiece ‘Highway 61 Revisited.'” If you’re a Dylan fan, this story will be interesting:

One hundred miles to the northwest, in Fargo, N.D., a little-known but portentous moment in rock ‘n’ roll history occurred when, on the night after Buddy Holly died in a plane crash on the way to a gig there in February 1959, an 18-year-old piano player named Robert Zimmerman (later known to the world as Bob Dylan) sat in with a fellow Minnesotan, Bobby Vee, whose band was conscripted at the last minute to fill Holly’s place on the bill.

If you’re not a Dylan fan, but you love tales of real America, you still might like this fine little piece of journalism.

This is why we need independent sources

The normally reliable eWeek did a dreadful job with a story last week that highlighted a report from Webroot Software. The story has the alarming title Webroot: Spyware Rampant in the Enterprise. And sure enough, in the second sentence reporter Paul F. Roberts writes:

Webroot Software Inc.’s State of Spyware Report for the second quarter of 2005, claims that 80 percent of enterprise computers are infected with some kind of adware or spyware.

Meanwhile, in the story’s 15th graf, we read:

A new enterprise version of Spy Sweeper, which is being released Monday, will be able to detect and remove sophisticated spyware that changes the configuration of Windows systems and interacts with the operating system at a low level, said Brian Kellner, vice president of enterprise products at Webroot.

eWeek didn’t interview a single independent source for this story. It was essentially a press release for Webroot.

I’ve asked Webroot’s PR department to send me a copy of the report and will comment more after I see it. You can get one from their Web site, but you have to provide a lot of personal information, including company name and the number of computers in your organization. Why not make this important study freely available for download? Hmmm. It’s almost as though they’re building a mailing list they can use for sales calls.

Update: Webroot hasn’t gotten back to me yet, but Paul Roberts of eWeek was kind enough to send me a copy of the report. I flipped to the Enterprise SpyAudit section to break down that frightening 80% number. And sure enough, on page 36 is this gem: “…cookies tend to make up the largest number of infections per enterprise machine.” Cookies! As I’ve written before, cookies are not spyware. In my opinion, Webroot is totally wrong to claim, that a computer containing one or more tracking cookies is “infected with spyware.” Ironically, Webroot even acknowledges this fact in a sentence buried at the end of the section (page 40): “Webroot will continue to monitor cookies until a definitive decision on whether cookies constitute spyware is determined.”

Meanwhile, there actually are some frightening statistics in that report, including the observation that 7 percent of the 60,000 enterprise PCs in their sample were infected with malicious spyware, which they define as “system monitors and Trojans.” If that data point is accurate (a point I’m not willing to take at face value, given the report’s willingness to exaggerate in other areas), it’s cause for great alarm. Even one such program is too high for comfort on any corporate network.

A fine job of debunking bogus quotes

Kevin Maney of USA Today deserves a medal for this article, which looks at a half-dozen widely circulated quotes from titans of the tech industry going back more than a century and finds that only one of them is actually legitimate:

Among the quotes is this widely circulated comment attributed to Thomas Watson, builder of IBM, in 1943: “I think there is a world market for maybe five computers.”

Except it’s doubtful Watson ever said such a thing.

I know this because I wrote a book about Watson. But it also got me thinking: There are about a half-dozen similar quotes that tech people use all the time. These quotes pop up in speeches, on posters, in PowerPoints, during sales talks and in pitches to raise money. They’ve practically become articles of faith in the industry.

But how many of them are real?

As it turns out, only one.

Definitely worth reading and bookmarking, so you can reply next time someone tells you that Bill Gates once said 640K of memory is enough for anyone. (Nope, he never said it.)

Orlowski’s a liar, too

I still have no intention of linking to Andrew Orlowski. But Thomas Hawk did, and he quotes from the latest installment in Andrew’s public feud with Robert Scoble. Orlowski prints a damaging e-mail supposedly sent by Scoble, and he goes on to slur Scoble’s reputation and suggest that Microsoft should muzzle him. Trouble is, the text of the e-mail that Orlowski printed doesn’t match what Scoble sent. I’ve seen the original, and Orlowski’s version contains sentences that Robert never wrote.  As Thomas writes:

Now getting a source who gives you bad information is one thing, but making up emails is something else entirely. If I were the Register I’d investigate this and if it turns out Orlowski fabricated an email then he should be fired.

It’s one thing to make up quotes, but e-mails that go through corporate servers leave lots and lots of traces, all of them monitored by lawyers who are going to safeguard them as if they were about to be subpoenaed in an antitrust trial. It will  be very easy to prove who’s telling the truth, and my money’s on Scoble, who has been scrupulously honest (if occasionally overenthusiastic) in all my dealings with him. Especially when the other guy makes shit up every day just for fun.

Seriously, folks, reasonable people can disagree about all sorts of things. They can yell and scream and question the other guy’s motives and even spin conspiracy theories. But if you start by distorting the available facts and then you make shit up on top of that, everyone loses.

One of these guys is a liar. I firmly believe Robert Scoble is telling the truth

In fact, Andrew may have stepped in a bigger pile than he bargained on this time around. Scoble’s professional reputation would be damaged severely with an accusation that he knowingly lied. If that statement turned out to be untrue and it was printed with malice or with reckless disregard for the truth (which kind of sums up Andrew’s working style)… Well, you’ve got all the ingredients for a libel suit that would have any lawyer grinning from ear to ear.

Couldn’t happen to a nicer hack.

More Orlowski hackery here . Counting up the A.O. errors is a Sisyphean task. Or, more accurately, I suppose, Augean.

Hackery, continued

Charles Arthur says he tried to post a comment here in defense of The Register’s Andrew Orlowski and his trash-talking about IE7 and Robert Scoble and “got denied for having ‘questionable content’ by the comments filter on Ed Bott’s blog.”

Sorry, Charles. I just checked, and apparently in the process of cleaning out comment spam yesterday (I get hundreds or thousands of attempts per day), I added a blank filter to the MT-Blacklist database, which means that anything posted by anyone was considered questionable. It made for a very enjoyable day for me, from an administrative standpoint, but that was probably a little too aggressive. It should be fixed now. For everyone else, here’s what Charles wanted to say in response to this post:

In fairness – or perhaps, better, precision – the top sentence says “some users”. Not all. So to say, Ed, that because you’re seeing it means that everyone is seeing it falls into what one might call “observer’s syllogism” (if it’s true here, it must be true everywhere).


Also, Robert Scoble saw this problem himself. And the story has been updated, twice, once with a link to Scoble’s blog.


(Disclosure: I write sometimes for The Register; I’ve worked with Andrew Orlowski on some stories, notably about online music stores’ cut of business: his figures were correct.)


Also, I’m never sure if “hack” is such a powerful insult for a journalist. It’s got a long and proud history in the UK.


First of all, Charles, IE7 is a developer’s beta. Of course it’s going to have bugs. (I’ve got a list of a dozen so far and I haven’t even been running it for 24 hours.) But it is disingenuous to claim that Andrew’s story simply mentioned “some users.” Go back and read it. Allow me to quote the part after that first sentence, with liberal use of emphasis to point out the really hackish parts:



Microsoft’s Internet Explorer 7 went on a limited beta release today and contains a nasty surprise for some users.


Users with search toolbars from Yahoo! and arch-rival Google have discovered that these vanish. [No qualifiers. Not “some users” – all of them. As Jeremy Mazner points out, “his implication is that the ‘some users’ who experience the nasty surprise == all users with search toolbars.” – Ed] Other third-party toolbars designed to block pop-ups or aid with form filling appear to be working normally, according to reports from Reg readers.click here


The default search engine is MSN Search.


There are sound compatibility reasons for Microsoft disabling third-party toolbars in an early cut of the software. [That evil Microsoft disabled these features deliberately! – Ed] 


The implication of Andrew’s story is that Microsoft deliberately or negligently blocked IE7 from working with software from two rivals. It’s the modern equivalent of the old “DOS ain’t done till Lotus won’t run” canard. Which also wasn’t true.


Finally, Andrew makes the bizarre assertion that Scoble saw this bug himself and then removed traces of his “confession.” I read every post in this exchange and never saw that. In a classic Orlowski move, when someone accuses you of making shit up, the response is … to make more shit up! Andrew claims to have seen this admission (no actual quote, mind you, just a bald assertion), and a decent journalist like Charles Arthur picks it up and amplifies it.


As for the observer’s syllogism… The fact that Charles once saw Andrew Orlowski write a story containing actual facts doesn’t mean that he does that regularly. Quite the contrary.


Oh, and on this side of the pond, a hack is “a mediocre and disdained writer .” Exactly.


Updated to fix some typos and add some context for anyone coming to the story late.

Andrew Orlowski is a hack

Robert Scoble points to yet another grossly inaccurate story from the keyboard of The Register’s Andrew Orlowski (no, I won’t link to it – go to Scoble’s site and follow the link if you must). Orlowski “reported” that the IE7 beta out yesterday doesn’t work with the Google or Yahoo toolbars. Actually, the headline was even more inflammatory: “IE7 nukes Google, Yahoo! Search.”

That’s complete nonsense, as Andrew could have found out had he actually done any reporting or testing, but it didn’t stop him from publishing a pack of lies. Scoble assembles a small mountain of documentation to prove that this story is categorically false (and I can attest to the fact that the Google toolbar works just fine on IE7, because I’m looking at it right here). He then says:

I wonder if Andrew Orlowski will link to my blog and correct his story because his report is HUGELY damaging here.

Sorry, Robert. Andrew Orlowski doesn’t correct anything. He got this story wrong, and even a legitimate journalist like Dan Gillmor got fooled into reusing Orlowski’s distorted report. Neither he nor the Register ever corrected it.

Lawrence Lessig caught Orlowski in another whopper. The original story is still up, with a sorta-kinda-not-really-a-correction-but-an-editorial-note at the very end that just muddies the waters.

Amusingly, I used the Google toolbar in IE7, as well as the Yahoo! Search capabilities also built into IE7 to find all these pages.

Yep. Andrew Orlowski is a hack. Don’t hold your breath waiting for a correction, Robert.

Boing Boing gets a big shovel, spreads BS

Follow the bouncing distortion:

It starts on CNET, with a story that quotes anonymous “sources” saying Microsoft is “in discussions to buy controversial adware maker Claria.” (The New York Times runs a similar story later the same day, with a few more details, including a quote from its unnamed source saying, “Those in favor of the deal … believe Microsoft could help clean up the adware field,” but that the talks are falling apart.)

Then the story goes to Slashdot, where the reference to “sources” is dropped.

A week later, another Slashdot story begins “A week after word leaked out that Microsoft was negotiating an acquisition deal with Claria…” Leaked out? All of a sudden the anonymous sources are confirmed?

Boing Boing, consistently rated in the Top 10 among sites in Technorati, picks up the ball and spikes it in the end zone in the Ultimate Fantasy Bowl:

MSFT acquires spyware firm, changes antispyware app to ignore its products
Microsoft recently acquired a spyware company called Claria (known for its spyware product, Gator). They have since updated Windows’ antispyware app so that it advises users to ignore Gator spyware.

So the rumor became a confirmed story and now it’s a done deal. Not to mention that the change in the status of Claria’s products happened as much as four months ago. Back in the reality-based community, the whole story’s still just a rumor based on a pair of stories from unnamed sources, but science fiction writer Cory Doctorow, who authored the Boing Boing piece, apparently has a news feed from a parallel universe.

And to compound the error, Prof. Brad DeLong picks it up and calls Microsoft a “company behaving very badly,” because if you read it on the Internets, it must be true.

As the good Professor might say, “Why oh why can’t we have a better blogger corps?”

Update, 7/9: Boing Boing has printed a “correction” that strikes out the words recently acquired and replaces them with is rumored to be acquiring. The headline now reads MSFT acquiring spyware firm. That’s wrong. The CNET and NY Times stories say that Microsoft is “in talks” to acquire Claria and that the deal is far from a sure thing.

Even worse, Boing Boing has left the original story up, uncorrected. See the original here, and the changed version here. If I use a search engine, I have a 50-50 chance of getting the original, bollixed-up version. Sloppy, sloppy.

And the sentence that follows the “corrected” one now makes no sense at all: Boing Boing says “[Microsoft has] since updated Windows’ antispyware app so that it advises users to ignore Gator spyware.” Since when? The implication is that this was a quid pro quo, but the change in detection status for Claria’s applications was made earlier this year (as Donna Buenaventura reported), and it affected other adware companies as well. The New York Times story says the talks between Claria and Microsoft started a few weeks ago. If there’s a quid pro quo, the timeline doesn’t support it.

The real story is that Microsoft has decided that high-profile adware makers who achieve a minimum threshold of disclosure (including Claria and WhenU) will be able to get an “Ignore” rating. Reasonable people can argue that that’s a bad decision, but the Boing Boing story doesn’t do that. It tries to create a story of corruption where none exists.

Update 7/10: The uncorrected Boing Boing story is still there. Meanwhile, if you want to know more about Microsoft’s decision to change the classification of Claria’s adware, see my follow-up: Dear Microsoft: Why should we trust you to detect spyware?

Update 7/12 6:00 a.m.: As of this morning, the original, uncorrected post appears to have been deleted from Boing Boing’s servers. The “corrected” version is still online and still wrong.

Update 7/12, 6:00 p.m. PDT: No deal. Boing Boing still says “MSFT acquiring spyware firm…”

Update 11/13/2006: The original story has never been corrected.

Today’s clinic in bad journalism

Here’s why I don’t trust or recommend BetaNews. Nate Mook of BetaNews writes an outrageously bad lede to a story on the Google toolbar this morning:

In a marketing tactic used primarily by spyware and adware companies, Google has begun bundling its Google Toolbar and Desktop Search software with the popular WinZip archive utility. The move comes as Google begins to expand its bundling effort with a number of leading software applications. [emphasis added]

WinZip quietly updated its download executable last week, which now weighs in at close to 4MB with the added Google tools. Users are given the opportunity to opt out of installing the Google software on the first WinZip setup screen; by default the tools are installed.

Good lord, I can’t even begin to count how many unwarranted innuendos are in this story. For starters, when you kick off the story by comparing the two companies involved to spyware and adware companies, you create an impression in the reader’s mind that is difficult to overcome. Mike at Techdirt calls bullshit on this story:

[T]he details don’t support the charge. It’s clearly displayed in the setup screen and it lets people choose not to install the bundled apps. Also, the purpose of the apps aren’t obfuscated by misleading language. It’s not a spyware tactic by any means — but the fact that so many spyware offerings use similar, if more underhanded, tactics means that such bundling is always going to be looked at suspiciously.

Here’s the screen that pops up during the installation of the evaluation version of WinZip 9.0:

Google__winzip_setup

Now, it’s true that you might see a similar-looking dialog box when you install a program that bundles spyware or adware. Except in those cases the intent is typically to mislead, and the bundled software often performs functions (such as displaying pop-up ads) that are not disclosed or are hidden in a license agreement deliberately written to confuse. For that matter, many bundled spyware and adware programs are downloaded without the user’s consent. But none of that is true here. This isn’t spyware. It does a good job of providing disclosure and asking consent (although I’d prefer that the default be off with the user being required to click yes or no to the installation of these two items). The bundled software doesn’t do anything that is remotely like spyware or adware. But none of those details are in the BetaNews story.

It gets worse. Here are a few more examples of loaded phrases, later in the story:

Google did not respond by press time to inquires about whether such a distribution approach could be seen as questionable from a company that boasts its ability to “make money without doing evil.” …

Unlike its rivals, Google does not have the ability to push its search tools via established software products…

But Google isn’t the only company resorting to bundling. Yahoo recently inked a deal to offer its Yahoo! Toolbar with Macromedia Flash and Adobe’s Acrobat Reader – both essential downloads for most computer users.

Let’s review: According to BetaNews, Google is “pushing” its software through “a marketing tactic used primarily by spyware and adware companies.” And to compete with Microsoft and Yahoo it has had to “resort” to this desperate bundling strategy, despite its “boasts” of not being evil. Can you blame Google for not responding when BetaNews asked them an obviously loaded question?

This is a great opportunity to discuss the nature of software bundling. As Techdirt notes, there’s a lot of room for confusion when bundling is involved. Users who have been trained to be suspicious of every unsolicited offer (for good reason) should be suspicious here. But a good journalist provides information that can help the reader figure out the real story, not throw more suspicion into the mix through sloppy reporting and inflammatory language.

And if a journalist is going to accuse a company of using unsavory tactics, it helps if they review their own standards and practices first. BetaNews uses Google AdSense ads (full disclosure: so do I). Ironically, this story is accompanied by a Google AdSense ad that links to Hotbar.com, which is identified in the Computer Associates Spyware Information Center as adware that includes a downloader and a search hijacker. (If that ad happens to appear on this page, I apologize. Please don’t click it.) Here’s a screen capture from the BetaNews story.

Google_ad_betanews

By the way, BetaNews uses a technology called IntelliTXT from Vibrant Media, which places ads directly in editorial content. This story contains a reference to Google’s rival MSN Messenger, which in turn links to an IntelliTXT ad that offers (I could have said “pushes”) Microsoft’s LiveMeeting software. I found this story through a link at The Office Weblog. I think Jason Calacanis, who owns the network that includes that blog, had the best description of this advertising technique: “The only publishers that will use this software are a) desperate ones, b) ones without ethics or c) people who make a mistake.” Ouch!