Van Smith assaults John Dvorak's column

This is a response to John C. Dvorak's latest "Inside Track", where he takes a number of pot shots at our now defunct Van's Hardware in general and the BAPCo SysMark 2002 article here in particular. Mr. Dvorak's article appears in the October 15th issue of PC Magazine.

Unfortunately, Mr. Dvorak did not research his article well. In it he claims that John Markoff of the New York Times states that no one has been able to get any details regarding what threats were made to us that influenced our decision to close down our site.

Mr. Markoff never asked me specifically about our site, but did inquire specifically about the BAPCo SysMark controversy. Mr. Markoff also later contacted me regarding Intel Itanium performance characteristics. I am sure if Mr. Dvorak had taken the time to question Mr. Markoff scrupulously, he would have avoided his embarrassing mischaracterisation of a fellow journalist.

As for the threats themselves, some of my articles have been provocative, even at times being attributed by mainstream news sources as undermining the stock prices of certain companies. Although most of the email I have gotten has been very supportive, I have also received a steady stream of hate mail from a vocal and typically anonymous minority. The severity of some of these responses border on pathological.

Even though some of these messages involved threats, I did not initially pay much attention to them. In fact, one man who threatened me also threatened the inquirer's Mike Magee around the same time several years ago, but at that point I still did not take these threats seriously.

However, that changed when our house was vandalised for a second time. This crime followed closely on the heels of a particularly large number of flames triggered by my work at InQuest. We reported this offence to the police who took note of the surrounding events and advised us to keep track of future menacing emails and phone calls.

Recently, I received a threat from an anonymous individual who called himself my "worst nightmare" and who promised that he would "introduce" himself to me when I least expected it, whether it be this year or further into the future. In the weeks preceding this note, we had been bombarded around the clock by an almost constant stream of hang-up phone calls. However, within days of receiving the email from my "worst nightmare" a man called our house and spoke briefly to our five-year-old daughter before she gave the phone to my wife, Kathy, who hung up on him, but only after he was able to say things that made it clear his intention was unhinge my family while tying his motivations to the work on our website.

Since then, as I have confided to our friends, Kathy has felt unsafe in our own home.

Keeping up our site was simply not worth it. Since we have closed it down, the number of flames arriving in my inbox has slowly diminished and the phone calls have ceased completely.

Returning to Mr. Dvorak's article, he mystifyingly attempted to dismiss AMD's arguments for pro-Intel bias in SysMark2002 through the diversionary strategy of labelling our site, which first reported AMD's accusations, as pro-AMD. This claim is neither true nor relevant even if it was.

Mr. Dvorak then threw out a red herring involving our protest regarding a specific ExtremeTech article which we believed had harmed the enthusiast community. Mr. Dvorak made no mention of the specifics of that issue as well. For background on that topic, here is a reply to ExtremeTech's Editor-in-Chief Bill Machrone I posted at the time of the controversy:

"Bill,

"AMD has established its New Product Review Program to allow a broad spectrum of media participants to preview upcoming products. Through NPRP, AMD provides CPUs and occasionally entire systems to reviewers and analysts, big and small. AMD also grants access to extensive information, under trust, so that derived reviews can be accurate and interesting.

"As such, NPRP is invaluable particularly to the smaller enthusiast sites that do not have the financial resources to procure these products in a timely fashion - if at all.

"Bill, as I am sure you are aware, AMD issued a note earlier today that indicated clearly that your organization's actions - and as Editor-in-Chief, *your* actions - have placed the future of NPRP as it exists today in jeopardy. If NPRP is redrawn so that AMD no longer releases hardware and related information beforehand then it will cripple the smaller sites.

"Your decision to publish information from a copy of an email sent to your own company under NDA has also served to embarrass good, decent people inside AMD who have been nothing but earnest, forthright and professional with the enthusiast community.

"As Editor-in-Chief of your organization, I suppose you closed your eyes when this news story went up, even though the note it was derived from was clearly marked as under embargo. Furthermore, I'm sure you knew -- or certainly could have found out with little effort -- that your own organization had this information and gained it through trust. Your organization violated that trust. *You* violated that trust.

"We get NDA material everyday involving products we are previewing, but we would never publish this information regardless who in our organization got the tip - be they involved with the NDA or not.

"Why? Because violating the intent of an agreement made under trust is wrong.

"I find your excuse to be hollow and disheartening.

Van Smith
Editor-in-Chief
Van's Hardware Journal

ExtermeTech is an Internet-only sister publication of Mr. Dvorak's PC Magazine. As reported by Cnet and, more recently, atnewyork.com ( here ), ExtremeTech and its ZD Media parent have encountered financial problems.

Mr. Dvorak, you would be doing your readers a greater service by simply stating the facts surrounding issues that you discuss and letting these readers come to their own conclusions. Instead you have chosen to completely ignore details while pontificating to your readership about what is and is not ethical.

Later in his curiously rambling article, Mr. Dvorak proceeds to label as "ludicrous" the notion that Intel could strong-arm journalists even if it wanted to. This dismissal is in reference to an article ( here ) appearing on our site several months ago, and is yet another attempt by Mr. Dvorak to divert the reader's attention from the central issue involving AMD's SysMark2002 accusations. And yet while he meanders through each detour, Mr. Dvorak continues to avoid the address any details.

For many, it is no secret that Intel will utilise whatever leverage it possesses in order to manipulate journalists. In addition to the information presented in our article, we have a number of emails clearly demonstrating some of the tactics the chipmaker will exercise.

Similar information ( here) has been disclosed several times on the pages of the inquirer.

Lastly, Mr. Dvorak returns to the BAPCo SysMark issue by attempting to dismiss the importance of benchmarks altogether, a notion that he tried to pass off on Mr. Markoff.

Either Mr. Dvorak is being deliberately disingenuous, or he has a shockingly misguided understanding of the dynamics of the computer industry. MPU products are largely differentiated and priced according to perceived performance advantages; that is why Intel can charge more for a 2.8GHz Pentium 4 than 1.7GHz Celeron, or why nearly all microprocessor manufacturers are constantly pushing their technologies to the limits in order to gain performance advantages over their rivals.

In the all-important high-end markets where chip producers are able to buoy profit margins and ASPs (average selling prices), the business is nearly all about perceived performance. This performance is typically measured and demonstrated ( here) through benchmarking.

As well they should, the highly influential collection of enthusiast hardware sites almost completely focuses on benchmark results. Of the collection of benchmarks commonly employed, BAPCo's SysMark is probably the one most commonly used to pronounce performance superiority on common applications.

According to AMD, SysMark2002 has potential bias that favors Intel's Pentium 4 processors which could significantly undermine the possibility of objective reviews, and innately and misleadingly skew assessments towards the Intel P4.

Having conducted many reviews myself, it is fundamentally important to be able to trust the tools you are using to provide accurate analyses. According to AMD's accusations, one of the most important benchmarks in use today appears to deploy clear, calculated measures explicitly to benefit one microprocessor architecture while undermining any possibility of the tool's objectivity as a serious application performance metric.

It saddens me to see Mr. Dvorak stoop to such a "hack job" as this version of "Inside Track" clearly is - and a poorly written hack job at that (I am still befuddled by his reference to IBM). I remember enjoying Mr. Dvorak's work digging up scuttlebutt on the back pages of InfoWorld roughly twenty years ago. Times have changed. Print media has been devalued as the Internet has taken over. Unfortunately, the quality of at least some print media seems to have suffered a similar decline.

But let none of these diversions take your eyes away from the central issue involving AMD's accusations regarding SysMark2002. The chipmaker's arguments are simple and compelling. Although we are dismantling our site, AMD's presentation ( here) remains downloadable from there. Read it and decide for yourself.

Van Smith