Letter to the Editor: Probe Cost Methodology

I am writing in regard to the article published in the April edition of Semiconductor International titled, "Practical Ways to Lower Probe Cost." The article deals with the supposed cost advantage of using Wentworth Laboratories' Accumax technology for probing bumped die vs. other technologies.

Although I have technical issues with the content, the primary issue I have with this article is that the author is listed at the head of the article as representing Infrastructure Advisors of Saratoga, Calif. The author's bio at the end of the article begins with his being president of Infrastructure Advisors and how they are "an independent analysis and consulting firm." The clear impression given is that the author is an independent evaluator of technologies with no personal interest in the technologies presented. In the last sentence of the bio it is noted, "He is a member of the board of directors of Wentworth Laboratories." This leads me to believe the byline at the head of the article is disingenuous. As the author has a clear personal interest in the matter presented, it should have been made clear to the reader that this was the case up front, so that the reader has knowledge of this as he or she evaluates what is presented.

I respectfully assume that SI has some editorial responsibility for what is published in the magazine. When data is published that compares products, there must be some validation of this data prior to its being published. As this can affect the business decisions made by your readers, it is reasonable to expect that data considered for publishing is verified in some way, at least to the point of having knowledge that such data exists. In the article, the data is presented as fact, with the stated validation that, "These values are based on actual user experiences." In correspondence with the author, he states, "It was written as a high-level viewpoint article and not a clinical study of benchmark data, which would have required much more exhaustive data collection. The main message was to point out that procurement costs are not necessarily the dominant factors in CoO, and this is backed up by the ISMI model based on the assumptions used." So, in essence, the author admits that the presentation of such data is not justified by the less vigorous methods he used since this was not a benchmark study. And, yet, the data is published and, in point of fact, the reader is left with the impression that this is a benchmark study. Indeed, the company for which the author is on the board of directors is presenting it currently to customers as a benchmark study. And they are bolstered by it being published under the SI banner, which validates their assertions by having the readers rely on SI's editorial standards. The article is clearly a comparison of the three different technologies and is stated as so in the table of contents, where it is even referred to as an analysis of "real-world scenarios." It is feared that the reality is stated but may not have been verified in any meaningful way.

I was told that SI reviews articles for:

• value to the readers
• potential interest of the readers
• quality of the writing
• whether the article was previously published

I was left with the understanding that SI does not have the technical expertise to validate data presented and does not verify data underlying the statements made in published articles.

Joe Baker, ICT, Email: JoeBaker@ICT-Probe.com

Editor's Note: Semiconductor International has replaced the byline on this article, which now reads, "Ron Leckie, Advisory Board Member, Wentworth Laboratories." SI stands by the publication of the article.