Sort of disclaimer

This is not a “legal” disclaimer, just an explanation of which technical and/or marketing information you may see in my posts here—and the information you won’t. Someone might want to convert it into a policy document for sites here. (But not my job.)

I subscribe to several technical journals and websites. You won’t see that information here unless it appears on a free-to-read/advertising-supported website. Even then, I reserve the right to ignore sites that repost information from behind paywalls. I don’t like paywalls. But I have to accept the site’s decision that they can’t survive on advertising alone. Many newspapers which make that decision are nice and let you read one article a month or so. There are other jerks, er aggressive sites that want you to subscribe for a year just to find out which news source actually provided the story, or that the headline is totally misleading. Of course, I don’t encourage them, and if I really want to refer to that story, I will try to find a better site/URL.

What about leakers? There seem to be three kinds of leakers. The first group, and usually worth paying attention to, are those who provide information from technical papers or presentations that are (or will be) in the conference proceedings for some meeting. The company involved has approved the publication and the conference usually benefits from being a source of such leaks, like HotChips. They state, “All HC33(2021) conference content is open to the public.” (Paper submission closes March 22 for HotChips34.) Converting what gets said and shown on slides into something you can act on is something I try to do.*

The second type of leakers are those who are trying to supplement their income by selling corporate secrets. I class those who are handing out résumés in this group. Normally the résumés circulated among those who might actually hire someone are usually harmless. In some areas everyone knows everyone else, and what they are working on, even if they don’t know the numbers. That sort of chatter is not recorded or written down, and I’ll follow that tradition here. As an example, I might know that Joe is working on a HARM (high-speed antiradiation missile) to be carried on fighter planes. If the project is not black (SAR Special Access Required), that level of knowledge is considered harmless. But there is a design spec somewhere with a classified appendix that has all sorts of interesting numbers. That knowledge is classified and leaking it can send you to jail.

What is the equivalent for the semiconductor industry? Yields are at the top of the list, wafers per hour and/or exposure times in DUV and EUV processes are up there also. Chip sizes, of course, leak as soon as some people have production samples. Gee, I just finished reading a lot of that stuff about ASMLs new 3600 series EUV scanners… And I started this note as a header for a chart of TSMC’s current and future processes. We will actually get there. It is now public.

Finally, some “leakers” royal pains who actually know nothing about the product, and just want to brag in a bar or to be thought of as a member of the group. If you can’t smell these people and rate their credibility correctly? You end up posting junk.

I try to stick to putting the pieces of the jigsaw puzzle together, but I have trouble with parsing the spoken words of lots of people. So I tend to rely on the written stuff. I should anyway. :wink:

  • I’m retired now, but when I was involved in an unclassified paper, it was usually as an editor who could check that the statistics were accurate and that no classified information leaked. Did I publish unclassified papers? A few. From two projects mostly, one to provide an interface between Cobol and Fortran, and the other to find various Doomdates for Y2K. Yes, there are other Doomdates. See https://en.wikipedia.org/wiki/Time_formatting_and_storage_bu… for a lot more Doomdates than I ever discovered. Unfortunately, many are there because they came into existence after January 1, 2000.
1 Like