Intel in hot water again, interesting reading! - test-drb@ccmp.vtda.org

12 Feb 1999

The solution's simple . . . don't use an intel processor.  Goodness knows
they're overpriced!
Has AMD started doing this?
Dick
----------
...
  From: Joe &lt;rigdonj(a)intellistar.net&gt;
 To: Discussion re-collecting of classic computers
&lt;classiccmp(a)u.washington.edu&gt;
...
  Subject: OT:  Intel in hot water again, interesting
reading!
 Date: Thursday, February 11, 1999 4:48 PM
 FYI
 >
 > <http://img.cmpnet.com/tw/newsletters/scoop-top.gif> The Scoop
 > http://img.cmpnet.com/tw/newsletters/blank.gif
 ><http://img.cmpnet.com/tw/newsletters/blank.gif>
 >http://img.cmpnet.com/tw/newsletters/blank.gif
 ><http://img.cmpnet.com/tw/newsletters/blank.gif>
 >http://img.cmpnet.com/tw/newsletters/blank.gif
 ><http://img.cmpnet.com/tw/newsletters/blank.gif>
 >By Fred Langa
 >By Fred Langa
 >
 >InformationWeek
 >
 >You probably saw the original coverage of Intel's announcement that it
would
...
  >embed an individual serial number in each Pentium
III and Celeron chip. The
...
  >96-bit ID can identify the user's PC to any
software that knows how to ask.
...
  >
 >Immediately after the announcement, various consumer watchdog groups went
...
  >ballistic. Epic, the Electronic Privacy
Information Center, launched a
 >boycott of Intel, calling it the "Big Brother Inside" campaign. Epic
says
...
  >the processor serial number, "would likely be
collected by many sites,
 >indexed and accumulated in databases...The records of many different
 >companies could be joined without the user's knowledge or consent to
provide
...
  >an intrusive profile of activity on the
computer."
 >
 >Intel immediately backed off a bit by announcing that although the serial
...
  >number would ship enabled on every chip, Intel
would provide equipment
 >manufacturers with a small software applet that could be used to prevent
 >access to the number. However, the software must work (it hasn't been
tested
...
  >yet); it must be properly installed on each PC;
and it must be run after
 >every reboot.
 >
 >Epic says that because this approach "relies on a software patch that must
...
  >run each and every time that a user turns on the
computer, it is susceptible
...
  >to tampering by other software programs." So,
Epic's boycott is still in
 >place: The group insists that Intel should disable the processor serial
 >number at the hardware level, where it will stay disabled until the PC owner
...
  >turns it on.
 >
 >To further muddy the waters, the processor serial number may not be very
 >secure. CMP Media's Electronic Engineering Times quoted cryptography expert
...
  >Bruce Schneier, who talked about the prospect that
the serial numbers can be
...
  >forged or stolen: "A system is only as secure
as the smartest hacker," he
...
  >said. "All it takes is for one person to
defeat the tamper resistance.
 >There's always someone who manages to unravel the protection. There
isn't a
...
  >copy-protected piece of software that hasn't
been stripped of its
 >protections and posted to hacker bulletin boards. This won't be any
 >different." (For the full story, go to "Intel ID Protection Scheme
Called
...
  >Insufficient.")
 >
 >Of course, there are legitimate and useful purposes for this kind of ID,
 >especially for resource-tracking within an enterprise. Indeed, some
 >workstation manufacturers already include similar functions on their
 >enterprise-ready boxes, and some enterprise software products use these
 >serial numbers for licensing. But Intel is attempting to broaden this
 >practice to an unprecedented degree by putting the ID number on every chip
...
  >and enabling it by default. Toss in only weak
assurances of the serial
 >number's security and only a weak turn-off option, and you're got a
 >firestorm of protests.
 >
 >Last week, I conducted an informal online poll among the readers of Windows
...
  >Magazine. The reaction was eye-opening: Out of
hundreds of posts, virtually
...
  >all were vehemently anti-Intel. And in that huge
majority, most people swore
...
  >their next PC purchase would be AMD-based, until
and unless Intel either
 >removes the processor serial number or allows it to be disabled in hardware.
...
  >One reader suggested the clever idea of
resurrecting the old "turbo" switch
...
  >approach and placing a simple serial number
enable/disable button on the
 >front of every PC. (You can read more on the controversy and see reader
 >reaction at Windows Magazine: Big Brother Inside?.)
 >
 >I was amazed at the absolute intensity of the reader posts. It's as though
...
  >the processor serial number was the last straw for
many people: Intel's
 >history of high prices and other public relations fumbles (like the
 >floating-point math bug) seem to have built up a huge reservoir of
 >resentment that's now spilling over. I think we're seeing the start of a
 >strong anti-Intel backlash, analogous to the anti-Microsoft fervor that's
...
  >changing the operating system landscape.
 >
 >Fred Langa is a senior consulting editor and columnist for Windows Magazine.
...
  >Fred's free weekly newsletter is available via
subscribe(a)langa.com
 ><mailto:subscribe@langa.com> . You can contact him at fred(a)langa.com
 ><mailto:fred@langa.com>  or via his website at http://www.langa.com
 ><http://www.langa.com> . http://www.techweb.com/
<http://www.techweb.com/>
...
  >