A relative called the other day, asking me to come over and fix her DUN. She'd gone with a new ISP, and didn't have the DNS settings right.

Before I could get there, she called Tiger (where she bought her computer) and they gave her "Technical Support" over the phone.

You can probably guess what they told her to do. Yup, the same thing that newbies get told to do when they wander into a rough neighborhood on the Internet: they had her format C:.

Of course, "format it" seems to be what passes for "tech support" these days. If the vendors were pediatricians, instead of saying "give the kid two aspirins and call me in the morning", they'd say "shoot your kid in the head then go get pregnant again".

Anyway, ask yourself what will happen when millions of folks -- who are running systems that are *only* compliant due to *patches* and *updates* they applied *over* the stock system on the CD -- get told to "format it and reinstall everything" *after* 12/31/99.

How will a non-complaint OS take to being *installed* in 2000+?

Hell if I know, but I wouldn't bet on it being pretty.

-- Ron Schwarz (rs@clubvb.com.delete.this), December 24, 1999

Answers

Ron,

You've got a good point ... but you're assuming that people will actually be able to get through to the "technical support" people during the first few days or weeks of January. Can you imagine what it's going to be like at Microsoft's help desk on Mon, Jan 3rd? I don't know whether it would be better for those help-desk folks to be on Valium, Prozac, or Jolt Cola.

Ed

-- Ed Yourdon (ed@yourdon.com), December 24, 1999.

#1 call in question to tech support (no joke/research proven)--

"Uh, where's the "ANY" key? Been lookin for hours, and can't find an any key...Please help."

-- Hokie (nn@va.com), December 24, 1999.

Mt Dew would be my choice.

-- Butt Nugget (catsbutt@umailme.com), December 25, 1999.

I heard on CNN that NASA has replaced the Hubble Space Telescope's CPU with an Intel 486 chip. The reason given for using should old chip is that it takes them so long to test and certify the CPU as being able to with stand the harsh conditions of outer space. Now I'm wondering what will happen to the Hubble's new/old 486 CPU when it hits the year 2000.

Will this be NASA's latest, greatest, and last blunder?

-- Ocotillo (peeling@out.===), December 25, 1999.

Ed: I wasn't thinking so much of the immediate post-new-year timeframe as much as months down the road. I can easily see this becoming sort of a virtual post-war unexploded bomb type situation, but without anything like the British "UXB" teams.

Re MS tech support, I *think* that except for contracted (read: to major corporate customers), most if not all OS- and application-level tech support is handed off to the OEM vendors. If they'd crack down on the "just reformat that bad boy" level of "support" the OEMs provide, things might be different. [g]

One further point to ponder: the vast majority of y2k- "complianceware" arrives as bits over a modem. Reformat the drive, and the *fixes* are gone, possibly forever.

I think we're going to be hearing a lot of muffled *thwump* sounds off in the distance, quickly followed by, "Oh NO!!!!" and a colorful stream of profanities, as Joe and Jane Six listen to the Nice Man and do what they're told, then realize where it left 'em.

-- Ron Schwarz (rs@clubvb.com.delete.this), December 25, 1999.

Ocotillo: there's nothing inherently flawed in a 486. In fact, the heart of a modern Pentium -- apart from the MMX stuff and a handful of rarely used opcodes (mainly, IIRC, multiprocessor-handling oriented), a Pentium is just a pair of streamlined 486s with some cache on a single substrate.

I'd be surprised if the new hardware going into Hubble wasn't y2k- compliant. I believe I read that one of the main reasons it was going up in the first place was to ensure that it (Hubble) would work past end of year.

-- Ron Schwarz (rs@clubvb.com.delete.this), December 25, 1999.

One of the technical points about the 486 being used in the Hubble or any air or space application: the cpu's in these applications don't use real time clocks.

The real time clock problem is largely what makes the embedded systems problem real. But the aerospace applications use precision (read: EXPENSIVE) crystal oscillators for timing, which simply count out billionths of a second for timing functions.

These things are basically "bulletproof", unless you get a batch which has frequency drift with temperature. And we work hard to weed that our during product testing of the aircraft and spacecraft computers.

And FWIW, the aerospace world is just getting around to starting to use Pentiums in prouct design. So using a 486 is pretty much "state of the art" for a proven and tested device being installed today.

WW

-- Wildweasel (vtmldm@epix.net), December 25, 1999.