Slashdot Mirror
9/9/99: News? Nein!
SEWilco writes "As Slashdot readers know, today's date abbreviated as 9999 may cause problems in some older computers. So far only one report of a Tandy problem.
9s-day no problem in New Zealand and Hong Kong, Guam OK and USA still has electric power on 9/9/99.
But seriously, folks, today is a big day for numerologists, pagans, and Nostradamus. So far today the NASA Near-Earth Object Program has not seen a comet coming to hit us. But what is so special about the Era of Alexandria 7491 anyway?"
If you are in Chicago, it's just more of ComEd's crumbling infrastructure rearing its ugly head. I'm half-expecting complete outages by Y2K, but the bugs aren't going to be related to 01/01/00 -- they are a result of years of criminal negligence on the part of an underregulated power monopoly, the results of which are just now coming to fruition.
Having said that, I am indeed skeptical of all of the "Y2K readiness" leaflets I receive in the mail. I suspect a good percentage of them are written from the standpoint of "we're not quite ready yet, but we will be, so let's sooth the unwashed masses." Unfortunately, as anyone who has ever written anything more than a trivially simple program realizes, deadlines have a way of slipping real fast in the world of software. I suspect there will be two major dangers resulting from the whole Y2k hoopla:
1) People being stupid and panicky (as many others have mentioned already), cleaning out stores and what have you before Y2K arrives.
2) People having done too little to prepare, trusting deceptive reassurances from those companies and services who weren't able to make their deadlines and be compliant on time, who then do an about face from blase' "it's nothin' to worry about" to full-fledged panic as they discover a whole host of inconveniences which combine to make their lives more than a little difficult.
In both cases the danger will be a result of panicy people, not technology. But to dismiss the notion that one should be planning for contingencies invites a whole host of problems of its own, quite possibly making a touchy situation worse. The best approach IMHO is to take some reasonable precautions:
* Have a little extra cash on hand in case the ATMs are down
* Have a hard copy of your statement from a day or two before the new year on hand
* Have a little extra food on hand (maybe a month's supply, instead of a week's)
* Have a few candles lying around
* Have a warm blanket handy in case the power does fail (thanks ComEd).
* Have a good book or two, for the same reason
* Relax. All those preparations were probably unnecessary, but now you don't have to worry even if things do come unravelled for a little while.
If one has made reasonable contingencies, one won't be one of the idiots consumed by panic when we usher in the new year with a few bumps. Of course, don't go shooting your gun into the air new years eve. Not only could the bullets injure someone when they fall back to earth, but, more importantly, you'll probably need that ammo come January first. (For the humor impaired: the last comment was a joke!)
The Future of Human Evolution: Autonomy
Obviously Sony must have used COBOL when coding for the Playstation 2.. see Playstation2 delayed.
So confidence-boosting articles are just fine by me! :-)
-----
New E-mail address! If I'm in your address book, please update it.
The real meaning of the GNU GPL:
"The Source will be with you... Always."
Apparently 9999 was used instead of an EOF marker. If the software came across it, it would assume that there was no more data after it.
I think this one is a dud; I don't know how it slipped past the "experts".
He told us that 28,000 cops were deployed last night here in HK. During one of their planning meetings he had mentioned to his supervisors that he thought this was just general paranioa. "Sir, I have never heard of a computer nicking a criminal".
The media are believing all the supposed computer "experts" that are coming to them and saying "oh my God the world is going to end".
Look at the Microsoft NSA story. Someone sees the NSA symbol in MS code, makes an assumption, and the media buys it hook line and sinker. But almost every media article quotes some "expert" as saying that it's a real - the media guys don't want to say that themselves. That way, when it turns out to be bogus, they're clean.
I think we could use some sort of trusted agency that would verify computer-related or security-related news stories. Actually we have this in the CIAC and similar agencies don't we? If a CNN story quotes the CIAC, and not "Security Expert Bob Fishpond from Funny Creek, Missouri" then I'd be a lot more likely to believe it.
On 6/6/66, I was little, I didn't know shit, and by
7/7/77, eleven years later still don't know any better
on 8/8/88, it's way too late for me to change
and by 9/9/99, I hope I'm sittin' on the back porch drinking red wine, singing
oooooooooh, french fries with pepper!
-- Mark Sandman, 1963 - 1999
I'm celebrating the way he would have wanted it!
--
Xenu loves you!
I've seen quite a bit of discussion about this on the MIDRANGE-L mailing list, and so I thought I'd mention a few things. A lot of people seem to think that 9/9/99 as a special date is mostly myth, largely because the computer would store it as 090999, which doesn't look as special. Surely 99/99/99 would be better. Other people pointed out that this sort of thing tended to crop up when the users wanted to add extra information in a field the programmer thought would only be a date. In such a program, 99/99/99 would fail because it was an invalid date, so creative users might be tempted to use a vaild date, but give it special meaning. Two common dates for things like "no expiration" or "not applicable" were 12/31/99 and 9/9/99. The former would be the highest date that could be entered in a two-digit year field, while the latter was easy to remember, yet still a good ways into the future.
9/9/99 problems are likely to be fairly rare, since the necessary circumstances would be somewhat rare, and, hopefully, many such problems have been caught by now. Where today's date does cause problems will likely not be noticed by the population at large. (i.e. no power outages, no broken ATMs.) At a guess, the most likely candidates for problems will be billing software running on older mainframs and midrange computers, and I'm sure the companies will do their utmost to bill you for anything you might owe them.
--Phil (I know the banks with my loans have been quite diligent.)
355/113 -- Not the famous irrational number PI, but an incredible simulation!
We're much more likely to see problems caused by idiots rather than computers. Think about what's probally already started happening. People are going to be buying more canned goods and other non-perishables. People are going to buy bomb shelters and take their money out of banks. Banks are going to go under, food prices will sky-rocket, as will any other materials used in a bomb shelter. Then guess what happens when jan 1st rolls around and the world hasn't exploded? Well with all of those non-perishables people aren't going to buy food, especially due to the increased prices because of the limited supply the day before. People aren't going to trust the banks with their money, after all, many went under because they weren't Y2K (read: moron) compliant.
Oh and we can't forget the inevitable rioting that's going to occur. Americans will use any excuse to riot. Your favorite team lost the world series, riot. Your favorite team won the world series, riot. Y2K aka armagedon is here, riot.
God bless morons one and all...
-matt
Computer experts have just determined that, after extensive testing, that the date 5/23/2001 will only occur once. "No computers have ever encountered such a date before," a noted computer expert said. "Who knows if they will work when they encounter it."
Already, several major firms have been created to certify systems an 5-23-01 ready. "People are urged to ensure that their banks, hospitals, and every other business they deal with are ready for this unprecidented event."
There has been a major event on 9/9/99, ... the evil Bill gates cartoon just took over After Y2K! Help us Nitrozac, you're our only hope!
9) Antichrist crashed party, demanded blood sacrifices to the Satan-spawn of Baal. :(
8) Panicing day-traders fleeing the inflationary wrath of Alan Greenspan caused a run on the dollar
7) Today's date actually expressed as 090999, thus disappointing a bunch of Y2K nuts hoping for a warm-up
6) Ricky Martin stole the MTV video award from Weird Al Yankovic, thanks to shameless ballot stuffing
5) Stayed at home working on computer to avoid superstitious fanatics, motherboard overheated, magic smoke came out.
4) Spilled hemlock all over myself at the coven swap-meet
3) "friends" is a re-run tonight
2) Couldn't play the new Dreamcast due to nuclear melt-down of national electric grid
1) That darn asteroid!
We want endless gardens of data, where the bits can flower, flourish and reproduce. -- Andy Mueller-Maguhn
While most posters have ballyhooed the 9-9-99 'bug', I have to relate the problem a company I have worked for ran into. They still use an document database running on an ancient 386. During Y2k testing for the 9/9/99 bug, the in-house server software that allows users to access the database remotely started to intermittantly dump. A bunch of the IS people sat down to discuss replacement / rollback / recode. While mulling over the possibility of rollback, one of the passing users chimed in. "You guys can't mess with the date, it would screw up our age number." There wasn't an age field. They made her show them. Some of the users were filling in the l_date field, intended for lease-end date, with the date the master lease became effective. As a result, a field that indicated days until lease end had a large negative number. When they had set the date to September 8th, one of the leases signed back in '82 managed to hit -9999 days until termination. Unfortunatly, when the machine rolled to the 9th, the field hit -10000, and it would no longer fit into the five char string. Every time someone looked at that particular lease, (which was often enough) the machine segfaulted.
The users were punished.
.sig: Now legally binding!
I studied COBOL in the mid 80's.
Some variables could only store numerical values, which were defined as a number of numeric digits, presumably using BCD as the internal storage value. It was necessary to use magic numbers to indicate special conditions, such as EOF or no more records. Since most files were indexed sequentially, all 9s was often used as special EOF indicator since it would always be sorted last.
However, by the mid 1980s, COBOL introduced a special value, HIGHVALUES, to which any numerical variable could be assigned, which was always higher than all 9s, and was to be recommended for such magic numbers. I think there was an equivalent LOWVALUES as well.
I do not know when these to keywords were introduced, but it was at least 15 years ago.
If 9/9/99 is a problem with COBOL, it must be using a very old variant of the language.