Y2K tests uneventful, securities industry says

TORONTO (CP) - An initial millennium-bug hunt in the computer systems of the Canadian stock, bond and mutual-fund industry has come up clean.

The Canadian Securities Administrators said Tuesday that the industry "successfully completed preliminary tests that simulated a variety of transactions covering trading, order entry, settlement, clearance and validation processes."

The beta tests were conducted March 13-20 as a dry run for full testing in May and June in preparation for the new year, when older computer systems that use only two digits to designate the year are expected to be befuddled by the change from 99 to 00.

Canadian securities commissions, which have been working on the problem since 1998, are running separate Y2K tests for the mutual-funds sector and for securities-related organizations such as stock exchanges, securities dealers and clearing agencies.

"The beta tests were a success," Randee Pavalow, head of the CSA year-2000 committee overseeing the tests, said Tuesday.

A similar bond-market test co-ordinated by the Canadian Depository for Securities Ltd. on March 6-10 was also a success, the CSA said.

) The Canadian Press, 1999

-- Norm (nwo@hotmail.com), March 24, 1999

Answers

Interesting ... most professional testers would argue that an "initial" test would be considered successful if they FOUND bugs, not if they FAIL to find bugs.

Of course, the optimistic interpretation is that there aren't any bugs at all. It's theoretically possible for this to be the case, but difficult to imagine when we're dealing with a large, complex legacy system. The only way that any professional software engineer would believe such a statement is to ask questions like, "How many test cases did you try? What percentage of the logic paths did you exercise with your tests? Did your tests include post-2000 dates as well as pre-2000 dates, and if the former is true, did you 'age' the date-fields on your database?"

Ed

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

"The beta tests were a success," Randee Pavalow, head of the CSA year-2000 committee overseeing the tests, said Tuesday.

Oh, and be sure to ask Bill Gates how his "Beta Test" merchandise comes through from start to finish without any bugs.... (sarcasm intended).

Mr. K
***has had to patch many finished software applications for bugs -- even from the esteemed, superior Microsoft guru group-- haven't you?***

-- Mr. Kennedy (not@all.convinced), March 24, 1999.

It's absurd to think that the securities industry is going to publicly admit the existence of any problems. The whole ponzi scheme is based on investor confidence. The industry itself is not going to do ANYTHING to to harm that confidence. Any announcement that 'there are no problems' can be disregarded. They wouldn't tell you if there were.

-- Spidey (in@jam.com), March 24, 1999.

i dont know if it is true but i did read somewhare on the net that the test was --not-- rolled over to 2000 but was a check for the present operating system.

-- bob (rcrozier@koyote.com), March 24, 1999.

"beta tests," Norm. Full testing not till May and June.

You get high honors for mastering the art of cut n' paste, Norm, but don't you think you should at least read these silly PR pieces before posting them?

Then maybe, just maybe, after becoming comfortable with reading, you might want to venture an *original thought* and attempt a thing we call "writing."

-- rick blaine (y2kazoo@hotmail.com), March 24, 1999.

Is that the Royal "we" Mr. Blaine?

-- Norm (nwo@hotmail.com), March 24, 1999.

Norm, you sure it's OK to be up & running around so soon after your frontal lobotomy? Maybe a few weeks off... rest up...

-- Lisa (lisa@poor.guy), March 24, 1999.

Well done, if you don't like the message, kill the messenger. Everyone screams that there is no good news and, when someone posts it, he gets ridiculed. Yes, they were beta tests and they went WELL. No one said that they were the final tests and that they were done. They even gave the date of the final test. Why can't you people accept the fact that progress is being made and not everyone has accepted your defeatist attitudes ("It's systemic, it can't be fixed, why bother!).

Ed, I am particularly surprised at your comments. Since the test results contradict your premises, obviously the people doing the tests must have been incompetent and didn't do them correctly. Because we all know that EVERY system out there has a Y2K bug somewhere so, if you say you are compliant, you are either a liar or an idiot or both. Although I disagree with most of your ideas, I would have expected better from your response. You ask for all sorts of specific facts and numbers from everyone who is actually trying to resolve the problems but then you publish your essays that are filled with "I believe", "we expect", "computers are likely", "it may", etc. etc. without a shred of hard evidence to back it up. And I have read everything on your site so it is not just your most recent essay. I agree that a great deal of this is based on opinions and educated deduction based on the facts available but why do you have a much higher standard for what others write compared to yourself.

ANP

-- Another NORMal Person (ANP@BettyFord.com), March 24, 1999.

Lets get liquid!

-- MarktheFart (quke@somewhere.com), March 24, 1999.

Let's see, time for a paradigm shift, have i got.... NOPE only $.18. So my $.02 worth. Quite a bit like Ed, EVERY system I ever tested, Alpha or Beta, was tested until we found errors. We used to ignore the minimum syntax errors in compiles, and once in a while they would come home and bite us, but every test session, until the user had had the orangutangs from data entry bang on it had errors. then they found more. Second test run/beta test series without errors?? Scares the excretions out of ME!! WHAT did we forget?? How many cases of that did we try? Etc, etc.

Chuck

-- Chuck, a night driver (reinzoo@en.com), March 25, 1999.

ANP = Norm

-- Lurch (watching@looking.com), March 25, 1999.

Nice reply, Lurch! Have anything intelligent to offer in rebuttal? Didn't think so!

-ANP

-- Another NORMal Person (ANP@BettyFord.com), March 25, 1999.

Let me try to put another slant on this report. We just got thru with our final testing of our revised laboratory system (hospital). Our original beta test also went well. The software did not destroy the database and almost all the external interfaces functioned. We thought that was great because it meant that most of the critical functions had been sucessfully fixed. Would the software have worked? No way! All the little problems would have overwelmed us within a day or two.

That was 18 mo. ago....Now we finally have a system that will funtion in daily production. We have tested in a time machine and found no rollover problems or problems across the rollover. But this was not true up until 2 mo. ago. Will we still have problems. I will almost guarantee it because it is almost impossible to test the system in every way that it will be used. But these problems should be manageable.

I can't imagine that a large system that the exchange uses, only now (with less than 9 mo. till rollover) going into beta can possibly make it without significant problems on the rollover. This does not mean that they will fail. They might get lucky and the problems will be easy to fix. But it's not the way to bet.

I will however give the above as a success story for all you pollyannas out there. We are done! And yes we are even done with the embedded chips! Now if the power co will give us electricity and all those reagant manufacturers can continue to make reagent for our machines to use. And the trucks have fuel to get it to us. And the bank continues to pay the techs........Oh you get the idea.

Got medicine?

LM

-- LM (latemarch@usa.net), March 25, 1999.

ANP, Norm, you've extrapolated evidence again to very, very wrong conclusions based on essentially no evidence - the mere existence of this test is itself good news, it means they have begun testing - TESTING THE PROCEDURE THEY ARE GOING TO USE TO TEST THE PROGRAMS. Later.

It doesn't mean anything else. These results don't mean anything other than this specific statement - "under the specific conditions that they tested, for the specific program versions and operating systems that they tested, for the specific test cases and input conditions that they ran the test under, they did not find the errors that they looked for." Are others errors left undiscovered? Will other operating systems or versions create errors? Will other test conditions create errors in output, input, or transferred data? Maybe.

(Actually, certainly, but I'll give you the benefit of the doubt that is the first error-free software since the Assyrians created the calender assuming the earth stood still and the sun revolved around it. That "program" worked fine until somebody invented the telescope and found a different test case.)

As Ed Y. alluded above - I'm very skeptical of tests that find nothing - in every case I've found in testing my own company's software, it only means I haven't looked hard enough.

-- Robert A. Cook, P.E. (Kennesaw, GA) (cook.r@csaatl.com), March 25, 1999.

The industry "successfully completed preliminary tests that simulated a variety of transactions covering trading, order entry, settlement, clearance and validation processes."

I may be missing something. This statement to me doesn't mean that there were no problems found. As those wise to the ways of IT have pointed out, when you do a test like this, you *want* to find bugs. So a successful test likely found some. This sounds like a PR version of an IT report saying that testing was done and that the bugs found are now being worked out. Could be wrong, hope not!

-- Tricia the Canuck (jayles@telusplanet.net), March 25, 1999.

Wonder why were not always accepting good news at face value, ANG?

See thread ...

More companies say don't panic

http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id= 000eNC

Could be more covert than overt reasons.

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), March 25, 1999.

Sorry, ANP, not ANG ...

(Can't resist .. Another Nincom Poop). Not. However, not seeing far enough either.

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), March 25, 1999.