I will document a small business problem I am familiar with. This is a long post. I hope it will prove insightful. I suspect it is happening all around the world, in many small offices, many times over.

First, some background: The company is a small, five-person independent insurance broker with ties to 5 insurance companies. They do 80% of their busisness with one big company (OBC). I have been friends with one of the persons for 20 years, and so agreed to help them upgrade their computers in 1995. We purchased good 486 class machines at the time with really no consideration of Y2K. Each year since then, I have spent perhaps 3 days of vacation (from my real job) doing simple upgrades, backups, new installs, etc.

Last September, I stopped by and installed the latest Win95 and MS Office '97 patches and [a famous maker of virus softwares's] Y2K suite. The Y2K program reported the computer was not compliant, but fixed the problem with a TSR. I innocently thought that these actions would cover their problems from a hardware/system-software standpoint.

At the same time, I reviewed all their application software and pointed out that the program they used to communicate with OBC was not likely Y2K compliant. My friend related that the OBC had earlier sent a Y2K version, but had later asked all their brokers to trash the CD and await a corrected version. She was in posession of this corrected version and would install it that same week. OK, fine.

On the Monday after the CDC, I was back on site (more vacation time from my real job). The computer day/time was OK, but the year indicated 1994. I could reset the date using Win95, but the year then reset to 2094 after reboot.

Here is a technical summary of why I think this happened (please, I am not really a technical person, so will take no offense if others care to correct my analysis): The real-time-clock (RTC) is a static 2-digit century register and reactive 2 digit year register. That is, the RTC software will change the year at midnight on 12/31, but the static century value is never touched by the BIOS/RTC. Unfortunately, when the RTC was updated to '00', the BIOS presumed the RTC was in its initial state and updated the clock to the BIOS earliest date of '94' - thus leaving the date at 1994. When I used Win95 to reset the date to 2000, Win95 was smart enough to reset the static century register to '20'. Upon reboot, the BIOS noted the RTC was in initial state and reset the RTC to '94'- thus leaving the date at 2094. Finally, when the date was set at 2094, the network cards failed to initialize for some reason.

My first attempted solution was to verify that the [famous maker's] Y2K software was inplace. It seemed to be installed, but obviously not working well. Famous maker's help line (unsurpirsingly) was impossible to contact, and I have still not received (after 2 weeks) the password required to access their online problem database.

We agreed upon a simple work-around. I used the DOS date command to set the date to 12/31/99 prior to starting windows and just after shutting it down. This seemed to solve the network problem, and meant that she would have to reset the date each morning, but no big deal.

Then I discovered that she had failed to install the Y2K version of the OBC they had received last September. I gave her some pointers and left with congratulations all around.

They have not done a lick of work since then. On Friday, I got a call from an on-site repair person trying to fix the OBC software. He explained that they needed new computers all around, because the new software would not work on a 486. Hmmm... I told him I hoped they would fix the software by the time my friends had purchased new the computers. He laughed and assured me there was no problem with the software. OK fine, I was happy to know my friend would soon be getting a new computer, installed by someone who did not have to take vacation time from their real job.

I believe I was as proactive as possible - installing all the required software upgrades and BIOS patches. IT was to no avail. The hardware still caused problems. The Fix-On-Failure approach adopted by the owners of the buisness for a major piece of operations sorftware compounded the problem.

They will lose perhaps a month of productivity. Will any client notice? Not if the problem is fixed by the end of the month. Has it been very costly? You bet. Could things still not work out? That risk is very high. Will it ever be reported on the news? NO!

I have to believe that many small businesses are even less prepared than my friend's business. Folks, this thing is just starting.

Sincerely,

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000

Answers

I'm So,

I'm not trying to disagree with your basic point that there are TONS of glitches and problems all over the place. But I think there's another aspect of Uhhmmm's story that should be kept in mind.

Since most small businesses don't have a full-time, savvy hardware/ software geek on hand to run around and constantly fix things, what they try to do is get themselves into a "stable" situation and then establish a firm policy that can be summarized as DON'T ROCK THE BOAT. Sooner or later, most of them learn that every time they try to upgrade something, install a new version of something, or make ANY change to their operational environment, there's a good chance that it will cause all kinds of nasty, unexpected, mysterious problems that will take days or weeks to fix. They get on the phone to the help-desk geeks at Microsoft or Lotus or Intuit or whoever, and they get a bunch of gibberish they don't understand, combined with the terrifying recommendation to "re-format your hard disk and reinstall EVERYTHING all over again," at which point they realize that they haven't done a backup in six months.

It's quite possible that the FOF strategy adopted by so many small business owners is a direct result of this bit of tribal wisdom: they may have sensed that their efforts to upgrade and fix their Y2K problems would just create MORE problems. I have to admit that during the brief six-month period when I succumbed to the Dark Side of the Force and tried to use Windows 95, I felt much the same way about my desktop PC: I had about 20,000 files on the machine, and once I got things working in a stable fashion, I was very reluctant to make ANY changes to the hardware configuration, or the driver software. Every time I tried to install some new software package, I found that it overlaid some DLL that made another program very unhappy ... and on and on and on ...

Ed

-- Ed Yourdon (ed@yourdon.com), January 16, 2000.

To the top.

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000.

This is a good example of why we should not attribute the lack of large-scale problems to excellent remediation. There is no way the lack of failures at rollover was from a near-flawless implementation of remediation programs.

-- ImSo (happy@prepped.com), January 16, 2000.

Excellent grassroots post, uhhmmm, thanks

-- Guy Daley (guydaley@bwn.net), January 16, 2000.

great highly informative post! thanks!

-- Carl Jenkins (Somewherepress@aol.com), January 16, 2000.

its a simple bios bug I had the same problem with award modular bios 4.50g go to award.com to download fix if you are using award modular bios I fixed on failure (in fact 2 weeks after), the intial work around was to always set date on power up. Folks this thing is over. Robert Menger

-- robert menger (robertmenger@email.msn.com), January 16, 2000.

Robert,

Thanks for the info regarding the Award BIOS. I am somewhat reluctant to get back into this mess, and my friend seems to be happily looking forward to new computers. But I will ask the owners if they are perhaps interested in a correction to their hardware problem.

Now, regarding your contention that (if I understand your point) this is a limited, insignificant, quickly fixed problem ...rant on

You seem to be a very good technician. However, most small business owners get memroy and DASD confused, few would know an Award from a Phoenix BIOS, and even you yourself could not fix the application stuff.

I believe there are many small co's that do not have access to your level technician and who are still attempting to FOF with their own skill set. They may eventually figure out the hardware stuff or seek services of a professional. The professional may help, or may just try to sell them more hardware.

Also, I believe there are many applications used by small co's that were not compliant, they did not understand this fact, and they are madly trying to solve this problem as we speak. There may be newer software available, or there may not. If available, different (or even newer) software must be located, purchased, installed, and mastered. This process requires money and (the same thing) time. None of this effort will be apparent until it fails and the company goes TU. IMHO, Y2K is that loud sucking sound we are beginning to hear that drains the profits and productivity of small companies everywhere.

I gave an example of one small business. I extrapolate based upon my limited knowledge. Perhaps this extrap is fundamentally flawed, but your very technical post just confirms my earlier thoughts. I remain convinced: this is not over because, IMHO, there are far too many small businesses still on the road to FOF. Some will successfully reach the end of this road, others will fun off the cliff (Road Runner style?).

...rant off

Sincerely,

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000.

Uhhmmm...

Do you have a sense of the denominater--that is, do you have a sense of how many gazillions of glitches of ALL varieties there are in ALL the computer systems all over the place?

This is why Y2K is over: whatever the Y2K problems are, they are lost in the larger sea of computer glitches we live with every day. The show was 1/1/00, and the show flopped--without thousands of failures happening SIMULTANEOUSLY, there are just thousands of failures of one type or another nested within tens of thousands of failures that we live with every day.

It's not "starting". What are you talking about? Computers have crapped out as long as there have been computers. How many small business do you work with? Their systems are always failing, just like everybody else's. Small business fail for these and a myriad of other reasons. Is this really such a difficult notion?

I mean, the determination to see a Y2K bogeyman REAL SOON NOW is quite incredible. Makes for a comical forum, but surely most analytic thinkers have moved on.

-- I'mSo (happy@prepped.com), January 16, 2000.

Ed,

Thanks for the words and thoughts. I believe you are quite correct. I do not know the average age of the small businesses in this country, but I suspect that the average age puts many beyond the time when anyone thought of Y2K as a problem. For this reason, I suspect, many small businesses have somewhat outdated computers and software. I think they probably purchased these systems the year they were born - just after the job got to be too much for one person alone.

[ImSo] seems to tout the number of problems we see everday, and to use the suposedly large number to discredit the problems caused by Y2K in the small business. I rather suspect that small co's see actually very few problems for the reason you mentioned - FEAR of Failure. They have avoided the change/problem cycle we see so often in larger co's with professional staffers. Now they must change or die, and the change itself has to be causing additional problems. Just one new problem in a static environ may be catastrophic.

BTW, Ed, you (and your books) taught me structured programming near the dawn of time. You are a giant, and have done more than any other single person to bring programming into the realm of the 21st century. Best wishes to you and your family in the new year. Please keep publishing.

Sincerely,

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000.

---h-m-m-m-m-, got an old mac, didn't download or install patch one, nuthing. Myabe friend buying new office puters might wanna take a peek- maybe, just maybe-at some macs. there's plenty of office and business software out there for it, far as I know. I know of several companies in atlanta area still using old macs that are networked, they all work fine. There's a few progs that weren't compliant, ya drag em to the trash, that's it. Unless you enjoy continually fixing glitches and bugs forever, then by all means stay wintell.

not a rant, just an observation, as long as folks keep giving money to the companies who build buggy software, they'll keep selling it to you, business or personal. And also not to say macs can't have bugs, but just my observations of the WEIRD problems I've seen just on this board with wintell stuff, and the hoops you have to jump through to get it "fixed", I'm amazed and almost can't believe believe folks still buy that stuff. Geez, the nationwide probglems must be in the quadzillions all the time, day after day, y2k or not. simply amazing. Just goes to show what advertising can accomplish in sales, I guess. Only problem I've had in mine is a slow modem, it works, just slow, and once got an unidentifiable bug, that really didn't shut me down, it just made it hard to use, so I reformatted and reinstalled from the cd's I had, been fine since.

anyway, good luck to you and friend, hope the business doesn't suffer too much financial loss or lost person hours of productivity, really. Just offering a longer term alternative to look at.

-- zog (zzoggy@yahoo.com), January 16, 2000.

Ed, I agree with your post completely. And I wonder, in the context of post Y2K, how many small businesses were just as well off doing nothing...

No doubt large systems that interact with others had to be fixed--and were fixed--over a broad span of time. But we had a lot of small enterprises who did nothing for the EXACT reason you mentioned (STABILITY) and so far they have done fine. One of my major criticisms of pre-rollover hype was the notion that your system had to become "Y2K" compliant or it would FAIL. Not just return a goofy date, or a goofy number (say, a bad bill) that had a workaround, but FAIL. And the tribal wisdom of leaving a working system alone turned out better for some than responding to the decree to "fix" as system prior to rollover. In a perverse way, older systems such as the 486 system mentioned above are more fragile and less likely to be "fixed" when you mess with them.

Put new lights on an old jalopy and overload the electrical systems, I guess.

I saw very few Y2K consultants addressing the possibility that upgrdading to Y2K compliancy might wreck your system...

-- ImSo (happy@prepped.com), January 16, 2000.

[ImSo] said...

But we had a lot of small enterprises who did nothing ... and so far they have done fine.

...said [ImSo]

I believe my conjecture is that small businesses are not doing just fine. We cannot really now know. My experience indicates otherwise, in fact. We will know soon enough.

Thanks for your thoughts,

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000.

zog,

I always take your admonitions to heart. Your experience is a welcome resource.

Thanks,

-- Uhhmmm... (JFCP81A@aol.com), January 16, 2000.