Re: ***** sp2 bREAKS eVERYTHING!!!!! *****

From: Don Taylor (
Date: 09/03/04

Date: Fri, 03 Sep 2004 08:31:52 -0500

Nathan McNulty <> writes:

>See inline comments

>Nathan McNulty

>Don Taylor wrote:
>> Nathan McNulty <> writes:
>>>First off, where are these posts about SP2 that haven't been explained?
>> Ok, where is the explanation for the string of folks who all have a
>> pretty similar description, "click on a file and Windows Explorer crashes"?

>Date of post and post title (and newsgroup)? I have yet to encounter an
>issue like this on the 20+ computers I have installed SP2 on as well as
>ever during all of the beta testing.

You posted to, or replied to, a posting that also appeared in

Newsgroups: microsoft.public.windowsxp.general,microsoft.public.windowsxp.basics,microsoft.public.windowsxp.hardware,microsoft.public.windowsxp.newusers,microsoft.public.windowsxp.help_and_support,microsoft.public.windowsxp.accessibility

and I'd guess in the last week there have been between fifteen and thirty
different people post in windowsxp.general, each describing something very
similar to what I just described. If you need to find them you can search
old postings for SP2 and Explorer, and then toss out those that meant
Internet Explorer. Or if you want I can do that, append the hundred or
so messages specifically addressing this and send it to you.

The number of different people with very similar descriptions of their
problem have led me to believe that this isn't just random problems,
these folks are seeing the same bug(s).

>>> The only real problems that can't be fully explained (but can still be
>>>fixed quite easily) have to do with messed up chipset drivers which
>>>introduce abnormal behavior.
>> I'd guess clicking on a filename might be less likely to be a chipset
>> issue, but I'm open to your explanation for this one. Thus far in the
>> thousands of posts I've read I haven't seen a single explanation for
>> this. Let alone a fix. (thus far there are two work-arounds that do
>> part of the job for some of the people with this until a real fix comes
>> along, run in safe mode or create a new user)

>Again, I haven't seen any posts with this description, though I haven't
>looked through the general newsgroups in a while. Here in the hardware
>newsgroups where I spend most of my time, it has all been chipset issues
>which seem to be resolved by reinstalling the chipset drivers. I don't
>even want to take a stab at why this is happening.

My reply had nothing to do with the hardware newsgroup. Whoever started
this just shotgunned the newsgroups.

>Also, in the hardware newsgroup, there have only been 400 posts dealing
>with SP2 (that is including replies too!). I am sure the General
>newsgroup is a zoo, which is why I like to stay out of there as it moves
>much too fast for me.

Not too bad, maybe a thousand or maybe more postings a day. And it gives
a much bigger view of what happened when SP2 hit the street.

>>>This can be fixed by reinstalling the
>>>chipset drivers. Give the Service Pack a little while for the MVP's who
>>>are still learning its in's and out's. Some of us who help out know
>>>more about SP2 than others and we are all learning to some degree.
>> I can certainly understand that. From reading all the posts there have
>> been very very very few postings that really have a specific explanation.
>> THere are a handful that suggested changing something and a person would
>> report that it worked. But it is still early for anyone outside of
>> Microsoft to really know why something doesn't work. And I haven't
>> seen any report of which chipset is responsible for Explorer crashing.

>I can tell you that certain Intel Chipsets with Prescott CPU's have
>caused quite a bit of fun. It is still early and most people are still
>learning how to troubleshoot SP2. Also, many hardware makers haven't
>released software that is specifically designed for SP2, but simply use
>a patch or something similar to allow Windows to run it. These problems
>will begin to resolve as time goes on since these drivers/software
>programs will finally be fixed. It is NOT Microsoft's fault when
>software or drivers don't work properly on their OS. It is the
>responsibility of the hardware/software vendor to make their product
>work with Windows. Microsoft provides the proper tools to these vendors.
>>>Here is what really causes all of the problems with SP2: Non-Technical
>>>computer users.
>> "All" is an impressively strong analysis for you to be able to claim.
>> But that is a market that Microsoft doesn't seem to want to walk away
>> from. If Microsoft wanted to refuse to sell products to anyone who didn't
>> pass an entrance exam maybe this would be different. But if they have
>> chosen who they want to market to then they accept the consequences.

>All is a very strong word and I agree, too strong. I would say though,
>that most of the problems I have helped with, especially when deploying
>the service pack at my college, have been caused by installing "crap"
>for a lack of a better word.

That's a wildly different claim.

Too much of what has been posted by a number of folks have a sample size
that is too small to mean much of anything. I'd dearly love to see the
results of Microsoft randomly calling 10,000 individuals who downloaded
SP2 and quiz them about what problems they had, if any. I'd guess that
they might even know that by now, but they aren't talking as far as I know.

But I'd still be curious how you have actually been able to really KNOW
that software on the system was the cause of the problem, how can anyone
REALLY determine that?

>>>This is the average computer user who knows how to
>>>browse the internet, but knows nothing about security or even how to
>>>install drivers. These are the people whose systems are inundated with
>> The chant "it is all viruses and spyware and stupid users" has been
>> posted in this newsgroup hundreds of times in the last week. These claims
>> are easy to make, just point the finger at someone else. Maybe some of
>> them are true. But I haven't seen any people who went back and checked
>> and confirmed that this was the case. Where is the evidence that this
>> claim is true?

>I say this simply from experience on a college campus. Now I admit
>colleges are a totally different crowd with all the P2P software and you
>don't even want to know what they are trying to download, but it would
>probably be something they don't want to show their parents. Along with
>this comes all wonders of malware. I would consider myself
>knowledgeable and keep my Antivirus as well as my Firewall up to date.
>I have not had a single problem in the last 5 years since I have learned
>how to keep my computer secure and up to date on the software/drivers.
>I do admit that it shouldn't take that kind of knowledge to keep the
>computer running though which is something Microsoft needs to work on if
>there is even anything they can do.

A mailer and web browser that would have kept that crap off the system
in the first place in almost all circumstances would have kept us out
of the mess that got us were we are today. This isn't rocket science,
but a mailer that can script reformatting your drives must be.

>>>I have installed SP2 on over 20 systems so far and not
>>>a single one has had an issue (except one that was a Prescott CPU which
>>>was my fault for not updating the BIOS).
>> For a particular collection of machines that is uniform enough I would
>> not be too surprised if you only had a 5% failure rate.

>These are all machines that I have built. There are only two of them
>that could be considered "uniform." A few of these are computers of
>friends that I have built, but most are my computers that I have here at
>home and at school.
>>>The problem doesn't point to the software, but more often the user and
>>>junk they have allowed on their computer. Don't blame SP2 when you
>>>can't take care of your computer. I would be willing to bet money that
>>>any of these people doing a clean install of SP2 integrated will have
>>>none of the problems they currently have. People just don't know how to
>>>take care of their systems.
>> I have antivirus software updated daily, I have a firewall updated daily.
>> I have spyware checks run once a week. All email comes in here in ASCII,
>> I NEVER let something like Outlook execute arbitrary code in mail. The
>> security settings on Internet Explorer are turned up high enough that I
>> continually accept or reject little popup windows, high enough that some
>> of the web pages at Microsoft will not correctly function, even if I do
>> accept all the warnings about the pages. There hasn't been a virus on
>> the system in four years, other than those I manually transfer via ftp,
>> to see if the antivirus software will sieze it the minute the transfer
>> is finished or whether it is too new for the av to see it, and then I
>> ftp it off to the av-lab. All the latest updates from Microsoft were
>> put on the system, other than things I will never use, like each new
>> bigger version of Microsoft Media Player. I've been in the software
>> business, writing and using, for more than twenty five years.
>> I install SP2 and Windows Explorer in every use on the system immediately
>> locks up, refuses any keystroke or mouse click.
>> I think I know how to take care of my system, and those that I do
>> support on. I've hammered the users so they never get crap on their
>> systems. You want to bet your money. I'm up for that, let's make it
>> an interesting amount. Put it on the table, let's do it.

>Have you actually tried a slipstreamed (SP2) install with default

At the moment I'm not screwing with the system, hoping MS support and
I can track down at least one of the Windows Explorer bugs and get them
to fix it before I just reformat the drive, the problem goes away and
Microsoft writes it off to another unknown problem that is forgotten.

I really hesitate to use the new method of debugging, "cycle the power
or reinstall it and see if you maybe don't bother calling us back."

>You being a software developer, you know how much coding goes
>into something like Windows. You really think they are going to be able
>to make an OS that is 100% compatible with absolutely every peice of
>hardware out there? You think that every software/driver developer is
>going to write perfect software/drivers? There is no such thing as the
>perfect code. There are going to be systems like yours or those with
>the special Prescott case that no matter what you do, Windows will not
>function as it should. You just have to ask what is preventing it from
>doing so. Is it actually Windows or is it your hardware/software/drivers?

First, you have changed the rules. You proposed putting money on the
table to back up your claims. I agreed we do that.

Now you have done what I've seen done a hundred times before, changed
the discusstion to "software can't ever be perfect so you can't complain
about how good or bad any of it is."

Just for a baseline, I spent the decade of the '80s finishing a product
with menus, hardware, multiple sources of commands, more conflicting
hardware constraints than Windows will ever imagine, etc. We put effort
into producing good software. Our measure of that was 1 user out of
every 2000 using the product full time year round would find 1 bug, the
other 1999 would never even see it, and that I would be able to quickly
find and fix that and have less than a 1% chance of introducing another
bug in the process. Put another way, imagine two dozen people using
Windows full time year around for 50 years and after all that there
would be >90% chance that not a one of them would have ever seen a
single bug, no matter how small, no matter who you wanted to point
the finger at claiming it was their fault, not yours.

When Windows code gets somewhere vaguely near that we can discuss quality.

At the moment I'd suggest that no user can likely determine why
something failed and it is probably the case that Microsoft can't
determine why a particular failure happened with a particular report.
If history is any guide, it is often the case that nobody REALLY
wants to know exactly whose fault it is, that gets embarrassing.

I still cannot understand why SP2 didn't do a virus/spyware check, do a
system file check, do a registry validation, and turn off or force the
user to turn off any and all running programs that were not essential
for SP2 and kill any unknown processes BEFORE the installation.

It wouldn't be that hard to have done that and if some of the claims
being made that these cause "all the problems" and that there are no
problems with SP2 are true then Microsoft could have saved themselves
from untold support costs and public relations expense. That isn't
even considering the cost the users paid for this, that's just
considering the savings for Microsoft. Even if that had only reduced
the failure rate by 90% it would have saved a fortune.

>> Don Taylor
>>>Nathan McNulty