matt@psuhcx.UUCP (05/23/88)
This is in response to <21045@think.UUCP> , and
<21061@think.UUCP> wherein
ejb@think.COM (Erik Bailey) quotes:
>Lineater, you'd better get your asbestos suit on pronto!
>***FLAME ON***
>In article <82@psuhcx.psu.edu> matt@psuhcx (Matt Cohen) writes:
(referencing difficulties with FLUSHOT PLUS)
>> I then rewrote the FLUSHOT.DAT file with only two programs,
>> command.com and a.bat checksummed. Flushot checked them on
>> startup, but did not perform as advertised when I ran A.BAT,
>> changed it, and ran it again.
>Well, what type of character did you put after the checksum? Only a +
>checksums a file when it is run. A - or ,1 checksum it at FSP load time.
Gee Erik, I thought since you were a beta tester for fsp that you had
read the manual:
(The following is from fsp.txt, the manual for FLUSHOT PLUS):
' Sorry. I got carried away.
' Seriously, there is more. When a "checksummed" file is loaded by
' MS-DOS, it will, by default, be checksummed again. So, if you
' had a line such as:
'
' C=C:\usr\bin\WS.COM[12345]
'
' the venerable old WordStar program (still *my* editor of choice!)
' would be checksummed each time you went to edit a file.
'
Noting that you have to enter each checksum manually using flushot each
time you add or modify a program , Erik comments:
>Nah, I don't think it would be TOO hard to extract the encodeing technique
>out of the code and rechecksum a trashed file...
I am sure a five line 'C' disassembler could do it in under an hour.
(By the way it is 'encoding', not 'encodeing')
>Is 5 minutes of entering a few numbers REALLY
>so BAD???
5 minutes for only one program, however virus protection is
pretty useless when only one program is protected. We noticed
that FLUSHOT simply DIES when we put in a list of only 37
programs? How many can it do, 10? (a dollar each :-)!
>Would you rather it do it FOR YOU? Gee that would be swift, wouldn't it?
>Awwww.... Can't read a number and type it in? Looks like you have a
>limited mental capacity.
You are starting to sound like the spiteful section in the FLUSHOT
manual.
A typical PATH may contain up to 400 programs. Assuming FLUSHOT could
handle that many programs (which it can't), by the time you
finished entering all the checksums, you would have spent the
better part of a day. You may enjoy that sort of thing. I
would rather do real work.
We are then presented with this convincing argument for typing in the
checksums manually:
>Lissen Matt, that is in
>there for YOUR protection!!!
Yes, I agree, it will protect me from doing anything while I am
repeatedly editing the FLUSHOT.DAT file.
In response to my description of how my CMOS RAM was trashed by
FLUSHOT, Erik had the following to say:
>>FLUSHOT PROTECTS CMOS RAM ?
>Yes, it does. Since I don't have an AT, I've not used this feature.
>It may be a bug.
I like the logic. It must protect CMOS RAM since you don't have any.
However, maybe it doesn't.
Finally, in response to my tongue in cheek speculation about FLUSHOT
PLUS having a time bomb to wipe out non-registered users:
>Well, I take offense to this. Why? I'm a close personal friend of Ross,
>and a major beta-tester for his software (INCLUDING fsp).
No wonder the bugs went out.
> Look. Ross
> didn't HAVE to write that program. In fact, way back on FLUSHOT v1.0, he
> just wrote it for the heck of it. He had no idea it would turn into
> practically a full-time job.
It wouldn't have if the software was designed right in the
first place.
> If you put that much effort in, I'd suspect
> that you'd expect some money in return.
If Ross wanted money, why didn't he just sell the thing to
someone or publish it? Because you have to get the nasty buggies
out first or you may be liable when someone's CMOS RAM or disk
gets wiped out!
The author chose to release FLUSHOT into the public domain with
the intention of providing something useful to a lot of people
without having to spend a lot of time designing and testing it.
Posting over USENET effectively puts something in the public
domain, even if the letter of the law is a bit shady.
>BTW#2 -- there is NOT a counter in the code.
How do you know, did you read the source code? Can we see it?
Pretty hard to program a checksum without a counter, eh? Maybe it IS
worth the $10!!
>BTW -- a *LOT* of people have
>registered FSP.
A lot of people ran the CHRISTMAS EXEC program without reading it
because it said to do so. I guess I have to take your word that there
is no time bomb in the code, since the author has chosen to not make
the source code available.
As promised, Eric forwarded my letter to Ross Greenberg, the author
of Flushot.
He had the following to say:
>"Well, Matt, I'm sorry that you found the program to be less than you
>expected. You certainly got your money's worth, though, didn't you?
Sure Ross, A FREE CMOS TRASHING! Sounds like a bargain. Anybody
else want one? Can I get a FAT scambling for an extra buck?
>I guess that part of the manual (the one with my phone number) must
>have escaped your astute observations as well as the "How to Use
>Flu_Shot" section must have.
I guess my calls and uploads (virusck.exe) to your board escaped
your observation, as well as others who tried to contact you
as well and got no response.
> But, I guess with people like you around, I should just stop enhancing
>FLU_SHOT, or trying to protect *you* from the bad guys. Hell, I can't even
>protect you from yourself.
Gee Ross, does that mean I am a WORM?? Is my fate that of the
WORM in the FLUSHOT manual?
Who are the bad guys, the guys that use programs without
testing them or the ones that release them to the public
without testing them?
> Look, the program does try to do a lot. One area I'v had consistant
>trouble with has been CMOS. It'll get pulled in the next release. Not
>because some people didn;t find it useful. Just because the bitching from
>the people who had problems with it isn't worth the lousy $10 that the other
>people pay. If you don't like it, don't use it. I'm certain that I won;t
>lose any sleep over it.
Gee I don't see why they are bitching. Just because they are
supposed to send you $10 after your program wipes out information
on their computer is no reason to be upset, huh? If you had
problems with CMOS, why did you unleash them on the world in
the first place?
"How's the asbestos suit. Any 'char's?"
The preceding statements are opinions only and are not to be
construed as offensive hard facts under penalty of the
misinformation act.
Matt Cohen (matt@psuhcx.psu.edu, matt@psuecla)