Re: Rubywand's new award! Plus, Answers

[Rubywand <rubywand@swbell.net>]

Jeff Blakeney wrote ...
> 
> On Fri, 27 Feb 1998 17:51:54 -0600, Rubywand <rubywand@swbell.net>
> wrote:
> 
> >No fair! Tripping over mutilated, sliced-up half-wits is not the
> >same as being shot down.
> 
> -= I =- take VERY personal offense at this statement.  I'm sure many
> others do as well.  I happen to think that my points, which you never
> bothered to address as far as I have seen, were very good, to the
> point and filled with FACTS.
>

     You have my apology for that. Your post ended up with the title
stuck off to the left of my Netscape display; and, since it was in a
relatively peaceful thread, went unnoticed.


     Now that your post has, just now, been read, it is unclear why you
should feel offended by the, admittedly picturesque, "mutilated bodies"
metaphor.

     Your post clearly states your position in a civilized manner
without resorting to deceit or insults. Although we disagree on some
points, there would be no motivation for any kind of direct or implied
hostile response.

     
 
> Basically, you just opened yourself up for flaming from a lot of
> people.
>

     One never knows for sure. Most likely, those offended are the ones
who were _supposed_ to be offended. 


 
> >     You mean like a game that causes machines running the standard OS--
> >the one all those other programs run on with no problem-- does not have
> >a BUG? Perhaps it has escaped your notice that there is _no_ popular,
> >installed OS under which the particular game will run without bombing
> >some user systems.
> 
> What is your hang up with the "standard OS"?  What exactly does that
> mean?

     It means System 6.0.1 without a lot of weird, undocumented,
untested add-ons.

     The "hang-up" with the standard OS is that it is the one which the
vast majority of users will have installed. It is, consequently, the
environment in which a typical GS program is expected to run without
crashing.



> All the operating systems that I have ever used have had bugs
> discovered and patches and new versions followed to fix those bugs.
> If version 7.0 of GS/OS were to come out, this game would run fine
> under it because it was written to follow the rules.  If the bug that
> is being triggered is fixed in this new version, Wolf3D would work
> fine with no modification necessary.
>

     You may wish to re-read your above statement. "If the bug that is
being triggered is fixed ... Wolf3D would work ...."?

     You admit that Wolf3D does not work correctly under any available
OS. Even were an OS patch in existence somewhere now, it would be months
before just those reading csa2 would be likely to adopt it.

     Which makes better sense to you:  1. Release a game which will run
correctly under the OS users have as well as under a future patched OS; 
or   2. Release a game which will run correctly only under some future
OS which nobody currently has?


 
> I'm curious to know, do you not have any of the publically available
> patches for System 6.01 installed on your machine?

     Two. One is an INIT patch for a Text Edit bug and one is a patched
HFS driver. 


>  If not, you are
> opening yourself up to a lot of problems and headaches that you could
> avoid if you installed these patches to the "standard OS".

     There is no disagreement regarding the desirability of repairing OS
flaws. 



>  By the
> way, there are many programs that don't work under "standard OS"
> either because they were written for older versions of the OS and
> didn't follow the rules or because they require add-ons to the
> "standard OS" to operate (ie. Spectrum, The Manager and TransProg III
> require the Heirarchic init or CDEV, Music Composer requires
> Clays.MTools, TCP/IP programs require the Marinetti CDEV).
>

     So, what? Is the fact that some old programs bomb under the current
OS a reason to release a new program which does, too?

     As for programs, like Spectrum, which require vendor-supplied
add-ons, one observes that the add-ons were supplied. Where was Wolf3D's
add-on when it was released?

  
 
> As far as I know, for Wolf 3D to crash your system you have to have
> the Heirarchic INIT installed on your system. ....

     The truth is, we do not know how widespread the bug's effects may
be. We have already identified one fairly popular configuration--
Spectrum 1 installed + the Hierarchic INIT; but, there may be other
vulnerable configurations. Nor can we be sure that the bug's effect is
always just a system crash upon exit



> It may have escaped
> your notice but the Heirarchic INIT was upgraded (my version is 1.6.3
> dated 18 June, 1997) and is now a CDEV.  Wolf 3D does not crash
> machines that are using the new CDEV version, which is backwards
> compatible with the INIT, so is it any wonder that only one playtester
> was affected?
>

     Thank you for the information. Where can one download this CDEV?
Installing it sounds like a better workaround for the Wolf3D bug than
moving Hierarchic to the Spectrum folder.
     

 
> >     By the way, as you may recall, we worried about, identified, and
> >discussed the bug for roughly a week on the world's largest Apple II
> >newsgroup without getting a word of feedback from the creator.
> >
> >     At the time, this caused some bad feelings because practically
> >everyone assumned that, of course, after posting the announcement on
> >csa2 and releasing the game, the author word be monitoring csa2 for user
> >responses.
> 
> Maybe because he doesn't usually read this group and just posted the
> notice about the programs availability as a consideration for the
> Apple II users here that don't read messages on Genie or Deplhi or
> check his web pages regularly?
>

     Really, Jeff, posting a IIgs New Game announcement only on Genie or
Delphi would hardly make sense. One might just as well stuff the press
release into a bottle and toss it into the Pacific Ocean.
"Consideration" or not, unless Wolf3D's release was supposed to be Top
Secret, it had to be posted to csa2.

 
> Maybe he was waiting for more reports of problems in case it was just
> one or two people who could be helped individually.
>

     Maybe he happened to be counting on his Delphi buddies who frequent
csa2. If he did, he was out of luck.

 
> Maybe he was sick and couldn't use the computer.
> 
> Maybe his ISP had system problems so he couldn't connect.
> 
> Maybe his keyboard broke from all the coding he did on Wolf.
>

     Maybe he should have bothered to check user postings on the largest
Apple II newsgroup. It takes, maybe, 30 seconds to click on
comp.sys.apple2 and get a messages download.
     
 
> There are any number of reasons that he took <sarcasm mode on> two
> whole days <sarcasm mode off> to reply.

     Correction. The first known bug report was posted on 2/17. This was
followed by the first posting with "bug?" in the heading early on 2/18.
There followed over twenty postings before the author posted a response
late on 2/22. The delay was more than _four_ days, not two.

     By the way, it is interesting that everyone seems to automatically
assume that hanging out on Delphi means no exposure to bug reports. Is
everyone there asleep?



>  I really don't think Sheppy
> should be blamed for not supporting his product because YOU are
> impatient.  A lot of people don't read the newsgroups constantly.
> Some only read them once a week.  This is a fact and has nothing to do
> with the quality of his program or his support for it.
>

    "Blame" only became an issue when those involved with Wolf3D's
release started claiming they were not to blame for anything. The delay
in responding to the posted reports on the largest (and, practically
speaking, only) Apple II newsgroup is blamed upon users who "failed" to
send emails. (Remember, Nathan's name is, like, on the credits? Every
post _was_ a report directly to the last person to actually work on the
program.) 

     The crash bug was blamed on the OS.

     Ask yourself: "What would Wolf3D's chief programmer have done had
the crash problem been detected a few days before release?"

     Answer: He would have changed _the_ _program_ so that the game did
not crash. This is the same approach any game programmer would take.


     The original postings pointing to the problem merely asserted that
there was a bug in the game. It was those who railed against this
assertion who insisted upon equating it to "blame" and began a series of
personal attacks equating the presentation of the bad news with
"ingratitude".

     Meanwhile, not one of those involved with the game expressed the
least gratitude for having the bug exposed in time to avoid becoming
responsible for many users taking apart their systems. When the
interaction with Hierarchic was identified and a workaround posted, not
one of those connected with Wolf3D expressed gratitude for their game
becoming runnable without crashing. 
     

 
> You also may have noticed in the Wolf 3D docs under the heading Help &
> Technical Support, Sheppy states that the best place for assistance is
> the Apple II Forum on Delphi.  For updates, news, utilities and stuff
> go to his web page.  He also lists his e-mail address for support
> questions.  Nowhere is there any mention of his intention to provide
> support here in comp.sys.apple2.

     Possibly, your presentation could have benefited by omitting the
above. It makes the author look like a nitwit.



>  The fact that he has is doing so
> says more about his intentions to solve any problems than your
> constant complaining about a problem that is NOT in his code.
>

     The fact that he is doing so shows that, despite original
intentions, he has adjusted to Reality and is not a nitwit. 


 
> It shouldn't surprise many of the long time readers here that many of
> the Apple II developers gave up on this newgroup because of the abuse
> they had to deal with.  Kind of reminds me of some of the messages
> being posted lately.
>

     To be sure, posting a bug report, posting a convenient workaround,
and posting an alternative, experimental game patch are serious abuse.

     
 
> >     That the CALL is risky does not necessarily make using it a bad
> >idea. It may, in some cases, be a very clever idea. But, really, this is
> >not the kind of thing you tack onto a major game just hours before
> >release. There are too many things which can go wrong and no time to do
> >the necessary testing.
> 
> You have been spewing this line all over the newsgroup.  Where exactly
> did you get this information?  I was in chat on Delphi with Sheppy
> when he discovered the problem with the RefreshDesktop call and he
> decided to use the SetTSPtr call to make RefreshDesktop do what he
> wanted.  This was at least one WEEK, if not two, before the program
> was released and the testers did get to test this change.
>

     The expression "last minute" was first used by me on 2/23:

>>
... While scanning
the program earlier in the evening, I came upon information which
indicates that, probably, the bug was the result of last-minute changes
and so escaped play-testing.
<<

     As you can see, the statement does not assert that the bug was a
last-minute addition. It is framed as speculation. The same kind of
qualification was used for "last minute" in discussing the matter with
the author; and, it was not corrected. If you or anyone else noticed the
expression "last minute" and did not like it and knew better, you had
only to email me or post an immediate correction.

     Only recently have details of the bug-introducing modifications 
emerged. Evidently, when "last minute" was believed to mean hours it was
'okay' because it helped defuse the notion that a bug had been
identified during regular testing and allowed to exist in the released
game.

     It is only when the focus shifts to programmer carelessness that we
are, now, told that, 'well, it wasn't hours, it was about a week'. You
and others, evidently, consider this to be important. Yet; you permitted
us to believe the 'last minute modification' scenario until it suited
your purposes to make a correction.


 
> There was a total of one out of ten playtesters that ran into this
> problem and the error couldn't be reproduced by others so they figured
> it was a rare setup that would cause the problem.  As it was not a
> fatal or dangerous problem (didn't initialize disks or anything, just
> makes you reboot) it wasn't that big of a concern.
>

     Certainly, a week of testing is better than a few hours. And, it
paid off. The bug was detected on at least one machine, but, evidently,
not identified. Under time pressure from the approaching deadline, the
game was released.

     Within three days of release, at least three users posted crash
reports on csa2.

     You say "it wasn't that big of a concern" because it "just makes
you reboot". Nice try. All the programmers knew was that the bug caused
a non-destructive system crash on one machine. _If_ this happens, the
user gets to reboot.

     That the test produced only one bug sample should have been enough
reason for setting back the release. With a mystery bug and just one
machine to present a failure sample, the programmers had no basis for
being confident that the bug would not, in fact, Init or otherwise trash
hundreds of user hard disks.

 
> Get a grip, relax and learn to be a little more patient.  You usually
> won't get instantaneous responses in newsgroups.  You can report a bug
> here but then it might take a week for you to get a response because
> it has to make it to the person responsible, they need to verify it
> and respond and then it has to get back to you. ....

     Reminder: We are not talking about AOL or Delphi or Genie. There is
no reason for messenger pidgeon delays on csa2. The programmers already
had good reason to expect problem reports and, sure enough, the problem
reports were posted.



> Especially when the
> documentation doesn't list this group as a place to get support.

     We are kind of down on supporting Bugs.


>  I'm
> surprised you don't seem to realize this.

     As a matter of fact, you probably are surprised.



> You've been around a long
> time.
>

     Perhaps not as long as you; but, sometimes, it seems like forever.


 
> The cool thing about all this is that the bug was discovered and being
> worked on before you mentioned anything ....

     Let this be clear, your words are not being taken to be an
admission that the game bug had been clearly identified so early. That
may be what you seem to be saying; but, it would not be consistent with
your position. You are, evidently, saying that the OS bug related to the
game crashes had been identified and that an OS fix had, very early,
been decided upon as a means to fix the game crashes.

     Even so, your statement helps explain why nobody cared about
monitoring csa2. Why should those involved with the program bother if
they've already decided that a fix is necessary and that it will be to
the OS. There was never any real interest in game bug reports, was
there? Indeed, they were actually an embarassing annoyance, weren't
they?

     Then, too, why bother informing users about your pre-release
detection of a problem until a week later when that "ungrateful"
Rubywand person forces the issue?

     Like, if you don't care a flying flip for the worry and possible
damage experienced by non-insider users, why say anything until, voila!,
you are good and ready to try forcing adoption of a new GUPP?


     For sure, your explanations have helped to complete the picture.

     Thank you for your post.


Rubywand