Desire to REMERGE Database and Program!!

[David W. Fenton]

Re: Desire to REMERGE Database and Program!!

pmiller@pksolut ions.com (Peter Miller) wrote in
<ikg7rv4vgk0td1 8q4pu1o0qtkg64i 2cmm4@4ax.com>:
[color=blue]
>Not only is merging a BE into a FE not
>necessary or desirable, but merging multiple BEs into a single BE,
>as you suggest as being relevant above, is likewise totally
>unnecessary.[/color]

Perhaps in the scenarios that have been explicitly discussed, but
that does not mean that there are no such scenarios where it might
be justifiable, as you seem to me to be suggesting.

I would agree that it would be the unusual situation where it was
appropriate and that one should attempt a split architecture first,
but that does not mean that particular circumstances can never
justify a different approach.

It's like with normalization. You try to normalize as much as
possible but in a particular application there may be reasons to
denormalize. I see this as precisely the same situation.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[David W. Fenton]

Re: Desire to REMERGE Database and Program!!

pmiller@pksolut ions.com (Peter Miller) wrote in
<g1l7rvkjvnreot or63m6rm48r0rm8 u9bpa@4ax.com>:
[color=blue]
>On Thu, 13 Nov 2003 17:41:41 GMT, dXXXfenton@bway .net.invalid
>(David W. Fenton) wrote in comp.databases. ms-access:[/color]
[color=blue][color=green]
>>You choose your solutions based on your assessment of the
>>requirement s. In this case, I think the requirements make the
>>unsplit back end a plausible solution.
>>
>>However, we don't actually know if Heather's circumstances meet
>>all the requirements I've outlined. But I was getting frustrated
>>that people were ignoring her question entirely.[/color]
>
>Well, now that she's posted further, it appears that many of your
>assumptions were correct.[/color]

She hasn't posted further, Peter.
[color=blue][color=green]
>>As you know, Peter, it is very common for me to question the
>>premises behind a question posted in this newsgroup and say that
>>there is a fundamental problem with the basic approach.[/color]
>
>Precisely! Which is why your initial response surprised me.
>[color=green]
>>In this
>>case, I thought the circumstances made the choice plausible and
>>felt that it was useful to post the code.
>>
>>As it turns out, parallel to my posting the code to do what
>>Heather asked, you started a discussion with her about relinking.
>>Seems to me she's learned both bits and all is right with the
>>world.[/color]
>
>Agreed.
>[color=green]
>>Why are you in such a snit about it?[/color]
>
>I guess we just saw (and see) it differently. I think a fe/be
>split is a very important aspect of good application design, and
>like most good application design aspects, serves a wider range of
>cases better than more limited designs (like in this case, a
>single file approach). I saw the arguments against a fe/be split
>as being trivial (ie, harder to copy two files than one), and that
>it would take more time to execute a single file solution than the
>current fe/be one. So I saw a move away from a good general
>design principle being made for weak reasons and requiring extra
>work. What isn't there about that combination to raise one's
>hackles?
>
>That said, with further clarification from Heather, I don't see a
>problem with the single-file approach in this limited scenario,
>especially if the app is to be run from cd.[/color]

Er, the clarification that talked about the CD was, in fact, in the
message to which I replied with the air code to do the job of
unsplitting. And it was the same message that you replied to
explaining about having code for reconnecting automatically.

There has been no further clarification beyond the information both
of us had available to us at the time we replied in parallel.

But, apparently, *you* didn't read her message carefully, since you
didn't see all these "clarifications " that I did see.

This is why it has seemed to me that you'd gone off for no good
reason, as you now acknowledge that Heather's scenario makes
unsplitting a plausible alternative (though not necessarily the
best solution based on the information we had). In short, it seems
to me that you went off on a tear because of a predisposition that
caused you not to read Heather's message carefully, and therefore
to ignore the information that I accounted for in my reply, which
you criticized as perhaps "toying" with her.

Heather has not posted to this thread since her replies to your
solution and mine, and I have no more information about her
situation than I had when I posted my code, and that you had when
you posted your suggestion. What we do have is discussion of that
incomplete scenario, and my additions to the scenario that you
acknowledge make the unsplitting a reasonable (if not optimal)
alternative.

So, basically, at this point you agree that I was providing an
answer to a reasonable request, and not being whimsical or
malicious. But you had all the information you needed to understand
that at the time you sent your "toying" post.

I'm not offended or anything.

I just think there's a lesson here that we need to read posts more
carefully, and when you see someone like me making a post that you
see as uncharacteristi c, you should perhaps consider that I gave it
some thought and chose my answer to fit the question asked.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[Peter Miller]

Re: Desire to REMERGE Database and Program!!


David,

On Fri, 14 Nov 2003 20:16:02 GMT, dXXXfenton@bway .net.invalid (David
W. Fenton) wrote in comp.databases. ms-access:
[color=blue]
>She hasn't posted further, Peter.[/color]

<snip>
[color=blue]
>Er, the clarification that talked about the CD was, in fact, in the
>message to which I replied with the air code to do the job of
>unsplitting. And it was the same message that you replied to
>explaining about having code for reconnecting automatically.
>
>There has been no further clarification beyond the information both
>of us had available to us at the time we replied in parallel.[/color]

Not true. Heather posted again yesterday morning, prior to my posts
and many of yours, stating in depth what her actual situation was.
[color=blue]
>But, apparently, *you* didn't read her message carefully, since you
>didn't see all these "clarifications " that I did see.[/color]

Before you go off on a rant, check groups.google.c om. You're
obviously mistake, and so your conclusions about my misreading things
are obviously off base.
[color=blue]
>Heather has not posted to this thread since her replies to your
>solution and mine, and I have no more information about her
>situation than I had when I posted my code, and that you had when
>you posted your suggestion. What we do have is discussion of that
>incomplete scenario, and my additions to the scenario that you
>acknowledge make the unsplitting a reasonable (if not optimal)
>alternative.[/color]

Why do you insist that Heather never followed up? Not only are you
wrong, but if there was any confusion about who may have seen which
posts, you could easily have verified this by a google search.
[color=blue]
>So, basically, at this point you agree that I was providing an
>answer to a reasonable request, and not being whimsical or
>malicious. But you had all the information you needed to understand
>that at the time you sent your "toying" post.[/color]

Wrong.
[color=blue]
>I'm not offended or anything.
>
>I just think there's a lesson here that we need to read posts more
>carefully, and when you see someone like me making a post that you
>see as uncharacteristi c, you should perhaps consider that I gave it
>some thought and chose my answer to fit the question asked.[/color]

<chuckle>

Yes. *We* do.

Peter Miller
_______________ _______________ _______________ _______________
PK Solutions -- Data Recovery for Microsoft Access/Jet/SQL
Free quotes, Guaranteed lowest prices and best results
www.pksolutions.com 1.866.FILE.FIX 1.760.476.9051

[David W. Fenton]

Re: Desire to REMERGE Database and Program!!

pmiller@pksolut ions.com (Peter Miller) wrote in
<rlmarvcp6ggs72 v76ua6qkgkv20dl 3f1t0@4ax.com>:
[color=blue]
>On Fri, 14 Nov 2003 20:16:02 GMT, dXXXfenton@bway .net.invalid
>(David W. Fenton) wrote in comp.databases. ms-access:
>[color=green]
>>She hasn't posted further, Peter.[/color]
>
><snip>
>[color=green]
>>Er, the clarification that talked about the CD was, in fact, in
>>the message to which I replied with the air code to do the job of
>>unsplitting . And it was the same message that you replied to
>>explaining about having code for reconnecting automatically.
>>
>>There has been no further clarification beyond the information
>>both of us had available to us at the time we replied in
>>parallel.[/color]
>
>Not true. Heather posted again yesterday morning, prior to my
>posts and many of yours, stating in depth what her actual
>situation was.[/color]

OK, I didn't see that post until after I'd replied to you, because
my newsreader sorts threaded by oldest post in a subthread. Her
post stood at the top of a subthread (because the post it was a
reply to had alread been marked as read), so it sorted last.

But it didn't supply anything but confirmation of the suppositions
I had already made based on the information she *did* provide.
[color=blue][color=green]
>>But, apparently, *you* didn't read her message carefully, since
>>you didn't see all these "clarifications " that I did see.[/color]
>
>Before you go off on a rant, check groups.google.c om. You're
>obviously mistake, and so your conclusions about my misreading
>things are obviously off base.[/color]

You had just as much information as I did before this last post of
Heather's, and you could not conceive of a situation in which
unsplitting was a worthwhile thing to do. Yet, based on the
information Heather had supplied, I was able to construct a
scenario that you agreed was a plausible justification for
unsplitting.
[color=blue][color=green]
>>Heather has not posted to this thread since her replies to your
>>solution and mine, and I have no more information about her
>>situation than I had when I posted my code, and that you had when
>>you posted your suggestion. What we do have is discussion of that
>>incomplete scenario, and my additions to the scenario that you
>>acknowledge make the unsplitting a reasonable (if not optimal)
>>alternative .[/color]
>
>Why do you insist that Heather never followed up? Not only are
>you wrong, but if there was any confusion about who may have seen
>which posts, you could easily have verified this by a google
>search.[/color]

While you're literally correct, her final post did not really
provide anything other than confirmation of the possible scenario I
outlined based on the information she'd originally posted.
[color=blue][color=green]
>>So, basically, at this point you agree that I was providing an
>>answer to a reasonable request, and not being whimsical or
>>malicious. But you had all the information you needed to
>>understand that at the time you sent your "toying" post.[/color]
>
>Wrong.[/color]

Yes, all the information was there, because I figured it out
myself. You're at least as smart as I am, Peter, so there's no
reason why you shouldn't have been able, based on the same
information, to construct a scenario where unsplitting was a
reasonable course of action.
[color=blue][color=green]
>>I'm not offended or anything.
>>
>>I just think there's a lesson here that we need to read posts
>>more carefully, and when you see someone like me making a post
>>that you see as uncharacteristi c, you should perhaps consider
>>that I gave it some thought and chose my answer to fit the
>>question asked.[/color]
>
><chuckle>
>
>Yes. *We* do.[/color]

Heather's final post did nothing but confirm the validity of the
scenario I constructed based on the information in her previous
posts.

We had the same information, Peter, from the beginning, yet you
couldn't see any situation that would make unsplitting a plausible
solution. Now you agree that it *is* a valid approach to the
particular problem.

I think you didn't see the possibilities precisely because of a
predisposition to a particular approach to a problem. We all have
such predispositions , but the vehemence of your disagreement with
me that unsplitting was worth contemplating leaves you in something
of an Emily Litella situation -- rather than asking further
questions, you had an unequivocal answer.

I approached the question for what it asked, based on the
information provided.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[Peter Miller]

Re: Desire to REMERGE Database and Program!!

On Sun, 16 Nov 2003 00:00:10 GMT, dXXXfenton@bway .net.invalid (David
W. Fenton) wrote in comp.databases. ms-access:

<snip>
[color=blue]
>Heather's final post did nothing but confirm the validity of the
>scenario I constructed based on the information in her previous
>posts.[/color]

As you've repeated four times, and I'll take you on your word that
although your scenario exactly matched that which Heather clarified in
her follow-up, that you did not in fact see her post before posting
the many posts where you reference the scenario.
[color=blue]
>We had the same information, Peter, from the beginning, yet you
>couldn't see any situation that would make unsplitting a plausible
>solution. Now you agree that it *is* a valid approach to the
>particular problem.[/color]

Yes.

But remember, just because I say ok, it wouldn't be a mistake to
unsplit given a slew of conditions, doesn't mean I think it should be
done. Personally, I think that if an app is in a well designed state,
it often makes more sense to keep it flexible, than to limit in a way
that may work well for a very specific implementation, but will need
to be reworked if that situation varies even slightly. If any of
Heather remote offices ever needed more active involvement in the
process, her re-merged solution fails quickly and needs to be
reworked. Had she stuck with the original better design, she could
easily handle multiple users at each remote site, the ability for them
to edit data, etc, etc.

So while I agree that the scenario outlined does not make an unmerged
implementation as crazy as it first sounded to me, I'm not in any way
suggesting I would unmerge the files.
[color=blue]
>I think you didn't see the possibilities precisely because of a
>predispositi on to a particular approach to a problem. We all have
>such predispositions , but the vehemence of your disagreement with
>me that unsplitting was worth contemplating leaves you in something
>of an Emily Litella situation -- rather than asking further
>questions, you had an unequivocal answer.[/color]

I'm sorry. I didn't see the post where you asked for further details
before posting a solution. Did I miss it?

I posted that I thought it better to rethink the need than seek the
code because that is what I thought was appropriate. I still think
she is better served by thinking through the issues than pouncing on a
workable piece of code and implementing it.
[color=blue]
>I approached the question for what it asked, based on the
>information provided.[/color]

As did I. We just had different views on the suitability of the
requested solution.

Peter Miller
_______________ _______________ _______________ _______________
PK Solutions -- Data Recovery for Microsoft Access/Jet/SQL
Free quotes, Guaranteed lowest prices and best results
www.pksolutions.com 1.866.FILE.FIX 1.760.476.9051

[David W. Fenton]

Re: Desire to REMERGE Database and Program!!

pmiller@pksolut ions.com (Peter Miller) wrote in
<g76ervgq5rqbbj jaectrj5c7kbnmd r0g31@4ax.com>:
[color=blue]
>On Sun, 16 Nov 2003 00:00:10 GMT, dXXXfenton@bway .net.invalid
>(David W. Fenton) wrote in comp.databases. ms-access:
>
><snip>
>[color=green]
>>Heather's final post did nothing but confirm the validity of the
>>scenario I constructed based on the information in her previous
>>posts.[/color]
>
>As you've repeated four times, and I'll take you on your word that
>although your scenario exactly matched that which Heather
>clarified in her follow-up, that you did not in fact see her post
>before posting the many posts where you reference the scenario.
>[color=green]
>>We had the same information, Peter, from the beginning, yet you
>>couldn't see any situation that would make unsplitting a
>>plausible solution. Now you agree that it *is* a valid approach
>>to the particular problem.[/color]
>
>Yes.
>
>But remember, just because I say ok, it wouldn't be a mistake to
>unsplit given a slew of conditions, doesn't mean I think it should
>be done. . .[/color]

I agree with that, of course.

But I thought the point of contention was whether or not I was
being malicious or not in posting the code for unsplitting. Given
that you acknowledge that there are scenarios where it is a viable
alternative (if not the one you consider optimal), I don't see that
you can say that it was any kind of mistake for me to post the
code.

And I had the same information that you had at the time I posted
that solution and had made this determination based on that
incomplete information.
[color=blue]
> . . . Personally, I think that if an app is in a well designed
>state, it often makes more sense to keep it flexible, than to
>limit in a way that may work well for a very specific
>implementation , but will need to be reworked if that situation
>varies even slightly. If any of Heather remote offices ever
>needed more active involvement in the process, her re-merged
>solution fails quickly and needs to be reworked. Had she stuck
>with the original better design, she could easily handle multiple
>users at each remote site, the ability for them to edit data, etc,
>etc.[/color]

This is a different set of questions, and I'm certainly inclined to
agree with you. However, the issue I thought we were discussing was
whether or not it was a good thing to post the code for
unsplitting. You seem to agree now that it was OK, as it's a viable
approach (if not the one you would choose personally). I post
answers to many questions about techniques I don't use in my own
coding, because I prefer other methods. I see nothing inconsistent
about this.
[color=blue]
>So while I agree that the scenario outlined does not make an
>unmerged implementation as crazy as it first sounded to me, I'm
>not in any way suggesting I would unmerge the files.[/color]

Who ever at any time suggested that you do anything of the sort?
[color=blue][color=green]
>>I think you didn't see the possibilities precisely because of a
>>predispositio n to a particular approach to a problem. We all have
>>such predispositions , but the vehemence of your disagreement with
>>me that unsplitting was worth contemplating leaves you in
>>something of an Emily Litella situation -- rather than asking
>>further questions, you had an unequivocal answer.[/color]
>
>I'm sorry. I didn't see the post where you asked for further
>details before posting a solution. Did I miss it?[/color]

I offered the solution, rather than saying "the solution you are
asking for is inappropriate." I didn't question the validity of her
approach, I simply answered her question.

You took issue with her approach absent the details, which, I
infer, you never considered.
[color=blue]
>I posted that I thought it better to rethink the need than seek
>the code because that is what I thought was appropriate. I still
>think she is better served by thinking through the issues than
>pouncing on a workable piece of code and implementing it.[/color]

She responded positively to both our posts.
[color=blue][color=green]
>>I approached the question for what it asked, based on the
>>information provided.[/color]
>
>As did I. We just had different views on the suitability of the
>requested solution.[/color]

Which is an issue that's entirely separate from the suggestion that
I was toying with her by posting a solution which you acknowledge
now is perfectly viable.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

[Peter Miller]

Re: Desire to REMERGE Database and Program!!


David,

On Sun, 16 Nov 2003 21:58:24 GMT, dXXXfenton@bway .net.invalid (David
W. Fenton) wrote in comp.databases. ms-access:
[color=blue]
>But I thought the point of contention was whether or not I was
>being malicious or not in posting the code for unsplitting.[/color]

But I never said (or meant to imply) malice on your part. I used
'toying' to imply a sense of play, yes, at the posters expense, but
not that you wished them any harm. If your post was a sincere attempt
to help someone and the lack of caveats was because you saw no problem
with such a solution (as you clearly state was the case) then so be
it. I was wrong to think you held back with your concerns. No big
deal. I *did* have concerns about such a solution, and I expressed
them. End of story.
[color=blue][color=green]
>>So while I agree that the scenario outlined does not make an
>>unmerged implementation as crazy as it first sounded to me, I'm
>>not in any way suggesting I would unmerge the files.[/color]
>
>Who ever at any time suggested that you do anything of the sort?[/color]

The implication of my comments was that since I specifically WOULD NOT
solve this problem with such a solution because of strong
disagreements about the wisdom of doing so, and since I thought most
developers would share that sense of concern (someone with a split app
wants to merge it for a situation where merging has issues and is not
required), I equate not being willing to employ such a solution myself
with not being willing to post code to do that very same thing.

You suggested that since I had accepted that a situation could exist
where merging was not a problem, I was wrong to express complete
disagreement with merging at the outset. I explained why this was not
so.
[color=blue][color=green][color=darkred]
>>>I approached the question for what it asked, based on the
>>>informatio n provided.[/color]
>>
>>As did I. We just had different views on the suitability of the
>>requested solution.[/color]
>
>Which is an issue that's entirely separate from the suggestion that
>I was toying with her by posting a solution which you acknowledge
>now is perfectly viable.[/color]

Agreed. But I've already made clear that I did not mean to imply any
malice on your part, and accept that you had none of the concerns I
had about why this was such a bad idea when you posted your solution.
I don't think it was illogical for me to consider that you might have
such concerns, and since they were not expressed in your posted
solution, I commented on that. I've apologized for any unintended
implications, and see have no interest in continuing this thread if
you are unwilling to accept that.

Actually, I have no interest in continuing this thread - period.


Peter Miller
_______________ _______________ _______________ _______________
PK Solutions -- Data Recovery for Microsoft Access/Jet/SQL
Free quotes, Guaranteed lowest prices and best results
www.pksolutions.com 1.866.FILE.FIX 1.760.476.9051