Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!udel!princeton!pucc!EGNILGES
From: EGNILGES@pucc.Princeton.EDU (Ed Nilges)
Newsgroups: comp.sys.mac.hypercard
Subject: Re: Hypercard 2.0
Message-ID: <12209@pucc.Princeton.EDU>
Date: 8 Jan 91 01:13:32 GMT
References: <12208@pucc.Princeton.EDU> <47827@apple.Apple.COM>
Reply-To: EGNILGES@pucc.Princeton.EDU
Organization: Princeton University, NJ
Lines: 106
Disclaimer: Author bears full responsibility for contents of this article

In article <47827@apple.Apple.COM>, jkc@Apple.COM (John Kevin Calhoun) writes:

>
>In article <12208@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU
>(Ed Nilges) writes:
>>It's pretty arrogant to convert a pre-2.0 stack to 2.0 format, such
>>that the stack CANNOT be used with an older version of Hypercard and
>>without any backup...
>
>Of course, the reason you can't open converted stacks with HyperCard 1.x
>is that the file format has changed to include new information, such as
>window location, styled text in fields, shared hilight states for
>background buttons -- lots of things to support the new features of 2.0.

You could have done this such that the format was upward compatible:
such that stacks would work with 2.0 and older releases, as long as
they did not explicitly require the new features.  You chose not to,
and I feel that marketing considerations played a role in your decision.
Either the pre-2.0 stack format did not accomodate expansion, which
was a bad decision at that time, or else the 2.0 stack format did not
take advantage of expansion capability...which is a bad decision NOW.

I am speculating since I am not privy to either stack format.  I'd
welcome insight into that, but I understand that Hypercard is not
open-architected at the level of the stack.
>
>In HyperCard 2.0, when you choose "Convert Stack..." from the File menu,
>you get a dialog that says "Convert stack to 2.0 format?".  This is
>inadequate, for two reasons:
>
>1) The possibility of problems in running 1.x stacks under 2.0, though
>   small, is real.
>2) Users didn't get the manuals that describe stack conversion and
>   advise making backups, because HyperCard 2.0 is being distributed
>   in a way very different from the way we thought it would be.

Thanks for your honesty.  I am sitting here with a DayTimer stack of
my own devising (I am responsible, by the way, for using that copyrighted
name to describe this noncommercial stack) that has effectively been
trashed by Hypercard 2.0.  It is in 2.0 format, I cannot convert it
back to 1.x format (please advise if there's a utility), it runs a
LOT slower because it does a lot of text editing, and I use a shared
copy of Hypercard 2.0 on a file server that is not always available.  I
should have made a copy of the stack before the convert, especially
since I am aware of Hypercard's overall philosophy, which is to treat
disk and RAM as a one-level store, altering disk without telling the
user: but it still burns my ass that I have to use 2.0 now all the
time.

I suppose I have a way of getting my own copy of Hypercard upgraded
to 2.0, but it's still slower for MY applications and I prefer, being
a control freak, to upgrade when I am good and ready.
>
>Therefore, in HyperCard 2.0v2, we changed the dialog.  It now says,
>"Only HyperCard 2.0 and later versions can open converted stacks.  You
>may want to save a copy before converting this stack."  I hope this is
>sufficient warning for new users.  The "Save a Copy..." menu item is
>in the same menu as "Convert Stack...", so there's no problem in making
>a backup before conversion.

Better, but I have been using Hypercard ever since it was announced
and I did the ConvertStack using 2.0v0, so it is not only "new users"
who need this warning.  I feel it would be best to provide an undoConvert,
an automatic backup, or a utility to convert back to 1.x format, and
it still seems that market considerations drive the decision not to
provide these tools.
>
>>                                                     ...right now
>>I'm pissed (angry pissed, not squiffled pissed.)
>
>Yes, I can tell, because you've made two generalizations from scant
>evidence:  2.0 behaves differently with all stacks that work under
>older versions, and 2.0 is slower than 1.0.  Neither of these is true.

I just reviewed the initial post.  I did say that 2.0 is slower.  That
has been my consistent experience on a 68020 platform.  Have you tried
Hypercard on such a slightly older platform?

I actually qualified my claim that Hypercard runs different, not just
more slowly.  I said that it appears to be the case.  I found that
"go to it", for example, works differently when "it" contains "card
id nnnnn".  If you want more detail I will create a test stack and
send it to you.
>
>The differences between 1.0 and 2.0 that may cause problems with older
>stacks are really very minor.  I'd be happy to help you figure out what
>the problem is with the stacks you've converted.
>
>As for HyperCard 2.0 being slower than 1.x, it's true that for some
>operations, especially those that require support from TextEdit, 2.0
>is slower.  But, overall, 2.0 is easily faster than 1.0.

AAaaaargh, that's precisely what I do a lot of...text editing.  My
stacks consist of a lot of functions running on top of Hypercard
string manipulation...things like "getPhrase" to get word n thru m
from a string, and so on.  Does anybody have an XCMD/XFCN of C-like
text facilities?

Where's my Hypercard compiler?
>
>Kevin Calhoun
>HyperCard Team
>Apple Computer, Inc.

Thanks for your reply, Kevin!  Hate to be such a pain in the butt,
but I actually like Hypercard.