NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/os/os2/advocacy/12735

Text File | 1993-01-24 | 9.8 KB | 210 lines

Newsgroups: comp.os.os2.advocacy
Path: sparky!uunet!gatech!destroyer!fmsrl7!lynx.unm.edu!umn.edu!csus.edu!netcom.com!xtifr
From: [e-mail] (Chris Waters)
Subject: Re: PRO/CON on Mirrors (was Re: AmiPro for OS/2)
Message-ID: <[e-mail]>
Organization: Netcom - Online Communication Services  (408 241-9760 guest)
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Distribution: world,local
Date: Mon, 25 Jan 1993 02:24:53 GMT
Lines: 198

In <[e-mail]> [e-mail] (Robert S. Rodgers) writes:

>In article <[e-mail]> [e-mail] (Chris Waters) writes:
>>In <[e-mail]> [e-mail] (Yamanari) writes:
>>
>>>In article <[e-mail]> [e-mail] (Greg Franklin) writes:
>>>>In article <[e-mail]>, [e-mail] (Timothy F. Sipples) writes:
>>>>>
>>>>> AmiPro for OS/2 2.0 will be full 32-bit, Workplace Shell aware,
>>>>> multithreaded, HPFS aware, etc.  Ditto 1-2-3 2.0, Freelance Graphics,
>>>>> and cc:Mail.
>>
>>
>>>    Well, now.  Will it use Mirrors?  What, in this context, does
>>>    "Workplace shell aware" really mean?
>>
>>Does it really matter?  I've heard some people saying that everyone
>>should boycott and whine about programs that use Mirrors, but I think
>>that this is ridiculous.  An app should be judged on its own qualities,
>>and not on the API used by the developers!

>    My real objection to apps that use mirrors is this:
>    Mirrors is going to be compiled into the code of the program.

Mirrors is a DLL, is it not?

>    As time passes, we can assume that newer, more efficient
>    versions of mirrors will come along and most of us (tired with
>    the lousy performance that Mirrors programs will likely
>    give--Word for PM used tricks very similar) will probably upgrade.

Upgrade the mirrors DLL?  So?  Your point?

>    As long as you're going to be running Windows code, why not buy the
>    versions that are more likely to be less buggy, updated more
>    often, and are only dependant on the quality of WINOS2 (which can
>    be updated independant of the application) for efficiency?

Because a native OS/2 app is likely to perform better than a Windows app
under OS/2?  Because a Mirrors app can use OS/2 features that are not
available to a windows program?  Because otherwise, the company may
decide that OS/2 development isn't worth the bother, since nobody seems
to be buying OS/2 programs?

>    The fact is, buying Mirrors-based products only encourages more of the
>    waffling on OS/2.

And not buying them only encourages companies to give up on OS/2
entirely.  I agree that none of the choices are necessarily optimal, but
I would still prefer to have a native OS/2 program if possible--*IF*
other factors are equal.  Which they rarely are.  As I said before, the
use of Mirrors rates barely above the color used on the box in my rating
of the importance of things.  If a Mirrors-based app works better than
it's native competitor, I will use the Mirrors app.

>    So yes, it really does matter.  If they're using mirrors, chances are
>    they're just having their Windows programmers do the job of quickie
>    patches to a windows core.  A sure fire way to get a lousy, buggy
>    product.

If it's a lousy, buggy product, it should be relegated to the dust bin
because it is lousy and buggy, not because it uses Mirrors.  I'm afraid
that I don't consider this a cause-and-effect scenario.

>>I see it as a way for companies to easily dip their toes into the waters
>>of the OS/2 market.  Companies that use Mirrors, in fact, may be the

>    Dip in, pull out at the first sign of trouble.

Which is probably what they would do in any case, except that they may
not spend the resources to even dip their toes in the first place w/o
Mirrors.  And, if Mirrors apps are rejected by consumers out of hand,
it's certain that many companies *will* pull out who might otherwise
start to make money, and realize that they could make even more money if
they addressed the OS/2 market more directly.

[OS/2]
>    needs companies willing to sit it out and build a market,
>    creating applications in the meantime that draw users
>    to OS/2.

OS/2 needs companies willing to produce apps for it, period.  This will
create attention, as well as competition.  Which will be good for OS/2.
If Mirrors helps further this goal, then three cheers to it, I say.

The most successful OS to date (MS-DOS) also has the highest number of
lousy apps.  Coincidence?  Of course not!

>    Mirrors products are the antithesis of this goal.

Having more products will hurt OS/2?  I don't understand this "logic".

>    How likely is it that a Mirrors app will run any faster than
>    a win app under the 2.1 beta?  Under the 2.1 GA?  Not very,
>    considering that the 2.1 that I've seen runs them at about
>    95% to 101% of the speed (not, of course, seamless).

But seamless is a big issue.  A Mirrors-based product will, of course,
run "seamless" in any resolution, because it is an OS/2 program, not a
Windoze program.  And with a Mirrors-based product, you don't even need
to have Win-OS2 present on your disk.  (I still have a copy on my disk,
but I haven't fired it up in over two months, and it may not be there
much longer.)

Nor is speed the only issue.  A Mirrors-based program can still use
advanced OS/2 features that are not available to mere Windoze-based
programs.  And, while there's no evidence that a Mirrors program will
run faster, there's also no evidence that I know of that a Mirrors
program will run slower.

>>ones that need the strongest encouragment and response from the market
>>to help broaden the OS/2 vendor base.  If their Mirror-based product
>>doesn't sell, they may give up on OS/2, whereas, if it does well, they
>>may switch to native API in a future version.  Thus, it might be argued
>>that it would be better to buy a Mirror-based program if other things
>>*are* equal.

>    Wrong.  Buying a mirrors based product only encourages
>    more of the same ("whatthey really want is just a straight
>    port of the Windows version").

Do you have any proof of this assertation?  I suggested one possible
scenario.  You claim that mine is false, and present your own.  I won't
deny that your scenario is possible, but I don't see it as any more (or
less) likely then mine.  In fact, I suspect that some companies will
react one way, other companies will react another.  And, to a large
degree, how the company reacts will be based on the reaction from
consumers and competitors.  This is simply not cut-and-dried.  Which is
why I don't even bother to figure Mirrors into my buying decisions.

>    If they want a strong response for the market, they can write
>    an OS/2 native versionwithout using a hack like mirrors.  Then
>    they'll see a response.  If AmiPro is a native app, you can
>    kiss DeScribe goodbye.

Even if it were a Mirrors-based app (which, I hear, it's not), you could
still probably kiss DeScribe goodbye.  Most people, like me, will base
their buying decisions on the quality of the app, not on the API used to
implement it.  If a company wants a strong response from the market, the
intelligent thing to do is write an excellent program and market it
well.  This is 100% orthagonal to the use of Mirrors.

>>But basically, I see the use of Mirrors as being *very* far down the
>>list of qualities I'll look at when evaluating a program.  If someone is


>    Speed?  Is the user interface OS/2 consistent?  Bugs?

The former and the latter are very high on my priority list.  The second
is low, but still higher than the question of whether the program uses
Mirrors.

>    All of these are not just program issues, but *mirrors* issues.

How so?  Mirrors auto-magically adds bugs?  As for speed, you may
speculate that Mirrors apps will be slower than Windows apps, but I
seriously doubt it.  Certainly, if I don't have to launch WinOS2, I'd
consider that a benefit, as well as a speed boost.

>>IMO, one of OS/2's biggest strengths is all the options it offers--DOS
>>and Windows *plus* native OS/2--and Mirrors is merely one more option.

>    Mirrors is just a sneaky way to sell Windows products to
>    OS/2 users and fool them into thinking they're getting
>    a real OS/2 product.

No, Mirrors is an easy way for a company to create an app that runs
directly under OS/2, albeit, not necessarily in an optimal fashion.  The
statement above smacks of rampant paranoia!

>>It seems like some of the loudest shouting against Mirrors comes from
>>anti-OS/2 crowd.  Do any other OS/2 *advocates* have any comments or
>>opinions about Mirrors to share?

>    It's always the anti-OS/2 crowd argument.  Some grand and global
>    anti-OS/2 conspiracy, for instance, that keeps OS/2 apps (ha)
>    out of stores.  Some omnipotent pack of OS/2 haters that keep
>    IBM from creating even remotely clever advertisements, let alone
>    honest ones.

Huh?  I asked a simply question, and get accused of seeing
consipiracies.  Did I comment about apps in the stores?  Sure, I'd like
to see them (and I think that Mirrors merely makes it more likely that
I'll soon be able to), but I'm hardly accusing any group of deliberately
blocking this.  I *have* seen OS/2 bashers on *this group* post
diatribes against Mirrors, and I simply wanted to widen the dialog.

>    Mirrors is a lot like computer companies that talk in their ads
>    about the high quality of the parts, etc--and charge twice as much
>    for the system.  When you get it home, you take a peek, and see
>    nothing but the industry standard names that your friends clone
>    has, but he paid 1/2 as much.

An interesting viewpoint.  I don't agree, but I appreciate the feedback.
It's certainly food for thought.  Obviously, this is not a simple issue.
--
Chris Waters    | the insane don't | "Imaginary guitar notes [...] exist only
[e-mail]| need disclaimers |  in the imagination of the imaginer" --FZ