Autor: Sławomir Stachniewicz (stachnie_at_alf.ifj.edu.pl)
Data: Tue 10 Feb 1998 - 09:15:15 MET
==================BEGIN FORWARDED MESSAGE==================
>From owner-vpascal_at_as60053.pc.nus.sg Tue Feb 10 00:49:59 MET 1998
>Received: from as60053.pc.nus.sg (majordom_at_as60053.pc.nus.sg [137.132.218.180])
> by alf.ifj.edu.pl (8.8.7/8.8.7) with ESMTP id AAA16458
> for <stachnie_at_alf.ifj.edu.pl>; Tue, 10 Feb 1998 00:49:40 +0100 (MET)
>Received: (from majordom_at_localhost)
> by as60053.pc.nus.sg (8.8.5/8.8.5) id HAA16014
> for vpascal-outgoing; Tue, 10 Feb 1998 07:21:15 +0800
>Message-ID: <199802092309090160.005ACF35_at_smtp-gw01.ny.us.ibm.net>
>X-Mailer: Calypso Evaluation Version 2.40.39
>Date: Mon, 09 Feb 1998 23:09:09 +0000
>From: "Allan Mertner" <mertner_at_ibm.net>
>To: vpascal_at_as60053.pc.nus.sg
>Subject: Further clarifications
>Mime-Version: 1.0
>Content-Type: text/plain; charset="ISO-8859-1"
>Content-Transfer-Encoding: 8bit
>X-MIME-Autoconverted: from quoted-printable to 8bit by as60053.pc.nus.sg id HAA16012
>Sender: owner-vpascal_at_as60053.pc.nus.sg
>Precedence: bulk
>Reply-To: vpascal_at_as60053.pc.nus.sg
>Status: R
>
Hi again,
Thank you for all the feedback to the "Directions" mail I posted yesterday - please let me answer some of the issues raised. Also note, that although I refer to v2.0 as a product below, it does not exist, and it is still up in the air whether it will be.
- The Math unit will most likely be supported by the new v2.0.
- MMX/PPro instruction sets will not be supported in v2.0 (this can wait).
- We will not convert the C-style Open* header files for OS/2 for inclusion. If you wish, this can obviously be done by anyone with access to IBM's developer toolkits for OS/2.
- Certain 32-bit DOS-extenders is likely to be supported as well, probably including some remote debugging facility.
- We are not going to build a RAD-tool a la Delphi for OS/2 from VP; if this is what you want, I recommend you buy a copy of Sibyl. We are convinced that there is not room for two (or probably even one) product of this kind for OS/2, and Windows already has Delphi.
- We also are not going to produce Java bytecode as output in the foreseeable future. While this is a neat idea, it still suffers from lack of useability: We don't have any libraries for this - not even System, Dos, Crt, Math - not to mention the VCL, of course. And if it's such a good idea, why hasn't Borland done it already? :)
- The proposed VP v2.0 for Win32/OS2 will be based on the text-like VP/2 v1.1 IDE and will primarily target text/console-mode apps (though it is of course possible to create GUI applications with it).
- I am glad to hear support for the sell-online-proposal. I think it makes heaps of sense, and also buy most of my own software this way.
- The product currently suffers from not having a linker. Any Win32-linker capable of linking .OBJ files to produce PE-style executables can be used.
- We also do not have the Windows unit; for internal use, we use the one from Delphi - but this one obviously cannot be shipped with our product. Someone needs to spend a week converting the Windows.h file to Pascal :)
- VP v2.0 is essentially a bugfixes and optimised version of VP/2 v1.99, with the addition of Win32 and some Dos32 support. This includes a Win32 IDE capable of generating OS/2 executables and vice versa. Each IDE can debug programs in its native environment only (obviously).
- The Os2Base unit cannot (and should not) be ported to Win32; instead, the Windows unit should be used. In places where OS/2 APIs are called directly, the code should be changed to call a system-independent routine (like the RTL units we provide) that uses conditional compilation to select the correct API to call.
That should cover it for now. For those of you who have followed me this far, a question on the Win32/DOS lists floating around in the list: How do you see this pricing implemented? It's obviously nice to see that some people will pay $500 for a Win32/Dos version of VP, but I have difficulty in seeing how this can work in practice, other than by us setting a price that is low enough to allow as many of you as possible to join - but then obviously the total is much smaller as well :)
For example, the [OS/2] list contains $3460 pledges - nice! But if we are to sell the product at the lowest price suggested, $100, selling it to everyone on the list will only get us $1600 (minus various fees, etc). The same calculation for the Win32/Dos list gives a higher result: About $3400 at a price of $200. Food for thought.
Again, thanks for listening.
Yours,
Allan Mertner
===================END FORWARDED MESSAGE===================
+-------------------------------------+
| Slawomir Stachniewicz |
| mailto:stachnie_at_alf.ifj.edu.pl |
| http://solaris.ifj.edu.pl/~stachnie |
+-------------------------------------+
To archiwum zostało wygenerowane przez hypermail 2.1.7 : Tue 18 May 2004 - 15:16:18 MET DST