Delivering the OS/2 API Commitment: Difference between revisions
No edit summary |
mNo edit summary |
||
(2 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
''by [[Jeri Dube]]'' | ''by [[Jeri Dube]]'' | ||
When IBM started talking about developing a 32-bit version of OS/2 (in the incarnation of OS/2 Version 2.0), developers were | When IBM started talking about developing a 32-bit version of OS/2 (in the incarnation of OS/2 Version 2.0), developers were sceptical as to why they should bother to develop 32-bit applications or convert their 16-bit Windows or OS/2 applications. This is when IBM started to state that the OS/2 32-bit API set would be the path to the future. | ||
==Portability, Durability, and Scalability== | ==Portability, Durability, and Scalability== | ||
The OS/2 promise was for 32-bit API portability, durability, and scalability (from | The OS/2 promise was for 32-bit API portability, durability, and scalability (from 4 MB systems to 16-way multiprocessors). By promising API portability, IBM was assuring developers that if they were faithful in their use of the 32-bit API set, they would be able to easily port their source from one hardware platform to another with only a recompile. This implied that developers need to maintain a single OS/2 32-bit program source to support and keep multiple hardware platforms up-to-date. | ||
The durability promise meant that IBM would not abandon the OS/2 32-bit API set. This was a strategic API set that would endure for many years. These 32-bit APIs would be robust enough to withstand the phenomenal rate at which hardware technologies change. Not only would these APIs support 32-bit hardware, but they would extend to 64-bit hardware as well. | The durability promise meant that IBM would not abandon the OS/2 32-bit API set. This was a strategic API set that would endure for many years. These 32-bit APIs would be robust enough to withstand the phenomenal rate at which hardware technologies change. Not only would these APIs support 32-bit hardware, but they would extend to 64-bit hardware as well. | ||
Line 16: | Line 16: | ||
* OS/2 for SMP | * OS/2 for SMP | ||
* OS/2 Warp | * OS/2 Warp | ||
The promise of durability and scalability has been more richly fulfilled with each of these releases. And now as the initial testing and demonstration phases of OS/2 for the PowerPC progress, the promise of portability to another hardware platform is fulfilled for the first time. Developers such as Sundial Systems Corporation and Pinnacle Technology have proven that their 32-bit OS/2 applications can be ported quickly and easily from the Intel hardware platform to the PowerPC platform. When using the Alpha version of the Software Developer's Kit, these two developers found just how easy it was to port applications that were written using only OS/2 32-bit APIs to the PowerPC. As OS/2 for the PowerPC moves into the Beta phase, many other developers will have the same satisfying experience of a promise being fulfilled. | |||
The promise of durability and scalability has been more richly fulfilled with each of these releases. And now as the initial testing and | |||
Both '''Dr. Randell Flint''' of '''Sundial Corporation''' and '''Charles Dircks''' of '''Pinnacle Technologies''' changed little to no source code and completed the conversion within six days of beginning the task. Both developers were absolutely amazed and delighted at the incredibly fast results from their porting effort. Randell Flint said, ''"It is hard to call this a conversion. It just happened."'' | Both '''Dr. Randell Flint''' of '''Sundial Corporation''' and '''Charles Dircks''' of '''Pinnacle Technologies''' changed little to no source code and completed the conversion within six days of beginning the task. Both developers were absolutely amazed and delighted at the incredibly fast results from their porting effort. Randell Flint said, ''"It is hard to call this a conversion. It just happened."'' | ||
Line 25: | Line 24: | ||
Making the move to 32-bit provides additional benefits, such as: | Making the move to 32-bit provides additional benefits, such as: | ||
* Improved performance because of an increase from | * Improved performance because of an increase from 64 KB to 4 GB of near access data/code | ||
* Less cumbersome and time-consuming storage management tasks, since 32-bit allows for paging rather than segment management | * Less cumbersome and time-consuming storage management tasks, since 32-bit allows for paging rather than segment management | ||
* The ability to more easily manage large chunks of data for spreadsheet, graphic, and other data-intensive applications | * The ability to more easily manage large chunks of data for spreadsheet, graphic, and other data-intensive applications | ||
* An increase in unconstrained logical object size from | * An increase in unconstrained logical object size from 64 KB to 4 GB | ||
OS/2 developers now continue to reap the benefits of the investments they have made in OS/2. These benefits will only compound as the OS/2 promise of portability, durability, and scalability provide an ever-broadening customer base for OS/2 applications. The OS/2 investment is an investment in a strategy that always made sense and now has proven that it works. | OS/2 developers now continue to reap the benefits of the investments they have made in OS/2. These benefits will only compound as the OS/2 promise of portability, durability, and scalability provide an ever-broadening customer base for OS/2 applications. The OS/2 investment is an investment in a strategy that always made sense and now has proven that it works. | ||
{{IBM-Reprint}} | |||
[[Category:Developer Connection News Volume 6]] | |||
[[Category:Developer Connection News |
Latest revision as of 00:24, 4 November 2022
by Jeri Dube
When IBM started talking about developing a 32-bit version of OS/2 (in the incarnation of OS/2 Version 2.0), developers were sceptical as to why they should bother to develop 32-bit applications or convert their 16-bit Windows or OS/2 applications. This is when IBM started to state that the OS/2 32-bit API set would be the path to the future.
Portability, Durability, and Scalability
The OS/2 promise was for 32-bit API portability, durability, and scalability (from 4 MB systems to 16-way multiprocessors). By promising API portability, IBM was assuring developers that if they were faithful in their use of the 32-bit API set, they would be able to easily port their source from one hardware platform to another with only a recompile. This implied that developers need to maintain a single OS/2 32-bit program source to support and keep multiple hardware platforms up-to-date.
The durability promise meant that IBM would not abandon the OS/2 32-bit API set. This was a strategic API set that would endure for many years. These 32-bit APIs would be robust enough to withstand the phenomenal rate at which hardware technologies change. Not only would these APIs support 32-bit hardware, but they would extend to 64-bit hardware as well.
Last but not least, the scalability promise meant that applications serving a broad range of processor sizes and functional roles would be supported. The OS/2 family of products would span from small mobile systems to networked client systems to fully functional servers running on a large multiprocessing system.
Getting Better all the Time
Since OS/2 2.0 was announced, there have been four additional releases. They are:
- OS/2 2.1
- OS/2 for Windows
- OS/2 for SMP
- OS/2 Warp
The promise of durability and scalability has been more richly fulfilled with each of these releases. And now as the initial testing and demonstration phases of OS/2 for the PowerPC progress, the promise of portability to another hardware platform is fulfilled for the first time. Developers such as Sundial Systems Corporation and Pinnacle Technology have proven that their 32-bit OS/2 applications can be ported quickly and easily from the Intel hardware platform to the PowerPC platform. When using the Alpha version of the Software Developer's Kit, these two developers found just how easy it was to port applications that were written using only OS/2 32-bit APIs to the PowerPC. As OS/2 for the PowerPC moves into the Beta phase, many other developers will have the same satisfying experience of a promise being fulfilled.
Both Dr. Randell Flint of Sundial Corporation and Charles Dircks of Pinnacle Technologies changed little to no source code and completed the conversion within six days of beginning the task. Both developers were absolutely amazed and delighted at the incredibly fast results from their porting effort. Randell Flint said, "It is hard to call this a conversion. It just happened."
And There's More
Working with a 32-bit API set has value beyond the promises made by IBM. After all, even other companies have realized the importance of moving from a 16-bit to a 32-bit operating system and are trying to deliver what is available in OS/2 today.
Making the move to 32-bit provides additional benefits, such as:
- Improved performance because of an increase from 64 KB to 4 GB of near access data/code
- Less cumbersome and time-consuming storage management tasks, since 32-bit allows for paging rather than segment management
- The ability to more easily manage large chunks of data for spreadsheet, graphic, and other data-intensive applications
- An increase in unconstrained logical object size from 64 KB to 4 GB
OS/2 developers now continue to reap the benefits of the investments they have made in OS/2. These benefits will only compound as the OS/2 promise of portability, durability, and scalability provide an ever-broadening customer base for OS/2 applications. The OS/2 investment is an investment in a strategy that always made sense and now has proven that it works.
Reprint Courtesy of International Business Machines Corporation, © International Business Machines Corporation