• Welcome to OS2World OLD-STATIC-BACKUP Forum.
 

News:

This is an old OS2World backup forum for reference only. IT IS READ ONLY!!!

If you need help with OS/2 - eComStation visit http://www.os2world.com/forum

Main Menu

Ecomstation 64 Bit Kernel

Started by PAUL555, 2008.04.07, 05:45:52

Previous topic - Next topic

saborion2

#45
Re:
Quote
Actually, nothing you point to says anything similar. Actually, if you read into some of those articles (and others on the Internet), you will see that early NT morphed from early OS/2 code. While OS/2 was written from the ground up to be a replacement for DOS based operating systems.

Regardless, we dont need to wait for IBM for an answer... the core components of OS/2 (ie: not including the device drivers and added subsystems) all hold IBM or IBM/MS or (in the most recent version) IBM/"Others" (as IBM started referring to MS) copyrights - and there are a list of IBM patents out there that are for OS/2 technologies and tied chronologically with it's development.

Certain things were ported to OS/2 after it was already long since a mature, stable OS (like the AIX firewall) - while others were re-written, nearly in their entirety based off designs and concepts from AIX (like JFS - which if memory serves, was then ported BACK to AIX as JFS2 with larger per file size limits).

From your writing (which attempt to give the impression that you are an authority when it comes to the overall knowledge of the development of OS/2 and related applications); also, the apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself; and, FYI as a long time IBM Software Business Partner repeating known facts does not appear to be very useful and helpful in cases where one is attempting to grapple with well thought our issues - hence, the appreciation, reference and usage of the OS/2 WF Bounty System/Process in place.

What is there to convince us that all the/our focus is on the "message" and that some is not diverted towards the "messenger/s".  8)

Regards.

RobertM

#46
QuoteFrom your writing (which attempt to give the impression that you are an authority when it comes to the overall knowledge of the development of OS/2 and related applications);

I claim no such position - but I would guess I am far more knowledgeable than you. Unlike you, when I started as an IBM Business Partner, they actually cared about OS/2. You entered the game far later than that.

Quotealso, the apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself;

You are welcome to express yourself - within the boundaries of accepted forum ettiquette. But when you are wrong, expect someone to point that out to you. All I asked was that instead of you posting the same thing over and over again, you post further justification. You haven't - you instead claim I am attacking you for disagreeing with you.

Quoteand, FYI as a long time IBM Software Business Partner repeating known facts does not appear to be very useful and helpful in cases where one is attempting to grapple with well thought our issues - hence, the appreciation, reference and usage of the OS/2 WF Bounty System/Process in place.

I've been an IBM Business Partner since 1995. I even helped beta test components of OS/2 long before they became available in OS/2 (such as the 32bit print engine in Warp 4) and talked directly to the developers of those subsections.

Also, "repeating known facts..." if someone asks for clarification of your point ("What's the status of COMFETAR LIVE?" "Why do you think that OS/2 wasn't a new OS but something relabelled?") then obviously they are not known facts - especially when no one else seems to agree... so enlighten us (with something different).

QuoteWhat is there to convince us that all the/our focus is on the "message" and that some is not diverted towards the "messenger/s"

Because both Robert Deed and I pointed out what we knew about OS/2 - which has nothing to do with you. Thus... "the messenger" has nothing to do with the subject, so stop taking disagreement personal.

Because I started with OS/2 with v2.0 Limited Edition (GUI and all) and the version of C-Set/2 that was released with it - thus again my knowledge and experience on the topic and my response has nothing to do with you personally.

Because I have been an IBM Business Partner for 13 years (exclusively for OS/2) - thus again my knowledge and experience on the topic and my response has nothing to do with you personally.

Because I've actually worked with the code developers at IBM to help resolve issues that are currently fixed in OS/2 (such as, already mentioned, problems in the 32 bit print engine that you and no one else saw because I and a few others tested it long before the first public beta) - thus again my knowledge and experience on the topic and my response has nothing to do with you personally.


SO......
Back to the topic at hand... the original kernel (and probably the current derivatives) are probably part MS property. And even if they are not, they are covered under the IBM/MS cross licensing agreement - either way, getting the source would be difficult. Yet, getting the source would be required in order to make a 64bit version.

The PPC kernel is supposedly a "clean room" implementation of the Mach kernel, and is a more likely candidate - but would require more work since it is written for a different platform/architecture. But, having the code to it would provide the ability to duplicate much of the functionality in the current Intel based kernel, leaving only the more recent portions to be recreated. It's a start - even if not the best of starts.

The (or a) Linux kernel is an alternative, but would cause issues in getting it as scalable and as threadable as the OS/2 kernel. As of now, much of the work on some of the new Linux threading models - as well as 64 CPU support - is actually based off OS/2's threading and CPU model (one of IBM's contribution to the Linux community). Unfortunately, it's not enough info or code to be useful to recreate the OS/2 kernel. While that means that a Linux kernel may eventually be at a usable point, it introduces numerous other issues into the mix (like updating other core Linux components (to utilize the new threading model) that would be needed to recreate OS/2 on top of them). Heck, just the granularity of process and thread levels alone on OS/2 is amazing... 4 levels, each with 63 sub-levels for priority. Then of course, you have the issue that many OS/2 subsystems use CPU Ring 2... ooops! Need to modify the kernel to handle that too - or re-write everything else as Linux components.

And all of that leads to other areas that would need to be addressed - such as writing a new PM and WPS and SOM and DSOM for Linux (and far more) which is being addressed in other threads, so I wont go into detail here about that issue.

Robert




|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


djcaetano

  Roberts,

  First of all, I would like to say I am a great fan of both of you and I really enjoy your comments on every thread, but regarding a new kernel, OS/2 for PPC parts, Voyager and such things... I would like to point out something:

  A) Most of OS/2 for PPC higher level subsystems are just Intel-OS/2 recompiled code. I had read about it in several places, but there is a book
         REICH, D. E. Designing high-powered OS/2 Warp applications - The anatomy
         of multithreaded programs. United States: John Wiley & Sons Inc., 1995.
     Where this is clearly stated. I don't know PMShell was just rescompiled, but WPS was for sure. The same should apply to MMPM/2. I believe only lower level subsystems like GRE, device drivers, file systems and such things were rewritten from scratch.

  B) Please, do not forget: the Operating System experience *is not* directly related to the kernel. Remember that OS/2 for PowerPC is MACH based, something much more like a UNIX than like an OS/2. And OS/2 for PPC, as far as I can tell, is a normal OS/2 from the users' point of view. It has a CONFIG.SYS and everything. The kernel is very important when talking about device driver model and performance (memory and CPU management), and that's all!
   Of course OS/2 kernel is great, but it is old (outdated) and has various problems. The most important: we are trapped with a limited driver model... 32 bit drivers are mostly a hack into the kernel. Also, even with access to the kernel source, it is not possible to create a direct 64-bit backward compatible kernel in the way old applications "simply run", without recompilation or any kind of virtualization.

  There is the need to use a new kernel for the future and, be it OS/2-kernel based or non-OS/2-kernel based, it *will* be incompatible (directly) with existing applications. It simply doesn't matter what kernel is used.

   And this has nothing to do with APIs or the kernel itself, it has to do with memory model and processor operating modes. The bright side is: whatever kernel is used, since OS/2 applications are based on system API calls, as long as the API is mostly preserved, it is always possible:

  1) Recompile old OS/2 applications for the new operating system, no matter what kernel was used.
  2) Run old OS/2 applications (for which we do not have the source code) inside a virtal 32 bit environment, *not* emulated, in the same way OS/2 runs DOS applications. Notice that MDOS *is not* a DOS operating system running into a Virtual Machine Box. This kind of virtualization is needed because a new 64 bit kernel should operate in 64 bit CPU mode, and it has nothing to do with the kernel used.

   Whatever the kernel is selected, if the target was running it in 32 bit mode, it would be possible to run binary OS/2 applications with a scheme similar to ODIN, but a more effective one, since it could be integrated into the kernel AND OS/2 API is not a moving target. In short, binary compatibility can be achieved seamlessly even with a foreign kernel; the problem is not what kernel, but the CPU operation mode / memory model.
 
  Maybe I am wrong, but I don't know if it is possible for 64 bit operating systems to support 32 bit applications without using some kind of virtual operation mode. But this *doesn't* mean running an emulator, nor loading the entire old 32bit-OS/2 operating system to run an application. As long as the 64 bit APIs are the same to old ones (just recompiled with different variable types), the virtual environment have just to remap 32 bit API calls to the 64 bit API calls.
  If I am wrong and it *is* possible to run 32 bit apps inside 64 bit CPU mode, then full binary compatibility is possible, but the API will have to be reprogrammed anyway. This would only be avoided if: a) 32 bits programs can run in 64 bit mode; b) OS/2 32-bit kernel source is released and c) OS/2 API code is released. But I am afraid none of these can be satisfied.

  That said, I am against Linux kernel for two reasons: Linux' threading model suck (incomplete, deficient) and Linux' device driver model suck. A user should never have to recompile the kernel just to enable a new device. But Linux is not the only kernel. There are LOTs of other kernels which *doesn't* have this problem and are very good. IBM used MACH (which was the one used for MacOS X too), but there is L4 or even Haiku kernel, which is already very good. The down side: there will be almost no device drivers available. This leaves one with the possibility of using ReactOS kernel... but hey, before we think in a kernel, there should be "things" to run on top of this kernel. We need to reprogram the API, or modify some parts of it. What about 2000's CAIRO instead of 1980's OS/2 GRE? What about re-creating WPS? When a new open WPS, using a new open graphical engine are ready, THEN one should think "on top of what will we run this beast?". THEN it will be the time to decide which kernel will be used and maybe we have, at that time, more options of well supported kernels then just NT (ReactOS) and Linux. Maybe Haiku's or even MacOS'.

  But hey... this is just the direction Project Voyager took!
  The only thing I hope is they development team keep the work and more and more people join them in the effort. Just complaining because they had not decided what kernel to use will not help at all. And I am sure of that.

   Just my 2 cents.

Saijin_Naib

Wow, epic posts. Thank you both! I've learned something here.

RobertM

#49
Hey djc!

I was thinking of writing a lengthy post commenting on each section of yours... then I realized, it would amount to "Yeah, I agree." or "Good point!" - so imagine I broke your post into sections and wrote one or both of those after each!  ;)  ;D

I actually commented on some of that stuff in my post that beat yours by seconds...

As for the kernel, one of the reasons I was interested in the PPC one (which I have little to NO knowledge about - ie: I know it exists) is because it would be a good "reverse engineering" starting place to get more hints about what and how the kernel worked (and more of a possibility of something that IBM could release - if they could find it)... and though it is based off the Mach kernel, I am sure it included several key things for Intel-OS/2 similar threading and scalability...

So... yeah, I agree! Good Points!

Robert

::)


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


saborion2

#50
Re:
Quote
I claim no such position - but I would guess I am far more knowledgeable than you. Unlike you, when I started as an IBM Business Partner, they actually cared about OS/2. You entered the game far later than that.

Quote
also, the apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself;

You are welcome to express yourself - within the boundaries of accepted forum ettiquette. But when you are wrong, expect someone to point that out to you. All I asked was that instead of you posting the same thing over and over again, you post further justification. You haven't - you instead claim I am attacking you for disagreeing with you.

Actually, I pondered for a while whether or not I should respond to your post but then again of what benefit would it be if I do not attempt to correct you with regards to some of your assumptions so much so even wanting to make a comparison with regards to time line of IBM Partner Membership in a public debate believing that such an issue is most trivial an uncalled for when more important issues should be the topic of the conversation; and, not to say the least that my earlier interface with the dumb terminal at the University of Sheffield, England during the mid-to-late nineteen seventies, the early nineteen eighties development banking and micro computing experiences through contact with Inter-American Development Bank and United Nations consultants, followed by micro-computing class instructions at Baruch College predates both of our relationships with regards to IBM Partner Membership (in my own case holding both IBM and Lotus Development Corporation Partner Memberships; but, so much for that.

Now,
Quotethe apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself;

You are welcome to express yourself - within the boundaries of accepted forum etiquette
First of all I have taken the liberty of changing of the word which I believe should be "etiquette"  ;) Why ask for certain details; and, As a point of reference here is an extract from a 1998 Lotus Development Corporation communication; 
Re:
Quote; Re: Concerning the issues with 1-2-3 that are talked about in the documentation you gave me, most of the issues are related to converting files between older and newer versions of product and converting documents between Lotus and Microsoft. Anytime a file is saved backwards or saved with an older file format than the format the file was created under, such as saving a 1-2-3 , 97 file for Windows 95 into a WK1 format for DOS, then naturally we are expected to loose certain features due to technology and features that are present now that were not present 8 - 10 years ago. Similarly, if we try to convert a file from Lotus into Excel or Excel into Lotus, due to differences in the products not every feature will be converted perfectly with the file filters that are available. Both Lotus and Microsoft create similar spreadsheet programs; however, there are several differences in both programs and these differences will remain to distinguish the products apart. We do try to design conversion filters that will allow as much of the file formats as possible to be exchanged and converted without disrupting the actual file design and format.

In one of your letters you made mention of the @IRR and @ERR functions in the 1-2-3 product. By design the @IRR (notably "absent" in Open Office) will calculate the Internal Rate of Return; where the @ERR is used in conjunction with other formulas, posted was an "ERR" showing an error was received in the calculations. As far as I can see in the program I cannot find an @ERR function that will allow us to calculate an Economic Rate of Return"
then you completely ignore the real issues at hand. Surely in your "wisdom" you do not wish us to believe that applications like the ones referenced above are expected to run on their own without the aid of the underlying Application Programming Interfaces (APIs) integrated into the Operating Systems - do you. So, at which point in the development processes of the Operating Systems are the Application Programming Interfaces (APIs) issues are expected to be addressed by developers. Were your company a candidate for a competitive bidding contract being offered by our company then your company may have just been eliminated from the process.  8)

BTW, there are words like those of "the adoption of a philosophical approach" to certain scenarios.

Enjoy the rest of your afternoon.



djcaetano

  Hi Robert,

Quote from: RobertM on 2008.04.24, 19:59:27
As for the kernel, one of the reasons I was interested in the PPC one (which I have little to NO knowledge about - ie: I know it exists) is because it would be a good "reverse engineering" starting place to get more hints about what and how the kernel worked (and more of a possibility of something that IBM could release - if they could find it)... and though it is based off the Mach kernel, I am sure it included several key things for Intel-OS/2 similar threading and scalability...
So... yeah, I agree! Good Points!

  I really think OS/2 for PowerPC not being released (even if recompiled for Intel!) was a big shame. I fell in love with its design and really think that should be the way computing should go. But this is past and we should think about future. Anyway, I agree it would be a great starting point. The problem with this approach remains the same as using a unknown kernel: we would have the need to write all device drivers, something very cumbersome.
  BUT, being OS/2 PPC kernel based on MACH kernel (although not the same "release version as MacOS X, AFAIK), maybe it's possible to add MacOS X device driver support to it in a not-so-painful way. But at this time, this would only means we would be able to run OS/2 on MacIntel computers... something that is not bad at all :) , but it is not a perfect solution also.

   Anyway, there are lots of things to happen before this decision. If IBM decide to release OS/2 PPC Kernel source code, this could be natural way to go (even if others are possible). Let's the time show us the better way.

RobertM

Quote from: saborion2 on 2008.04.24, 21:22:53
Re:
Quote
I claim no such position - but I would guess I am far more knowledgeable than you. Unlike you, when I started as an IBM Business Partner, they actually cared about OS/2. You entered the game far later than that.

Quote
also, the apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself;

You are welcome to express yourself - within the boundaries of accepted forum ettiquette. But when you are wrong, expect someone to point that out to you. All I asked was that instead of you posting the same thing over and over again, you post further justification. You haven't - you instead claim I am attacking you for disagreeing with you.

Actually, I pondered for a while whether or not I should respond to your post but then again of what benefit would it be if I do not attempt to correct you with regards to some of your assumptions so much so even wanting to make a comparison with regards to time line of IBM Partner Membership in a public debate believing that such an issue is most trivial an uncalled for when more important issues should be the topic of the conversation; and, not to say the least that my earlier interface with the dumb terminal at the University of Sheffield, England during the mid-to-late nineteen seventies followed by micro-computing class instructions at Baruch College predates both of our relationships with regards to IBM Partner Membership (in my own case holding both IBM and Lotus Development Corporation Partner Memberships; but, so much for that.

So... you answer is "but I was in college before you were." Congrats. And yes, perhaps I should have added Lotus to my list as well.

Quote from: saborion2 on 2008.04.24, 21:22:53
Now,
Quotethe apparent convenient avoidance of relevant issues when taken together give the indication/impression of the feeling of being challenged in a world in which one is free to express oneself;

You are welcome to express yourself - within the boundaries of accepted forum etiquette
First of all I have taken the liberty of changing of the word which I believe should be "etiquette"  ;) Why ask for certain details; and, As a point of reference here is an extract from a 1998 Lotus Development Corporation communication; 
Re:
Quote; Re: Concerning the issues with 1-2-3 that are talked about in the documentation you gave me, most of the issues are related to converting files between older and newer versions of product and converting documents between Lotus and Microsoft. Anytime a file is saved backwards or saved with an older file format than the format the file was created under, such as saving a 1-2-3 , 97 file for Windows 95 into a WK1 format for DOS, then naturally we are expected to loose certain features due to technology and features that are present now that were not present 8 - 10 years ago. Similarly, if we try to convert a file from Lotus into Excel or Excel into Lotus, due to differences in the products not every feature will be converted perfectly with the file filters that are available. Both Lotus and Microsoft create similar spreadsheet programs; however, there are several differences in both programs and these differences will remain to distinguish the products apart. We do try to design conversion filters that will allow as much of the file formats as possible to be exchanged and converted without disrupting the actual file design and format.

In one of your letters you made mention of the @IRR and @ERR functions in the 1-2-3 product. By design the @IRR (notably "absent" in Open Office) will calculate the Internal Rate of Return; where the @ERR is used in conjunction with other formulas, posted was an "ERR" showing an error was received in the calculations. As far as I can see in the program I cannot find an @ERR function that will allow us to calculate an Economic Rate of Return"
then you completely ignore the real issues at hand. Surely in your "wisdom" you do not wish us to believe that applications like the ones referenced above are expected to run on their own without the aid of the underlying Application Programming Interfaces (APIs) integrated into the Operating Systems - do you. So, at which point in the development processes of the Operating Systems are the Application Programming Interfaces (APIs) issues are expected to be addressed by developers. Were your company a candidate for a competitive bidding contract being offered by our company then your company may have just been eliminated from the process.  8)

BTW, there are words like those of "the adoption of a philosophical approach" to certain scenarios.

Enjoy the rest of your afternoon.

Well, I couldnt ignore the "real issues at hand" that you just brought up since you just brought them up - so let's tackle them. If you read through this thread (skipping your posts and the replies to them, that is) you will see that we have been discussing APIs and more (if even only by inference).

So, since you are now going off track with that... let me ask you this... just how many APIs do you think there are in the OS2KRNL? Just curious. You do realize that OS2KRNL isnt even a megabyte, dont you? As a matter of fact, OS2KRNL, OS2BOOT and OS2LDR combined dont equal a megabyte.

So perhaps you are talking about APIs for various subsystems such as MMOS2, WPS, PM, networking (LanMan/Samba, TCP/IP, etc), SOM, DSOM, REXX/C, etc... in which case we are once again straying off topic as this thread is about "eComStation 64 Bit Kernel"

Now of course, there are the basic drivers (BASEDEVS) that hook directly into the kernel - but that's a different topic as well - and others have already managed to replace some of them with versions the eCS community has the code for.


So once again, I am not sure what you are trying to point out...

Robert

PS: Enjoy your day too... I am enjoying mine :)


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


saborion2

#53
Re:
Quote
Well, I couldn't ignore the "real issues at hand" that you just brought up since you just brought them up - so let's tackle them. If you read through this thread (skipping your posts and the replies to them, that is) you will see that we have been discussing APIs and more (if even only by inference).

So, since you are now going off track with that... let me ask you this... just how many APIs do you think there are in the OS2KRNL? Just curious. You do realize that OS2KRNL isn't even a megabyte, don't you? As a matter of fact, OS2KRNL, OS2BOOT and OS2LDR combined don't equal a megabyte.

So perhaps you are talking about APIs for various subsystems such as MMOS2, WPS, PM, networking (LanMan/Samba, TCP/IP, etc), SOM, DSOM, REXX/C, etc... in which case we are once again straying off topic as this thread is about "eComStation 64 Bit Kernel"

Now of course, there are the basic drivers (BASEDEVS) that hook directly into the kernel - but that's a different topic as well - and others have already managed to replace some of them with versions the eCS community has the code for.


So once again, I am not sure what you are trying to point out...

You know Robert, you do make me wonder sometimes with your kind/line of questions and the real intent... as a matter of fact do you really need to be reminded of a specific "bounty" that was posted in the OS/2 Bounty System that you are now saying here again
Quote
So once again, I am not sure what you are trying to point out...
Anyway, let me attempt
Quoteto point out...
to you once more (hopefully, one less time and trusting that it is not a waste of my time) therefore to make a long story short our efforts are intended to mirror a similar UNIDO's solution offering:
Quote
COMFAR

In 1983 the first generation of the Computer Model for Feasibility Analysis and Reporting (COMFAR), a computation tool for financial analysis of investments, was released. Since then, this UNIDO software has been developed further, to support the financial and economic appraisal of projects. In 1995 the third generation, COMFAR III Expert, was introduced to the market.

Today, over 1500 licensees are using COMFAR in about 130 countries for project analysis and appraisal.

http://www.win2biz.com/comfar/default.htm

that predates most IBM or Microsoft Office products that I am familiar with. I guess that you will notice the similarity in the product naming; and, such is our focus and directions all along without your insinuations of
Quotegoing off track
Now, how about you telling us/stating; putting aside the issue
Quotejust how many APIs do you think there are in the OS2KRNL
how many Inter-American Development Bank and United Nations financial, economic and technical consultants you, along with those IBM developers you spoke about sat down with to determine what "tools" they needed to address the tasks at hand  (like it is with
QuoteCOMFAR
and not having developed products that you thought that they would need. Maybe, one of the reasons why OS/2 is where it is today ???

Not below why (from my perspective) it was felt that OS/2 was good candidate for
QuoteCOMFAR
COMFAR III Expert is a computer software that permits the user to simulate the short- and long-term financial and economic situation of investment projects. The software permits the analysis of industrial as well as non-industrial projects, whether new investments, rehabilitations, expansions, joint venture or privatization projects.

COMFAR III Expert is operable under Windows 3.11, Windows 95/98 and Windows NT. For details, please refer to the Hard- and Software Requirements.

So, what is your next question if another attempt is not to be made to hide the  COMFAR-COMFETAR LIVE relationship as well as the relevant "functionality developmental" issues under the rug as they have been for years. ;)

Regards.

Robert Deed

This is actually a reply to your other message.   

PMSHell and WPS were just recompiles from the Intel code, however any toolkit which IBM did not have the source to or permission to use on PPC processors were removed.  So theoretically it has less 3rd party code in it.  Only IBM could really answer this.

MMOS/2 as far as PPC went was never completed.  However, it was a full ground up 32bit rewrite.  Actually, OS/2 PPC didn't contain any 16bit code.. so any parts of OS/2 which are 16bit (or were at the time of OS/2PPC) were either omitted or re-written.


Quote from: djcaetano on 2008.04.24, 21:25:17
  Hi Robert,

Quote from: RobertM on 2008.04.24, 19:59:27
As for the kernel, one of the reasons I was interested in the PPC one (which I have little to NO knowledge about - ie: I know it exists) is because it would be a good "reverse engineering" starting place to get more hints about what and how the kernel worked (and more of a possibility of something that IBM could release - if they could find it)... and though it is based off the Mach kernel, I am sure it included several key things for Intel-OS/2 similar threading and scalability...
So... yeah, I agree! Good Points!

  I really think OS/2 for PowerPC not being released (even if recompiled for Intel!) was a big shame. I fell in love with its design and really think that should be the way computing should go. But this is past and we should think about future. Anyway, I agree it would be a great starting point. The problem with this approach remains the same as using a unknown kernel: we would have the need to write all device drivers, something very cumbersome.
  BUT, being OS/2 PPC kernel based on MACH kernel (although not the same "release version as MacOS X, AFAIK), maybe it's possible to add MacOS X device driver support to it in a not-so-painful way. But at this time, this would only means we would be able to run OS/2 on MacIntel computers... something that is not bad at all :) , but it is not a perfect solution also.

   Anyway, there are lots of things to happen before this decision. If IBM decide to release OS/2 PPC Kernel source code, this could be natural way to go (even if others are possible). Let's the time show us the better way.


Robert Deed

Hey RobertM..  While we're playing the game.. I've been an IBM Development Partner since 1993 (I believe, OS/2 2.0).  I actually beta tested all versions of OS/2 since 1.3 (though I had a version of 1.2 which did not run on my PS/2.. ironic.)  I even beta tested different versions of CSet and Visualage =) 

However, my actual productivity as a development partner was rather limited.  Since most of my applications are closed source billing systems I never really took part in developer forums. =)

Though I do find it amazing that someone who came to the game later on can argue things which we actually watched happen.

RobertM

Quote from: Robert Deed on 2008.04.25, 05:57:46
Hey RobertM..  While we're playing the game.. I've been an IBM Development Partner since 1993 (I believe, OS/2 2.0).  I actually beta tested all versions of OS/2 since 1.3 (though I had a version of 1.2 which did not run on my PS/2.. ironic.)  I even beta tested different versions of CSet and Visualage =) 

However, my actual productivity as a development partner was rather limited.  Since most of my applications are closed source billing systems I never really took part in developer forums. =)

Though I do find it amazing that someone who came to the game later on can argue things which we actually watched happen.

Hey Robert,

At least he is amusing - if not redundant. I just wish he would spend even ONE post and explain why a now non-existant, defunct project and a VC group are the sudden salvation of OS/2... or for that matter what the heck it has to do with a 64 bit kernel since he hasnt provided the relationship - nor how he thinks that walking up to a VC group and saying "Hi! I am gonna port Notes, SameTime, et al to OS/2! Give me money!" would work...

Ah well... at this point I think I am going to let him enjoy his "opinions" and carry on this conversation with people who respond with "I think this because _____ (actual explanation)" :)

Anyway, I'm off for filming some shots for Star Trek New Voyages... I will see you all in 4 days.

Robert


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


Robert Deed

Curious...  What is star trek new voyages... =)

Quote from: RobertM on 2008.04.25, 06:30:32
Quote from: Robert Deed on 2008.04.25, 05:57:46
Hey RobertM..  While we're playing the game.. I've been an IBM Development Partner since 1993 (I believe, OS/2 2.0).  I actually beta tested all versions of OS/2 since 1.3 (though I had a version of 1.2 which did not run on my PS/2.. ironic.)  I even beta tested different versions of CSet and Visualage =) 

However, my actual productivity as a development partner was rather limited.  Since most of my applications are closed source billing systems I never really took part in developer forums. =)

Though I do find it amazing that someone who came to the game later on can argue things which we actually watched happen.

Hey Robert,

At least he is amusing - if not redundant. I just wish he would spend even ONE post and explain why a now non-existant, defunct project and a VC group are the sudden salvation of OS/2... or for that matter what the heck it has to do with a 64 bit kernel since he hasnt provided the relationship - nor how he thinks that walking up to a VC group and saying "Hi! I am gonna port Notes, SameTime, et al to OS/2! Give me money!" would work...

Ah well... at this point I think I am going to let him enjoy his "opinions" and carry on this conversation with people who respond with "I think this because _____ (actual explanation)" :)

Anyway, I'm off for filming some shots for Star Trek New Voyages... I will see you all in 4 days.

Robert

RobertM

Quote from: Robert Deed on 2008.04.25, 06:39:54
Curious...  What is star trek new voyages... =)

Sorry for the off-topic post/plug... Star Trek New Voyages (or Star Trek Phase 2 as we now call it) is the continuation of Kirk's original 5 year mission with new cast reprising the roles from TOS, rebuilt sets, state of the art cgi and scripts from some of TOS's most famous writers (like David Gerrold and DC Fontana), and even some TOS actors reprising their roles (such as George Takei and Grace Lee Whitney in "World Enough and Time" and Walter Koenig in "To Serve All My Days"). So far we've won the TV Guide Best Online Sci-Fi Webisode of 2007 (beat "Battlestar Galactica" and "The 4400" and the rest), been nominated for a Hugo, Peabody and Nebula award...

Check out http://www.startreknewvoyages.com/ - the episodes are free to download or watch streaming (in high definition) online.

Robert

Well, there's my one horrendously off-topic post for the month... good thing the new month starts soon!  ;D


|
|
Kirk's 5 Year Mission Continues at:
Star Trek New Voyages
|
|


saborion2

#59
Game! Games! Games!; and,  Hilariously Entertaining... Don't think the Gotham City OS/2 Meetings in those bygone days of the 90's were this entertaining. 8)

Where in the world is "Kim Chung" (Serenity Systems).  ??? ::) :D

QuoteThough I do find it amazing that someone who came to the game later on can argue things which we actually watched happen.

How about trying this in a "computerized" game for a start:  ;D
Quote

STAR DATE...... 1982


ORION RESOURCES INTERNATIONAL




A SIMPLIFIED GUIDE FOR ESTIMATING




MODERATOR NOTE: Article snipped to avoid possible copyright issues. Please post a link to such info if it is publised, instead of posting overly large snippets or entire documents.





BTW, As you enjoy your "game" of filling up at the "gas" stations in the coming days, months and years.... you may find some advantage by not having forgotten to integrate UC2 during your programming schedules.

Re:

Quote
Instant Messaging and Web Conferencing
      Lotus software
   
For unified communications and collaboration (UC²) solutions

http://www-306.ibm.com/software/lotus/category/uc2/


Enjoy comprehending the economic evaluation procedure as outlined by a
Quotedefunct VC Group
;D ;D ;D