• 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

What is the future of this platform?

Started by miturbide, 2012.05.27, 19:58:07

Previous topic - Next topic

miturbide

http://www.os2world.com/content/view/21749/2/
QuotePosted by Eugene Gorbunoff - Sunday, 27 May 2012


What applications do eComStation users need?

I consider this a good exercise, it is always good to know what the community is needing. But I still consider this community first objective is to get more users and developers. Accomplishing this is hard because of the OS competition on the PC market.

We have Windows and MacOS, close source, available for charge, and Windows came as a default on all PCs (with a hide charge) and MacOS came with all Apple computer. Both had a bast community of users and developers, and are interesting markets for software developers that sell licenses.

On the other had, we have a thousand Linux distributions, it is open source, available free of charge. There is also a vast user and developer community. And they are also an interesting server software market for it. (Oracle, IBM, other have their paid software available for this platform).

Sure, that platforms had their technical ups and downs, but they share something in common:
- they have more users
- developers
- and are a interesting market.

But here with the OS/2 and eComStation community we have the following reality:
- eComStation is only available at charge.
- OS/2 is abandomware, which means it is not legal to use it if you didn't acquire a license with it was available for sale.
- eComStation and OS/2 is open source. The source code and its improvements are controlled by a single company
- We have a reduced user and developer base.

So, what are the benefits of OS/2-eComStation compared to those other Windows, MacOS, Linux.

Being pragmatic, there is no benefit. We use/support this platform because we like it. It is the thing that we consider what is more appealing to us.

But when I try to think about an strategy to grown the user/developer base of this platform the following doubts appears:
- What is the future of this platform? Controlled by a single company and without it major developer (IBM) investing on it.
No answer.
- How can new users acquire OS/2-ecs to start testing it and developing over it?
The only solution is to pay a license fee.

So, trying to think on a solution about this the only way to start getting more users and developers is working on turning/cloning OS/2-eCS as open source software.
Making a clone of OS/2, just like other projects are working right now (Linux, ReactOS and Haiku OS).

Making it open source, will not give us user/developers at once, but will be provide a more decent answer to my questions:
- What is the future of this platform?
It is not controlled by a single company. Everybody can use it, develop it and share it for any purpose. The control does not depend on a single company or group.
The future of the platform is open for everybody to mark its path.
- How can new users acquire OS/2-ecs to start testing it and developing over it?
You can download it free of charge, and you will not be breaking any law because it is open source.

Sure, people may read this and try technically by refuting line by line, like saying is it "impossible" to make an OS/2 clone (which is not true, it is hard, not impossible). Or by complaining that ReacOS and Haiku are not technical apt.

But the main question remains when a new user/developer thinks about OS/2 and eComStation:

- What is the future of this platform?

Can you answer that?
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

crimso

#1
IMO it is rather pointless to discuss this.

The probability of OS/2 and eCS becoming Open Source is next to zero. IBM would have to invest money to sort out license issues and prepare the release. There is little to be gained there on the non-pecuniary side, so unless there is a business case of sufficient size, it will not happen.

Creating an OS/2 clone might not be impossible technically, but practically the situation is such that it won't happen anyway. Such a project would be huge and take an awful lot of manpower. The number of developers who are skilled enough to pull this off and are still into OS/2 and eCS is small. If they worked on such a project in their spare-time, it would take about forever. Even if they could work on it full-time, it would take quite long. While they were working on it, other projects that are vital to the community would have to be put on the back-burner and technical progress would move on. I rather doubt that this would be of benefit.

The only way I can see for OS/2 technology to survive in the long run is by


  • Coming up with something that is based on other OSS to minimize down the workload and is cool enough to draw developers from elsewhere, which is pretty much what Voyager was aiming to do first; or
  • Injecting it into a current platform that has a large developer and user base.

Given that user-interfaces of OSes/distributions like Windows 8 and Ubuntu are taking the mobile route currently, which seems to put off quite a few people, there might actually be a window for the second option.

danielnez1

In a perfect world we could do something along the same lines to what IBM did with OS/2 PPC; i.e. put a "OS/2 personality" on top of something else. The issue is how would we offer backward capability. Is the OS/2 API well documented and would there be legal issues if someone produced a open implementation of it?

In a perfect world we could maybe take DOSBox or FreeDOS for the DOS computability and use the newer builds of WINE for Windows computability and forgo the use of Win-OS/2.

TBH I strongly believe that Windows 8 is a train wreck in the making and I agree with crimso's view that it could be used to our advantage.

Of course, it's nice to knock ideas around but actually being able to do them is quite another. On the another hand, it has amazed me what the community has been able to do with OS/2, even without access to the source code!

Cheers,

Dan

miturbide

Quote from: crimso on 2012.05.28, 15:22:28
The probability of OS/2 and eCS becoming Open Source is next to zero. IBM would have to invest money ....
We all know that IBM is a dead end about this.

Quote from: crimso on 2012.05.28, 15:22:28
Creating an OS/2 clone might not be impossible technically, but practically the situation is such that it won't happen anyway. Such a project would be huge and take an awful lot of manpower....

I'm sorry, but I can not ignore how ReactOS had progress on cloning Windows. Sure, ReactOS is far from perfect, but they have a booting OS that can runs applications. I'm not sure how many manpower they used, but I can see the results.

I aim to this because this platform need to have a long term goal and try to make steps to reach that goal.


Quote from: crimso on 2012.05.28, 15:22:28
  • Coming up with something that is based on other OSS to minimize down the workload and is cool enough to draw developers from elsewhere, which is pretty much what Voyager was aiming to do first; or

This idea was also good and valid too. It can also be a part of a long term goal. It is too bad that Netlabs stopped the idea and there are no resources or people interested on coding it (just as there is no interest on an OSS clone)


The issue that concerns me is that there is no ideas/efforts to have a long term strategy for the platform. Patching old IBM binaries does not sound like a good long term strategy.

I just wish there could be a way to reactivate OSS projects to replace old OS/2 proprietary components.
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

aschn

With the current amount of developers, I see the following as the only
viable option (the idea is not new):

Make those features that we like about OS/2 available for Linux.
Create a window manager that feels like the WPS with objects that support
setup strings.

I don't see anything else what's missing in a current Linux distro.

Andreas

lewhoo

"I don't see anything else what's missing in a current Linux distro."

There are plenty of things missing or just made in a different way. As an everyday linux user (I have to use it at work, for scientific tasks), I find its archaic in many ways architecture very annoying. Complete rubbish in the file system, slow gui response (Xwindow), ideas that make using linux via ssh very convinient, but slowing down the system boot/response time etc... However, yes, I agree that this may be the only way to save something from OS/2. But for me OS/2 is not only the WPS, so it would be saving only a small part.

miturbide

I'm also not a Linux lover (even that I'm an open source lover), I personally dislike the Directory hierarchic structure (FHS), and the Monolithic kernel, but I agreed with Andrea's post.

It will be practical to stop worrying about the backend components and clone the top of OS/2 over it.

But on that case, which will be the steps to try to accomplish this?
Would it be like cloning, WPS, SOM and PM over Linux?

Would it be possible to start/focus on a project to clone WPS to turn it 100% OSS, and work it over eComStation/OS/2 with the goal of someday make it run on Linux?
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

melf

#7
Although there might be questionable things about the OS4 project, I think their contributions should not be disregarded. Future might tell us that the application of their work is not illegal. I guess that the kernel is the key to a future life, that is not based on Linux.
/Mikael

danielnez1

A bit of a side question, what view does Mensys have on this? Perhaps some of the revenue from eCS sales or the software subscription could potentially fund a replacement Kernel etc.

miturbide

hi melf

I also have good expectations about OS/4 project. For what I know they are using a IBM kernel leaked source code, but when they manage to replace it all there will be no legal doubts about the project (which it may take some time).

The only issue I have with the OS/4 team is that they do not care about which license to put on the development the made (commercial or open source). They are now in a legal void, they say come and code, but the rules of the use of the code are not clear. They just say, use it, it is free of charge.

Some of the russian team of this project, think that the GNU GPL and other open source license only allows big corporations to make million of dollars, which I think it is not right and I found exaggerated. They don't think that if they release it under a copyleft license, anybody that code any improvement (big corporations and individuals) will be forced to make it open source too, which in the long term it will help the project to grow. That is what happened to Linux, IBM, Redhat and even MS, when they try to make an improvement to linux components are forced to give the source code under the same license.

On this case, I also dislike releasing open source software under non-copyleft license, which mean that the derivative work can became close source. If we are trying to make the platform to grow I consider copyleft the best option.
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

miturbide

Hi danielnez1

I can not talk on Mensys behalf but I think we all know what is their current business model.

They sell eComStation licenses, and they pay the OEM fee to IBM. On that case their strategy is to make eComStation run on newer hardware to support their legacy customers.
Mensys has supported open source development like Qt4, OpenJDK, OS2AHCI, network drivers (which is good) but it is focused on trying to keep eCS running in the short term.

The reality for Mensys is that someday the OS/2 legacy corporate customers will eventually migrate to other platform, ATMs will get old and out of circulation, scada devices get upgraded, industrial machines will get upgraded once the economy gets better, so they will not require OS/2 anymore. At that moment (I think we can not specify when) it will not be a profitable business for them.

Since they sell licenses, they do not understand the open source model (services). But I can say for sure, if OS/2 became open source (cloning it) they will benefit in the long term, they currently have corporate customers which they can keep selling support, they have a developer network and they have business advantaged over any other company trying to make money with the platform.

The only one that will not benefit from making a OS/2 OSS clone is IBM, which will loose the OEM software fee.

Please notice I don't have anything against Mensys, thanks to them OS/2 is still on the market (since I don't know what happened to Serenity), and they keep investing on this platform.
I consider Mensys business model for eComStation not bad, but it is not sustainable on the long term if you want the platform to be preserved.

But Mensys is always welcome to slap me back in the face. What is the future of the platform? How are the going to grow the users base with this business model?
Martín Itúrbide
OS2World.com NewsMaster
Open Source Advocate

Skype - martiniturbide
Google Talk - martiniturbide@gmail.com

danielnez1

Hi miturbide,

I agree with what you are saying. I do believe that long term, Mensys would benefit from a more open version of OS/2 as they could still sell support or sell a beefed up version for cooperate customers etc.

As for IBM, I've always been puzzled by their attitude to OS/2 migration. For a business solutions company it seems a bit weak to suggest to their corporate OS/2 customers to simply migrate to Linux and/or use Java without providing any sort of backwards compatibility. However I guess that worked to the advantage for eCS.

DougB

QuoteAs for IBM, I've always been puzzled by their attitude to OS/2 migration.

I worked for IBM for 25 years. For the first 20, they were a company that went out of their way to be sure that the customer got the best service, and products. In the last 5 years (ending in 1992, when I retired), the whole thing changed to "make money, or you're outa here". OS/2 had a solid customer base, but IBM was not making much money from it. Development was expensive, and there was not a lot of service to be done. I believe that it paid for itself, but there was little profit in it, so IBM tried to cut it. The only reason it hung around for as long as it did, was because of the solid customer base who refused to let it go. IBM filled the gap by pushing JAVA, for a while, but that also didn't produce much profit, so they jumped on the *NIX bandwagon, and attempted to make money by selling "conversion support". I think that is coming to an end too, but I have no idea where they plan to go next. I would suspect that windows would make more opportunities for "conversion", and maintenance, and I have seen a few indications that that is where they are going. I also believe that most customers are realizing that they are not getting the service, or products, that they need (from anybody, not just IBM). There is an opportunity to pick up a lot of that business, but I don't know of anything that is available to do the job. Cell phones, and *pad products will eventually replace the laptop, and desktop (if they haven't already), but the servers that do the background work are available to whoever can make the most reliable hardware and software. Linux, seems to have that market, but I am not so sure that it is really doing the job that is desired. OS/2 (as eCS), can still do a credible job, but it is getting very limited with new hardware, the GUI parts (OS/2's strong point) are becoming less important, and the networking is getting so far out of date (no IPv6 support), that everybody can see the end of the line, and they don't want to commit to it, until that problem (among others) has been resolved.

Where does it go from here? Well, rumor has it that IBM has "lost the source". If that is actually true, then IBM would be the last company that I would hire to help me protect my data. If it is not, how can anybody trust anything that they say?  So, IBM will be no help with the task. We can get little information about new products (even simple things like video adapters, and NICs), to be able to write proper drivers. There are far too few qualified programmers working with OS/2 software to be able to keep up with the changes that are demanded by users. The bottom line is, that Mensys (eCS) has made great progress in keeping OS/2 working, but there have been no great advances since IBM backed out. Eventually, OS/2 (as eCS), will just not be able to do the job any more, and the users will have to move on, to get the job done. A lot of that has already happened, but there is a hard core who are still committed to keeping the platform alive. More power to them, but they need as much help as they can get.

rwklein

While not is out in the public. One thing that has improved and continues research is the ACPI section.
But also research in how to boot eCS from an GPT partition is progressing.

I don't think IBM has lost the sources. I know of one customer Mensys did not get that had a support contract with IBM then ran upto 2013 or 2014. So the must have source code somewhere.

Roderick

jdeb

#14
The way I understand it is if IBM wanted to release the OS/2 source code it would be unlawful for them to do so. My understanding is that they are not the sole owners of every part OS/2. OS/2 started as a joint project between IBM and Microsoft; because OS/2 contains code that belongs to both parties, IBM would have to convince Microsoft to also agree to the release. Microsoft would never do this.

With that being said, I am very happy with eCS and grateful for the advances it has made. I am also hopeful that future releases will run on more modern hardware; specifically wireless, video cards, and stable multicore use.
Foxconn P41A-P, Intel E7500, 2 GB DDR2 800, ATI X300, ADATA Sata II 64 GB SSD, 160GB WD Sata II HD, Intel Pro 100/1000, Antec Basiq 350W PS, Antec 300 Case, ECS 2.1