I consider that an OpenJDK port will be more usefull than a Odinized (sorry for making Odin a verb here) version of JRE for Windows.
Why?
First, because OpenJDK is open source (under GNU GPL V2 license), that means that the code will be shared and various developers can chip in and no one will have the exclusive control of an OpenJDK build for eCS-OS2.
Ex: If the main developer of OpenJDK dies then other can continue with the job. No company will have exclusive control on telling us how to use OpenJDK or try to restrict us to use it.
While making the Windows JRE to work on OS/2 means that, possible one day Oracle/Sun can change the license to forbide this, or stop giving the Windows JRE for free. This can not happen with OpenJDK (license GNU GPL). But I understand that an Odinized JRE will be a good workarround to have something fast, but it will not be a good solution thinking in the long term.
The second thing is that I prefer having a native port of the JRE. Possible Odin can work fine, but it gaves a better performance to have a native port.
The third point is a doubt that I have. Is the Innotek pack for making Odinized applications complete open source? I haven't seen any documentation on how to make Odinized OS/2 applications arround, like for exmaple how the Odinized Acrobat was produced. I think that Netlabs made some agreement with Innotek to get this technology before the Sun deal, but I'm just especulating, since I don't know how that ended. If anybody can give me some lights on this point it will be apreciated.
Roderick, about the Bounties, I share with you that is not being effective. But I think that if you share a problem with the community, it is a good idea to share also a possible solution so we can evaluate them. Or you just want us to close the bounty system definitive ? or if you have a better idea how to raise more money and not only a hundred of dollars, please let us know.
Possible to make a positive change on the Bounty system it is required to contact more developers and ask them how much money sounds good enought for them to be involved on making any of the Software ports proposed. And later anounce a target for the Bounty and let people know there is a developer interested if we reach $XX target.
So the Question for this Bounty is:
1) Which Developers are interested on making the OpenJDK port for eCS-OS2?
2) How much money is required to raise to make this port a reality? And later contact all the people in the community interested to Chip in for it.
So Roderick, the questions to Mensys are:
1) Is Mensys interested on this Bounty to port OpenJDK to eCS-OS2 ? Allowing access to the Binaries and Source Code complete ?
2) How much money do Mensys required to make the port possible ?
If there is interest we will see if we can raise enought money for this. And also thinking in the long term that eCS will have a native, open source, and free Java Runtime Enviroment to be included on it and can add more value to eCS.
Regards.
Martin.
It has been a while since I checked all of the Java stuff. The good stuff of the ODIN code is that it requires less work per build. I know the advantages of getting Java compiling native, that is not an ODIN build.
About the bounty system ? You don't have to close it down. Its just that
How much money is needed to get a current Java build ? I don't know. This depends on the developer of course and the road choosen! If you take the current ODIN code we have that is being updated to support Flash then that could be a done for maybe a few thousand dollars.
I don't know how much work would be needed to get JAVA working as a native port. With Open Office a lot of the old OS/2 code was still floating around CVS. So that made live easier, it was still a big job.
Price tag ? It depends who you can find todo the work. Or if we can find a team of volunteers. But as was written in this thread by Silvan, its going to be a lot of work. I'm not holding my breath for a volunteer to do this work. I know how much time our Open Office Developer needs to keep up! Its not a small operation :-)
That is I don't want to dis-curriage anybody to try and work on such a project in his spare time. The estimate is just that its going to chunk of work. How big that chunk is don't know.
So the price tag would depend on the developer. Could he do it as labour of love and for a lower price, is he from a country where lets say 1000 Dollars a month is a lot of money. Or is somebody from the US that just needs the full price paid per hour.
Roderick Klein
Mensys