Hi
ArcaOS was released like 6 months ago (May 15, 2017) and I think it will be a good time to have some discussion about what
YOU think about ArcaOS. I had seen people jumping on other forum threads giving their opinion on what should be a priority and what not. I think it will be better to use this thread to do that.
From my personal point of view I have my wishlist based on my needs and hopes which I complete understand differs from other opinions.
Short Term - DriversPeople still scream for drivers and there are different needs, but my personal needs is in this order:
-
Wifi (
Listed as Short Term Projects on AN Roadmap)
-
USB 3.0 - In my case I need it to install ArcaOS on devices that does not have USB 2.0 ports, CDs or Legacy compatibility for USB. (
Listed as Long Term Projects on AN Roadmap)
-
GTP : To make it easy to install ArcaOS on HDD if you want dual boot. (
Listed as Long Term Projects on AN Roadmap)
-
UEFI : It is where the market is going. There are some Intel PCs that does not have Legacy mode.
-
Support for accessing unpartitioned/non-LVM USB sticks (large floppy) greater than 2GB. To make life easier also. (
Listed as Long Term Projects on AN Roadmap)
Mid Term - AppsArcaOS does not have a roadmap of which software frameworks should be ported to have Apps under the platform. The easy way is to say it is "just and OS" and I don't care what it is running on it, but if you don't have apps, why use the OS on the first place?
It is obvious that they primary goal of ArcaOS was to run OS/2 applications made with OS/2's API (CPI, PM, SOM, WPS, REXX, etc) and I think that goal is accomplished with ArcaOS 5.0. Now what about the present? If someone wants to develop something for ArcaOS what it should use?
1) IBM OS/2 Toolkit was be repackaged and included on RPM, that was good. But maybe it needs more like a facelift to include more samples, and updated documentation on how to compile the samples today on 2017, and maybe even rebrand it with a new name and version to match ArcaOS.
2) ArcaOS Open Toolkit. I think it is worthy to promote that ArcaOS has open source libraries and tool that can be used to also create software under the platform. This should also be seen as an strategic part of the OS that allows developers to create and port software. (Even if there are people that counts the microseconds comparing a PM app vs. a Qt App, those can keep using the classic OS/2 API)
3)
Qt 5 is important to have it as part of the "open toolkit". Let's hope that can happen with Bitwise.
4) More frameworks. I think that there is missing one or two more modern frameworks on the platform, but this is something I didn't set my mind on. I feel like there are very interesting things on GTK and the Node.js frameworks.
Long TermMaybe six months it is still not enough time to get a long term strategy. This is the hard spot and where everybody will reply to me "Don't fix what it is not broken".
Does the OS/2 platform should have a long term strategy? Or the strategy should be "Will the last person leaving, please turn out the lights"
When you buy ArcaOS you need to pay the "IBM tax" (the cost of the OS/2 license that IBM negotiated with Arca Noae), that is in part what makes ArcaOS expensive (Mainframe guys, please don't compare prices here). Also the soul of OS/2, which I think are CPI, PM, SOM and WPS, are close source without any possibility to be enhanced, fixed or open sourced.
Maybe like Lynn H. Maxson said on some Warpstock, we need to progressively open source parts of OS/2. I had been told that is almost sure that if someone wants to clone PM or SOM, the first version will suck very bad. They tell it to me like if there is no way to improve software over time or like version 1.0 has to be 100% bug free, which is not how the software development cycle works. Since this is a long term strategy I think it will be good to nurture the open source replacement from crappy to something that can work over time, instead of thinking it as a Big bang miss or hit thing.
Maybe:
1) Evolving XWorkplace to try to reduce the dependency on WPS classes until it became a fully parallel desktop workplace object classes running on SOM. (little by little)
2) Hacking PM and CPI to replace and improve some of the API functions (little by little).
3) Revolving SOM by trying to get the leftovers of Netlabs SOM and SOMfree to see if some useful SOM replacement can be generated from it. (little by little)
Why do I still insist on Open Source? Because we already had been dumped three times, one by IBM, second by Serenity Systems and the third by Mensys/XEU. Open source is just a way to reduce the risk in case of a fourth dump and eventually (with a lot of hard work) it will release us from the IBM tax.
So, let's have fun, post your wishlist.
Regards