Quote:

That trend has actually started with Zune...

Quote:

Yes it has... but open source and closed source collide somehow. And they propably were not interested to become dependent on a closed project with a core component everything (including desktop) is based on.

Quote:

I'm not a developer, I don't have the talent.
Yes, I already am familiar with Fab, he helped me work out a problem I was having accessing my email via OWB, great guy.
And its already been mentioned that MorphOS has not utilized AROS sources for over ten years, so what impact would Deadwood's work have?
BTW - I'm not hostile to AROS (or its developers) in any way.
I still occasionally send messages to Staf Verhagen just to check in.

Quote:

I have been using MorphOS for only about ten years, and prior to that only had limited exposure to Amiga systems.
In fact, since I sold 68K based systems in the '80s and '90s, Amiga was the competition (and our four or five user systems started at $999, which was a lot less than what an Amiga commanded).

For that matter, I keep track of the development of AROS, and occasionally install it to evaluate its status.

So, while new users are not common, they exist.

@Jim

Deadwood has impact because he is the guy behind newest version of OWB today. He works on it over a year now and has also fixed bugs like the memory bug and is giving away the changes.

What new users? There is hardly anyone showing up. And to be honest people who not know of amiga will not join anyway and that former amigans join again is not very propable too because of lack of competitive software. That is true for all platforms.

[ Editiert durch OlafSch 13.09.2016 - 16:08 ]

> As to porting anything else, why would anyone feel compelled to do that?

Because the porter wants the software on his preferred platform, for instance :-)

> if members of our development team believe that part of the MUI4 repository that was
> used contained work from developers other than Stuntz, then it seems quite likely.

Even then it's still a difference whether the source code was obtained legally (from Stuntz, but with non-Stuntz code, so Maus/Böckelmann have been acting in good faith) or illegally.

> Hope you enjoy fixes by Deadwood from Aros camp.

As written in comment #252, there's nothing to enjoy yet for MorphOS users unfortunately, as Fab has not yet released a new version with integrated fixes.

Oh yeah, that deadwood.
OK, whenever that happens.
Personally, I'd like a Java JIT port.

> its already been mentioned that MorphOS has not utilized AROS sources for over ten years

I'm not sure if "not synced anything with the AROS tree" is exactly the same as "not utilized AROS sources".

> what impact would Deadwood's work have?

http://morph.zone/modules/news/article_storyid_2262.html
http://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=11529&forum=9

Quote:

Quick, before Andreas beats me to it:

You mean Javascript JIT?

Quote:

To clarify: I'm not aware of any of the initially imported AROS APL modules being updated with changes from the AROS tree in the last several years (if ever).

> I'd like a Java JIT port.

In case you refer to JavaScript here, I'm afraid such JIT can't be ported between ISAs. A JIT compiler which compiles JavaScript code to PPC machine code must be written in PPC ASM, and unfortunately we can't use the TenFourFox one because WebKit is dependant on JavaScriptCore. That's why bigfoot has been working on such JIT for the last 3 years.

http://morph.zone/modules/news/article_storyid_2075.html
http://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=11523&forum=32&start=4

Quote:

Just what Mark needs, another project.
Still, can't think of anyone better to pursue this.

Quote:

No no, no such thing as "good faith" is applicable here. Good faith does absolutely not make it *legal* to use other people's copyrighted materials, not in any way. Cluelessness (deliberate or not) does not put you above copyright laws. If laire, piru, kiero or jacadcaps (and probably others) should decide to make an official copyright infringement thing about this, they would probably have a pretty solid case (from change logs alone), since using their work without their explicit permission is simply illegal. How sources were *obtained* is perhaps of lesser importance, unlike the usage of them. They obviously were granted access to the source tree. Obviously. To make a MUI 3.9 port, that's for sure. In the source tree they also found the MUI4 sources. That is also obvious. Maybe they thought something in line with "hey, we've got permission to make a port of MUI period (no matter version), so let's download the MUI4 sources as well". But even the MUI4 code in 2009 contained work of others. Had I been managing a big project like MUI, I would have kept track on changes and contributions, and I see no reasons to why that wouldn't have been the case. Under any circumstances it falls onto *their* responsibility to ensure 100% IP integrity in what they release, when using external codebase. Their rather "liberal views" on things like copyright notices when it comes to MUI related files is publically proven, which could have "helped". Delete. Replace. Modus operandi. Et cetera. But this was not a 2009 affair. It was a 2015 affair. And a 2016 affair. And still counting. Had they had a valid *license* to MUI4+ (4.x+ in practice meaning "MUI NG", meaning beyond the "old world" 3.x MUI (there were major inner reworkings put into 4+)), they wouldn't have had to resort to "filling the gaps" with their own made up code, they could simply have based their port on the official sources they should have approved access to. The same with MUI5, had they really had a valid *license* to MUI5, there would have been a proper port from the official sources. But this was not the case. They created a new branch of MUI development in 2015 that was *not* the 2015 MUI, but something different, incomplete and incompatible. This while posing as the new MUI for AmigaOS developers, starting a new "official" site. Now in 2016 they released MUI 5, and with that it became even more "comical" since the major version bump to "5" was never at all about whatever classes or ui api they may have mimicked, but even more about inner workings and lower level improvements and merging with the OS. Perhaps a real MUI 5 is no longer even easily portable to systems like OS4, where MUI is something that is optionally run as a stand-alone, third party application. But paint a beef onto a cardboard sheet, and maybe it will go down just as fine together with som red wine sauce and brussel sprouts...?

Quote:

You sure you don't mean before Andreas beats me over the head with it?

[ Edited by Jim 13.09.2016 - 18:58 ]

> You sure you don't mean before Andreas beats me over the head with it?

Is that how comment #312 reads to you?

No, don't get too thin skinned on me.
But you do like to hop on any error.
However, the additional information, as always, is quite useful.
Over the years, your attention to detail and your persistant research has been invaluable.
You already know that when I need CORRECT information, you're the person I trust most.

> No no, no such thing as "good faith" is applicable here. Good faith
> does absolutely not make it *legal* to use other people's copyrighted
> materials, not in any way.

As you're not exactly known as legal expert, please have a read:

https://en.wikipedia.org/wiki/Good_faith_(law)

Both Stuntz and Maus/Böckelmann are German, so this applies here:

https://de.wikipedia.org/wiki/Treu_und_Glauben

> If laire, piru, kiero or jacadcaps (and probably others) should decide
> to make an official copyright infringement thing about this, they would
> probably have a pretty solid case

Yes, against Stuntz that would be, in case Maus/Böckelmann can prove he licensed their code to them. Of course, as soon as it was proven that Stuntz illegally licensed other people's IP without their consent, Maus/Böckelmann would have to stop distribution of their MUI version.

> using their work without their explicit permission is simply illegal.

...as is licensing their work to others without their explicit permission.

> How sources were *obtained* is perhaps of lesser importance, unlike the usage of them.

"Perhaps" is the key word here.

> In the source tree they also found the MUI4 sources. That is also obvious.

No, that's just one version of the story. The accused tell a different story.

> Under any circumstances it falls onto *their* responsibility to ensure 100% IP
> integrity in what they release, when using external codebase.

Please do as suggested and read about "good faith" or better "Treu und Glauben".

> Had they had a valid *license* to MUI4+ [...], they wouldn't have had
> to resort to "filling the gaps" with their own made up code, they could
> simply have based their port on the official sources they should have
> approved access to.

As you know, the accused claim they got from Stuntz a license to 2009 MUI4 source code, so it's obvious they had to "fill the gaps" with their "own made up code" in their attempt to match anything that was added after that to the MorphOS MUI4/5 source code.

> The same with MUI5, had they really had a valid *license* to MUI5,
> there would have been a proper port from the official sources.

As you know, the accused never claimed they had a license to MorphOS MUI5 (i.e. 2015+) source code.

> They created a new branch of MUI development in 2015

2013.

> you do like to hop on any error.

At least *I* didn't hop on your Java(Script) confusion :-)

> the additional information, as always, is quite useful.

My reply wasn't additional information but (in direct reference to what you wrote) the very explanation as to why such JIT can't be ported from another ISA.

Quote:

You and others make it sound like MUI4 or MUI5 is something completely different than MUI 3.x. It is not. It is still *very* largely the same stupid (sorry) thing. If the version jump was from 3.1 to 3.2 it would look much less dramatic than 3.9 to 4.x. For comparison the AOS Workbench (not ROM) version jump from AOS 3.1 to 3.5 contained hugely more dramatic changes then the jump from MUI 3.8 to MUI5. In absolute they may not have been all that huge but compared to MUI 3.x to MUI 5 they are immensly huge. No comparison whatsoever.

Quote:

I would have assumed that by default contributions made by others to my MUI sources end up under my control and if the contributors want special treatment (want to control who else gets to uses the stuff they commited) *they* have to ask me for permission to allow them to contribute stuff under this special conditions. Not the other way round. That to me feels totally crazy. It's like if MOS Team allowed someone to port MOS to ARM (if they are not interested to do it themselves) and later that guy wants permission from MOS Team to use the modifications he did (like modifications to original code to make everything endianess safe) for an x86 MOS port.

Quote:

Polluks has been doing quite a bit of using AND fixing some stuff from AROS lately. Do I know exactly what he's doing with it. Nope don't claim to. :)

Quote:

Sure, it's simply no longer based on the same object ownership paradigms the older versions were. Which in turn will enable even cooler things yet to come.

On a side note, I wonder if the fork version '5' can do this too...

Quote:

If dev docs about it are as good as for many other MUI4/MUI5 features (not existing or saying "yet undocumented. complain in mailing list") you may be safe for a while ...

There probably are not many (any?) apps using MUI4/MUI5 features written by outside coders who do not have access to MUI4/MUI5 source code (which those who have can use as docs) or close contact to those who have. Or are there?

Quote:

That was the case there, yes. MUI used to be developed on two repositories back then and whatever made it to stuntzi's repository was considered under his control. There's an issue of not crediting all the authors properly though, but this is something he should have addressed when handing over the sources.

(sorry for the delay, it took me a while to dig out old stuff and verify things)

Quote:

Actually this was documented in the intuition/BOOPSI headers a while ago...