Consequences of the AmigaOS 3.1 source code "leak", one year after?

[Pat the Cat]

Speculation: I suspect that one of the engineers took a last snapshot of everything before he left Commodore for a related job, e.g. at NewTek, Scala or 3DO. The time stamp of the last file modified is very, very close to the day when Commodore went bancrupt.

Indeed, but I did not look,  I used logic. The last known complete source distro of 3(.1) was CBM. Therefore, the leaker was once a CBM employee.

Thank you for helping to confirm my suspicions. I'm pretty sure I know who now.

The following may offend atheists, so look away now...


... God blesss 'em,  and all their leaks.

[Pat the Cat]

Well, instead they decided to include Arexx. Unlicensed, BTW, as far as I know.

Interesting. I never got feedback from IBM on Rexx. William (Bill?)l Hawes should know, one wasy or the other.

[Pat the Cat]

At least a much cheaper. 1$ per disk as opposed to 0$ per disk is an easy decision to take. However, unlike Basic, ARexx is not a beginnner's language, and it did not include an editor, or a GUI, or support for graphics... nothing like that. It's an entirely different type of product, thus hardly comparable.

This is why they charged for replacement floppy sets, of course. Most people were like "it's a disk, it costs pennies". They did not appreciate that the content was actually valuable and had cost money to develop.

Would it have really been sensible to keep funding a rival OS, purely to keep using the same Basic? I think CBM were smart. They knew alternative Basics would appear, having studied how software chains develop. They also knew that, for serious development of applications, Basic was very very limited.

You have a bleeding edge product. You want to attract developers. You do not include crap development tools with the base machine unless there is no alternative.

[Pat the Cat]

Thus: ARexx != Rexx, except for the language fundamentals, and Bill != IBM.

Rexx = IBM. Arexx = Amiga implementation of Rexx. Therefore, to certain degree, Bill has a relationship with an IBM product, in that he developed an Amiga implementation of an IBM product. He didn't do it blindfolded.

[Pat the Cat]

If you search through the CBM bug database (which you probably don't have), you'll notice that there were at least three versions of AmigaBasic (all from Microsoft, of course). The version 1.2, which was shipped on the Extras disk of Workbench 1.2.

I didn't know they had a bug database. I thought they just had a shredder for that sort of thing. CBM were principally known as a hardware provider. Any issues with software were kept very, very close. Bit like Apple in many respects.

If you weren't a registered developer (and I could not be) then they didn't talk to you. They certainly did not talk to me about jack %&$#?@!%&$#?@!%&$#?@!%&$#?@! that was technical. I had to find everything out from day one pretty much on my own, but at least I could make phone calls without worrying about the cost of that.

Reason I couldn't be a registered developer? Conflict of interests. You cannot sign a Non Disclosure Agreement and yet be a responsible, independent journalist who puts their readers first, last, and always. That was my problem to deal with, and I ended up giving up journalism. I made some cool Amiga toys instead. Now excuse me, I'll start digging them out for y'all.

[Pat the Cat]

CBM had a professional developer program, called CATS. Unfortunately, it costed money to join, and no, I never joined. I just bought the books (RKRMs, AmigaDos manual) and did some reasearch of my own.

Well, at least I called CATS. They kind of had different levels, for different pockets. I never registered either though. Some journalists did register, but you had to go out of the office, couldn't be staff, for that. I saw that as unprofessional and so did the magazine management generally. If you are keeping commercial secrets that you cannot disclose, then you can't have true freedom of expression.

Right, and correct. I also wrote letters to CBM, they probably went into the same paper waste as yours.

I never wrote to CATS. Never wrote to Commodore. Never had time, I had thousands of other words to write, constantly.

I believe everyone disliked the CBM "management" in all its glory. Customer relations costs money, and CBM was always short of money. Instead, ask developers to *pay* for your platform. I as a hobby developer did not buy into this type of marketing.

Well, they DID pay on Marketing. They paid Future to have Amiga Format "Get you Started" guides in the boxes. They were a sodding nightmare, in some respects. And I know they paid for that, because that was an out of house commission. But it did ease the journey for noobs. They could start having fun quicker.

Then CBM released the A500+ without even telling us. Gee, thanks guys. Really clever. "Oh, THAT new machine... Yes, we're not doing upgrades for WB2 for earlier Amigas for a couple years yet". In other words, they were going to keep control of the Dev and Release chain, and still not tell anybody how to use the bloody thing properly. Maybe they were too scared to talk at all. I read Dave Hs pieces on the period, and his releases. The prototype AA3000s were going up in smoke and falling apart, and nobody knew why exactly. Lot of conspiracy theories about that one. I think it was just a failed board run on a new production system, but there you go. It might have been intentional sabotage. Who knows?

[Pat the Cat]

s.

The dos.library in AmigaOS4 assumes more responsibilities than was the case in previous versions, in which file systems and client software was left to fend for themselves. The old way of doing things required a file system to implement an inane/insane number of operations, accessing crufty data structures according to very underdocumented rules, and this in turn led to stability issues. It was hard to write a robust file system, but poisonously easy to get it subtly wrong with dire consequences, for which it was extremely difficult to figure out what exactly caused them.

So, for example, dos.library now cares about how file change notifications work, how you would change the volume or device list after a medium change. This sounds like the most pedestrian thing to mention, but what goes on under the hood in dos.library V40 (and below) in these areas is a highly complex and error-prone process.

All very true. A classic OS will either work or lock everything up in this respect, and typically non developer systems don't have a hardware "quit" on a program to interrupt the softcrash. Mind you, if everything is talking to each properly, it doesn't happen. But out of control software, unstable software, is an issue with the Classic AmigaOS.

It's an issue with all OS, mind you. I hear you are a very nice and helpful Amiga guy who does nice things with networks. You've probably done scads more Amiga, like Thomas. Thank you.

[Pat the Cat]

A friend of mine wrote Regina, the free implementation of REXX, he too, has no relations to IBM. You see, REXX is a _language_, anyone can implement interpreter for it.

Sure. And if you do that for the WRONG operating system, and release it for sale, IBM will sue your arse off.

Just because Rexx has public versions, doesn't make all of Rexx open source and free to develop for, on any platform.

Try making and selling a product that uses HPGL. Hewlett Packard will have a few things to say, like "cease and desist or pay us lots of money". You can develop such a product, use it, give it away, no problems. Try selling it, different story. Not a smart thing to do.

[Pat the Cat]

No, you are wrong.

I'll give you a clue. Try it on a mainframe...

... you ain't got one. Ain't that a shame...

http://regina-rexx.sourceforge.net[/URL]

Yes, Regina is open source. Rexx isn't. If was, It would be somewhat easier to google the source code for it an IBM mainframe.

[Pat the Cat]

You have no idea what my job is and what I have access to.

Yeah, sure, and you take the mainframe home with you in your backpocket at the weekend. I don't care whose mainframe it is, I don't care if the it's the NSA backbone machine, it's not yours.

What - are you googling?

Again - just like with C, Rexx is the language, for which there are multiple implementations, some commercial, some not. IBM have their implementations, but they do not stand in the way of others doing theirs. It is quite common for them all to name the interpreter available in the command path "rexx". IBM has a long history with Rexx, since the language was born at their turf (by the hands of Mike Cowlishaw), and IBM has used it through-out their products, on mainframes as you say, and on OS/2 and elsewhere. They also have worked closely with and contributed to the Rexx Language Association, with funding of development as well as source code of their own development.

http://www-01.ibm.com/software/awdtools/rexx/opensource.html

Yes, and hell will freeze over before they release their mainframe source. Or rather, they will have a highly evolved replacement before they release it as open source. In the meantime, please stop repeating the same assumption, that because there is a "free and open" version, all versions are "free and open".

[Pat the Cat]

True? False?

Completely irrelevent to the Amiga. If you want to write Rexx for the Amiga, I have to point out, it's been done.

If you want to write Rexx for something else, that is also irrelevent to the Amiga.

If you just want to hold up Rexx as an example of something completely and utterly Open Source, I suggest you phone IBM in the morning and ask them how they feel about it. It's theirs, their call. Not mine. As you are a tech working on an IBM mainframe, this should not be a problem for you.

Quite frankly I'd have better things to do with my time, but if you really want to prove a point, I won't stop you. It might be CLM, a Career Limiting Move, professional suicide for you maybe, but that's not my concern. Or you might be correct, IBM mainframe Rexx is completely open source. Amazing. That doesn't mean Arexx is open source though. Oh dear. How sad. Nevermind.

[Pat the Cat]

Hyperion Entertainment holds an  exclusive, world-wide, perpetual source-code license to the intellectual  property of AmigaOS 3.1 and additional content as documented in the  publicly available settlement agreement between Hyperion Entertainment  and Amiga, Inc. which has taken the form of a stipulated judgement.
~ http://hyperion-entertainment.biz/index.php/news/38-corporate/167-amigaos-31-source-code-leak-official-statement

Untested in court. It was settled out of court, the court just rubber stamped it... and Amiga Inc never showed a direct path between CBM and themselves, in regards to the copyright on the source code.

I'm not saying there isn't such a direct path of transfer, but I think anybody trying to demonstrate proof of ownership to the copyrighted sourrce code is going to have a hard time of it - they'd have to show an agreement with the CBM liquidators, who were based in the Bahamas, as a starting point, which would include the term. If it wasn't included in the break up of CBM, it was never transferred from CBM.

Without that, nobody has a claim to it, but it's still copyrighted (to CBM).

[Pat the Cat]

It is not at all legally safe for anyone who worked on any original AmigaOS to work with AROS.

It's totally safe. They just can't receive money for working on AROS. That would be unsafe.

You cannot legally stop somebody from doing voluntary work without breaching their civil rights ("freedom of expression"), and any contracts signed with CBM are long, long void.

A right cannot be forfeited. A privalege can be forfeited, but NEVER a right. They are non-removable. 1689, Bill of Rights. This goes back to even before America was born.

[Pat the Cat]

Should I start with the highly inconsistent (and up to recently mostly undefined) syntax? Or the interoperation between the Shell and script execution ("Execute" and all its wierdos)? Or the amounts of black magic in the Shell start-up? Which are, for good reasons, undocumented. It spares the reader a real headache.

I know of maybe 10% of what you refer to. Personally I would miss such things, being one of those "Execute weirdo" types, but then again you are kind of in a no-win situation with that stuff. You can't please everybody even if you did nothing. You can't please everybody with what you release.

So at least you are doing something positive by trying something new that works better, even if some people miss the old black magic. They can still have it, but they have the old limitations too, and miss out on new features added.

The biggest kind of "black magic" I can think of was the official CBM line on the phrase "compatibility". It is the most nonsensical gibberish I have ever read in the English language, barring perhaps a Windows or MS-DOS EULA. :furious:

Do your best, Herr Richter, anybody and everybody involved in Amiga development work. No one can expect you to do more. They might moan and complain, but the real honour lies with the person who takes up challenges and meets them, as best they can.