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

[kolla]

Nice rant about how much distaste you have for Linux, now provide examples of how brilliant closed source products are

[kolla]

Just got a new SCSI2SD hardware here, with some linux software to install it.

Why would this need "special software" btw?

System is a pretty stable ("rotten"?) Debian system. Does the software work out of the box? Of course not! It misses "libudev.0", except that Debian runs (since ages) libudev.1, the next version, with a different interface. Why was that breakage necessary? Was it really necessary to create "just another incompatible" interface for udev?

Yes, udev is part of systemd, you just entered dragon territory.

If open source coders had some discipline in keeping their software interfaces stable, linux could be a much better system - but that is not the development goal of open source.

Wrong, it has _nothing_ to do with open source, it's just not the goal of certain popular Linux distributions.

It sounds me to that you would be much more at home running NetBSD, for example.

[Pat the Cat]

Huh, I am curious... where have you been since 1994?  It has been quite well established knowledge that the sources have been  around all the time.

Excellent. By whom? Who has claimed that? Where were the sources available from? How long have they been available for?

AFAIK this code was leaked just a year ago. Now you say it was always around. Where from? Who from?

Or was it just passed around the Dark net as a piece of exotic techno  contraband? That seems to be what you are talking about, maybe. I doubt  it, it would have surfaced a lot, lot earlier if that was the case.

You want 22 years of my personal history? Google is your friend, they say.

EDIT:
The official settlement, which was reached out of court, and rubber  stamped by the court in Washington, is here. It seems to make no  reference to the fact that AmigaDOS is itself based on a sublicense from  Metacomco... which everybody seems to have conveniently forgotten.

https://cases.justia.com/federal/district-courts/washington/wawdce/2:2007cv00631/143245/148/1.pdf?ts=1261252743

Anyway, I'll study it further. Metacomco were British. Y'all thought  everyting Amiga were totally developed in the States? Not quite true.  Dr Tim King certainly not. I never met him, but I did get to study some  of his 6502 code on a completely unrelated project, legitimately).

[guest11527]

Nice rant about how much distaste you have for Linux, now provide examples of how brilliant closed source products are

Installed SCSI2SD on Windows 7. Worked out of the box. No need to fiddle with old libraries. No, I do not use windows on a daily basis, leave alone the spyware windows 10 which we are not even allowed to use in our office due to privacy concerns . But, dislike microsoft as much as you like: Their stuff *works*, and they care about compatibility.

[ferrellsl]

You may remember that just about a year ago a file "amiga os source code 3.1.tar.bz2" popped up on a web site, was linked to, copied, and the contents even wound up on GitHub for a couple of days. This event was widely publicized, on Twitter, on personal blogs, and it even made the news.

That file would contain pretty much all the AmigaOS 3.1 source code, and plenty of other material which used to be available to Commodore developers back in 1994. It's safe to say that the contents of the archive are now very widely distributed, just not necessarily available to the general public.

Back then there was speculation as to who made the data available, where the data came from, and which consequences the availability would have.

It's been a year now, and I'm curious. What did the availability of the source code make possible?

(Careful: there could be legal strings attached to answering this question, so you might consider your options when posting answers here)

If you have to ask that question, then there have been no consequences.

[guest11527]

Why would this need "special software" btw?

To setup the parameters, for example the SCSI ID by which the device is seen.

Yes, udev is part of systemd, you just entered dragon territory.

Yes. Amongst the other "dragon territories", as such as "which init system do I use today", "which X11 replacement system do you prefer" and "how do I configure my printer with cups".

Wrong, it has _nothing_ to do with open source, it's just not the goal of certain popular Linux distributions.

One comes with the other, and that is the problem you do not (yet?) understand. It is the motivation of the developers. Of course, as an open source developer, I want my code to be nice, clean, orthogonal, and - sorry - I need to change an interface today for that because I did a bobo in first place, %&$#?@!%&$#?@!%&$#?@!%&$#?@! happens. Oh, sorry for the user, just reinstall...

As a paid developer, my motivaton is my pay-check, and this comes from the customer. If the product doesn't work, I'm fired. Full stop. Means, "bad code runs the industry", and AmigaOs is certainly that (to major parts): Bad code. However, it's bad code that works. If you leave this bad code to open source, you might get good code in the end, but no compatibility. BPTRs, BSTRs? Away with this nonsense. Legacy GlobVec initialization in dos.library? Away with this crap. graphics.library workarounds for bad programs? Sorry for them, away with the junk...

All nice and correct from a software engineering perspective, but still a bad decision from the user perspective.

[psxphill]

If you leave this bad code to open source, you might get good code in the end, but no compatibility. BPTRs, BSTRs? Away with this nonsense. Legacy GlobVec initialization in dos.library? Away with this crap. graphics.library workarounds for bad programs? Sorry for them, away with the junk...

Implementation and interfaces are two different things. So BPTR/BSTR/GlobalVev should always be part of an open source amigaos, no matter what language it ultimately ended up in. Bonus points if you can pursuade a c++ compiler to create 100% compatible libraries using a simple class and convert BPTR & BSTR to real pointers.

workrounds are always a problem as they are necessary because one thing assumes something about another thing that was never guaranteed. There are workrounds for hardware bugs, which might need changing when you make an unrelated change for example. For that you need to do decent testing, whether it's open source or not won't make a difference to that.

[Pat the Cat]

I ask again - with binaries floating around that were built using the leaked sources - why are the legal owners not doing anything?

Read the settlement. That just states Amiga Group have the right to enforce copyright.

Whether they choose to do so, or are able to do, is a different question. And whether their claim to do is legitimate technically hasn't been tested in a court (yet). That would be the "bottom line", I guess.

[slaapliedje]

Why would this need "special software" btw?

I was going to ask the same thing, I very rarely see any drivers on a CD for hardware, since almost all of it is handled and is in the kernel.  

Yes, udev is part of systemd, you just entered dragon territory.

As you pointed out, why any special software?  Copy the udev rules over and you're good to go.  Why would this depend on a specific version of libudev?  And even if it did, they actually DO use backward compatibility.

Wrong, it has _nothing_ to do with open source, it's just not the goal of certain popular Linux distributions.

It sounds me to that you would be much more at home running NetBSD, for example.

Yeah, not sure what the beef with Linux is that he has.  A while ago I bought Heavy Gear 2 for Linux.  Guess what?  Even though it came out in 2000,  it still works on my Debian Sid desktop.  That's 16 years ago, for a very much proprietary game.  This "Linux changes all the time!" is a mantra that Microsoft and the BSD guys have thrown out there to try to discredit Linux.  

Granted, as you said, systemd is a whole different beast.  It's still pretty heavy in development, but the api level stuff is all pretty solid, otherwise RedHat wouldn't have put it into RHEL7.

Back on topic; If AmigaOS went open source, even now, it'd help all of the AmigaNG projects, not to mention just improvements all around.  Would be nice to get updated kickstarts as well so we wouldn't have to use reboots to get it patched up to newer standards.

Even if no one was interested in hacking on it, it'd be good to be able to look through the code for academic reasons.  Personally I like looking at how operating systems are designed and how they're used, so from that perspective having source available is fantastic.

[eliyahu]

Back on topic; If AmigaOS went open source, even now, it'd help all of the AmigaNG projects, not to mention just improvements all around.  Would be nice to get updated kickstarts as well so we wouldn't have to use reboots to get it patched up to newer standards.

it wouldn't help hyperion since they already have the source; i don't know how it would help AROS or MOS -- and besides, if it's "available" as some posters suggest, then they already have access to it.

if you want updated kickstarts, again, check out the AROS project. open-source, compatible, and under continuous development. everything the crowd who keeps asking for OS3 to open-sourced wants. except the name and the history. and if that's what you want rather than 'open source,' upgrade to AOS4.

-- eliyahu

[slaapliedje]

I think the sole problem of the AROS kickstarter project is that it's still a reimplementation of the 3.1 API, as the rest of AROS.  

You're right about Hyperion not needing it, since they already have it.  I figure the AROS kickstart is much like EmuTOS for the Atari's.  Open source, but they're still reimplementing things that they wouldn't have to spend as much time on if TOS/GEM had gone fully open source.  EmuTOS doesn't have full compatibility with the hardware banging software out there, and neither does AROS' kickstart.  That's neither here nor there though, since AROS I believe is a recreation of 3.1, whereas we already have 3.9+BB1-BB4 that updates all of this.  

I really should get an burner and try out the AROS kickstart on my A4000D to see how well it works, I've only used it under UAE, but then again I should do the same with EmuTOS.

But (with the exception of Thomas of course ) we have lots of 'hackers' that could/would probably do something amazing with the source if it were actually opened.  

But knowing what I do about various licensing, it's almost impossible to do that.  For example, even if Linus wanted to shift Linux to GPLv3, he probably couldn't because of the clause where if you want to re-license code, you have to have the okay by the developer.  Some of the coders have surely passed on from AmigaOS, so getting things relicensed so that people could reuse it without someone bitching up a storm (whomever that someone is).

[EvilGuy]

Wrong, it has _nothing_ to do with open source, it's just not the goal of certain popular Linux distributions.

And the need some people have to "upgrade". If you've got a working system that does everything you want, why would you want to upgrade and potentially break things. Just to whinge about how bad open-source is?

This is universal, it applies to open-source and closed-source environments.

It's why people have old DOS boxes still hanging around, or Amiga's controlling heating systems, or C64s controlling machine tools..

[guest11527]

And the need some people have to "upgrade". If you've got a working system that does everything you want, why would you want to upgrade and potentially break things.

Because then somebody else will break the machine for you. Seriously, if you connect a system to the internet today, and you do not keep the system up to date, then you're soon running into trouble.

Just to whinge about how bad open-source is?

It's not about "good or bad", it's about "what is the goal of the development". Open source makes the developer happy. If you are the developer, then that's good for you. If you just need a system to be productive in some other way, it's not good for you.

This is universal, it applies to open-source and closed-source environments.

Actually, the way how the problem is handled is different. M$ spends great effort to write software wrappers around software wrappers to keep legacy software working. This is, from an engineering perspective, not very satisfactory, it makes the system slower than necessary, and more bulky than necessary. For MS, it is important that the system is sold. This type of motivation does not exist in the open source world. The motivation is the "satisfying arcihtecture". Which means that drivers and programs have to be ported and ported over again, from version after version, to adapt to a changing infrastructure. Which is, from an engineering perspective, probably satisfactory, but it leaves users alone with anoying compatibility problems.

With such simple problems I just observed (probably expectedly) with the SCSI2SD installation software. You cannot take a piece of binary executable and expect it to run on your average Linux box. There will always be some form of compatibility issue between some library somewhere. Linux means: The only way how you get software working on Linux is either take it from your Linux distribution, or port it and compile it yourself. With windows, I install the binary and it works.

That's again the experience over and over again - which of the two models is more satisfactory *for you* is then of course another question. If I just want to solve a single problem ("get my SCSISD working"), the windows solution is quicker.

[Minuous]

AROS I believe is a recreation of 3.1, whereas we already have 3.9+BB1-BB4 that updates all of this.

Exactly. I don't know why AROS is considered to be a "NG" AmigaOS when it is just a clone of OS3.1. OS3.5 and OS3.9 have been out for about 17 years now. Still waiting for them to be supported by AROS.

[Pat the Cat]

Exactly. I don't know why AROS is considered to be a "NG" AmigaOS when it is just a clone of OS3.1. OS3.5 and OS3.9 have been out for about 17 years now. Still waiting for them to be supported by AROS.

Just my opinion, but... I can see no advantage in moving up. I mean, I want to run Amiga software. That means, something written for a machine with a 3.1 ROM. Or earlier.

How many full size applications are there specifically for 3.5 or later? Erm... not enough.

[EvilGuy]

Because then somebody else will break the machine for you. Seriously, if you connect a system to the internet today, and you do not keep the system up to date, then you're soon running into trouble.

Not every machine necessarily needs to be connected to the internet :-)

That's again the experience over and over again - which of the two models is more satisfactory *for you* is then of course another question.

Of course, and in my experience I've seen the most computer-illiterate people use Linux on a daily basis without any of the hassles you're describing. Upgrades and all that mess as well. These people are nowhere remotely close to what anyone would describe as a "developer".

If I really wanted to mess with them, I'd give them an AmigaOS 3.1 machine ;-)