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

[EugeneNine]

That's not what I'm saying. I'm saying exactly what I said above: "AmigaOs needs a maintainer, not a bunch of hackers". And that's exactly going to happen if you open source it.

Look at Linux: It's a nice operating system for developers. It's a poor operating system for end users. The average open source guy develops for his particular needs, and not for the need of the user - which means that anything that is of utter importance for creating a working software infrastructure is ignored: Stability of software interfaces.

Open source projects may start out as a simple hacker tool but as they become bigger/more popular they tend to fall into standard code practices (many of which were developed out of open source projects).  For example myself , as an end user of Linux and/or KDE have submitted requests and saw those implemented.  On the other hand myself as an end user of a certain closed source OS have paid $200 to open an incident to make a request only to have it denied.  So I've found that Open source and Linux works better for end users myself.

The GNU/Linux system - and I do not talk about the kernel interface in particular - changes on a daily basis. If you get a binary from yesterday, you do not know whether it will continue to run today because somebody surely played with the interface of some system library somewhere.

This is not acceptable for an end user product - breaking legacy software isn not an option, and even less so in the Amiga environment which only has legacy software.

Yes the code changes but if your an end user and not a developer your not pulling the daily code changes, your running the stable versions so these changes don't affect you unless you intentionally upgrade portions of the software your running and you should be reading the change logs to see what changes first.

Examples? Ok, here are two: Years ago, there was a nice XMMS plugin to play Amiga "chip" tunes through an UAE interface layer. XMMS changed the plugin interface when porting to Audacity for no apparent reason, and the player broke. I took great effort to port it to the "new and improved" Audacity interface then - for my own needs - just to find out that the open source "hackerz" changed the interface *right again*. Why? There was no reason to - it worked the way it was.

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

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. The customer is not the user. The customer is the coder.

Now, consider what that means for the Amiga "market"? It means - already - a lot of frustration due to a lot of incompatible software floating around, and a software infrastructure that consists entirely on legacy software.

This sounds like a plan for utter failure for me. If you want open source, nothing beats Linux. I'm using it myself, works for me. But that's a different market with different goals, and that should not be confused with the Amiga ecosystem, which is something entirely different. I *cannot go along* and change the interface of "layers" just for the fun of it, and break old code. It's a big no-no.

How many examples of the same issues happening to close source software do you want, I can provide just as many as this issue is not specific to Open Source, this issues happens with any software.

[EugeneNine]

How should I know? Ask ASUS, it's their product, after all. If you installed an operating system the system was not designed for, then that's certainly not ASUS' problem and they will tell you. The product probably worked when it arrived at your Mum's home. Did the system come with Windows 10? Probably not. So if you install something on the machine ASUS doesn't give you a warranty for, that's then entirely your problem.

No, I'm not running a windows repair shop, I'm not interested in this work.

Same finger pointing issue I run into I bet.  Bought my wife a laptop with windows 8.  It got a forced upgrade to 10.  Dell says its Microsoft's fault, Microsoft says since it was an OEM windows originally its Dells support issue.  Meanwhile wife can't print on legal size or use headphones.

Daughters netbook got forced from window 7 to 10, it takes 28 minutes to boot.  I bought her a $200 'refurbished' laptop and put linux on the netbook, took less than 20 minutes to do the whole install.

But users are stuck in the middle, they buy a system, then get a forced (windows) OS "upgrade" and when it doesn't work the OEM points at Microsoft and Microsoft points at the OEM.

[EugeneNine]

I'm not unhappy about my choice. I'm just telling you what the drawbacks of open source are, and that you cannot expect the average user to handle such a system. Open source is for developers. You are a developer, I'm a developer, but my Mum is not. She cannot install a printer with cups - I can. She *probably* could install a printer on windows, but not on Linux.

The "drawbacks" you list open source are specific to open source, those drawbacks are common to software in general (open or closed source).

Have you tried to install a printer in current versions of windows?  Its a lot easier in Linux now a days.

[EugeneNine]

Who forced you? I'm pretty sure I was never forced to upgrade to anything.


If Microsoft provides an upgrade that doesn't work on your machine, it's Microsoft's fault if the upgrade doesn't work. Unroll the upgrade, and the problem is gone. If Dell claims that this machine supports Windows 10, it's a Dell problem.

The forced windows 10 "upgrade" is well known and takes lots of steps to prevent it that are not end user friendly and even 'hackers' have to keep up with the new methods every time Microsoft figured out a way around the other methods of preventing it.

Well unrolling the upgrade now requires me to pop in the CD and spend a couple hours watching windows reinstall then reinstalling all her software.  So its not time cost effective to do.

[EugeneNine]

Insert the CD, click on install. For Linux, I first need to hand-pick a vendor that supports Linux. HP is in general fine, though it requires for new printers not supported by the distribution a binary blob that - typically - does not work out of the box on a given machine, due to incompatible and unstable software interfaces. Been there, done that.

Yes, I got everything working, but it was a hassle.

Its windows 10 and an HP laserjet.  So inset HP CD, wait for it to install then update itsself from the web and then have her try to print on legal size and it cuts off the print at letter size so we open a case with hp who says reinstall their software.


Or I run linux so I plug HP printer into the network, enter my root password and its done.  I then try to print her legal size statements and they print in legal size like they are supposed to.

[EugeneNine]

Microsoft did use some pretty sneaky tactics there, for a bit.  For example when they changed the [X] close gadget to an "implied consent" to allow the upgrade to proceed.  Even they admit that was a "poor decision" on their part.

Still not the same as "forcing" someone, however.

Both systems were running windows 8 and 7 when they went to bed and both had 10 when they woke up, no prompting.  And I had did some of the "prevent windows 10" but I hadn't had a chance to see if there were any new methods to prevent the windows 10 upgrade that day.

I don't know, I have very little sympathy for someone who says that  their computer runs poorly.  Or their car, or the plumbing in their  house, or anything else, for that matter.  Take the time to understand  it and make it better.  That's the whole "if you want something done  right do it yourself" attitude, lol. :lol:

Thats why I went OS.  I can't fix some of windows issues without the source.

I opened a ticket with MS on typeperf.exe.  In windows 2000 it was a resource kit tool.  It was bundled as part of the OS in server 2003 so it then became officially supported.  If you call the cpu % utilization with an * for the CPU/Core # it will sometimes give a - number.  I was able to duplicate it and opened a case and MS decided to not fix it instead giving a workaround.  So I provided the workaround to the monitoring team and for the next couple years they hated me because every time they put an invalid cpu % utilization ticket in my queue I'd send it right back to them with MS's workaround (which wasn't feasible to implement BTW).

My point is that every 'fault' of open source is in reality a 'fault' with software development in general no matter if the source is opened or closed.
For every open source developer that doesn't value their users input there are just as many that do.  Likewise there are close source developers who think they know what the end users needs/wants better than what their users do (Apple/Microsoft).

I seem to recall Commodore developers were often called out for not listening to what their users needed/wanted.  But when it comes down to it if you get stuck with a close source package that has unresponsive developers your stuck.  If you use an open source package and the developers are unresponsive then you can simply say fork it and make it the way you think it should be.  You don't have that option with close source.

I wish they would just OS the code and let us work on it.  For now AROS is the next best and when I don't have an application in AROS I need Linux works best.  I don't have to check for the daily trick to prevent upgrade with either.

[EugeneNine]

That's your first mistake.  Figure even if you bought something brand new, off the shelf today, that it's been sitting in that box for months and is almost certainly outdated.  Never, ever use the CD that comes with a piece of equipment.  Always download the latest drivers off the manufacturer's website.  Maybe that will help?

replace cd with download software, whatever.
Either way its been uninstalled, reinstalled, etc multiple times and still won't print legal size.

[EugeneNine]

Try installing it on a different Windows PC.  Isolate if it's an issue with the printer, or with the software.  My guess then is that it's a stuck setting somewhere.  Sometimes they're buried several layers deep.

Daughters netbook was the only other windows 10 pc.  We've managed to keep the rest at 7 which works fine.
Her netbook was useless in 10, took 28 minutes to boot to a login and then was so slow as to not be usable, don't know why they would even try to push it.  She had a chromebook and used it until I found another good deal on a 'refurbished' latitude and she uses it now.

Just not worth the time to troubleshoot anymore, I just hand her a Linux laptop when she needs to get work done.  Windows it pretty much just a game os anymore.

Mine all have Linux from my current Latitude E6230 down to the Latitude C400.

[EugeneNine]

My experience with HP printers specifically and printers in general is  that if they are supported in CUPS, they WILL work a lot better under  Linux than under Windows.  Windows has always had a really crap printing  system.  I mean it's just absolutely terrible.

I recall having  to do some weird voodoo to get my mother's printer to even work after it  randomly stopped printing.  All with install drivers, make sure it's  not plugged in, plug it in when it tells you, oh wait, it didn't detect,  okay unplug, reboot, remove drivers, reboot again, install drivers,  plug in printer.. oh is going to work?  Nope, try again...

Ohh, the whole plug it in, unplug it, plug it in again thing is because USB is crap in windows too.

This is a network printer though.

But I've setup other printers in Windows too and its still too much work in Windows compared to Linux.

So I mentioned earlier that software changes affecting you are not just an issue in open source but also close source.
Many years ago I tested Office 2010 in my company and they asked for feedback so I took the time and wrote up all my issues and then rolled back to 2007 and a couple weeks later got an e-mail claiming all the issues I reported didn't exist as both my helpdesk and Microsoft had thoroughly tested office 2010.  When SP1 for Office 2012 was released I found all of my reported issues in it as fixes.  So I have to at least give them credit for fixing those even though they denied they existed.  I still have a couple old KB numbers that haven't been fixed.