Jump to content

Ubuntu and the Unspoken Rules


securitybreach

Recommended Posts

From what I have read online, I think Linux Mint plans to stay with X until any successor proves to be better and as stable. I would think LM17 LTS will be an X based distro. LM has other problems being based on Ubuntu, such as more limited support timeframes for "interim" distros like LM 16.

Link to comment
Share on other sites

http://en.wikipedia.org/wiki/Mir_%28software%29#Software_architecture

 

Longtime Linux kernel developer Matthew Garrett criticized choice of licensing for Canonical's software projects, particularly Mir. Unlike X.Org Server and Wayland, both under the MIT License, Mir is licensed under GPLv3 – "an odd [choice]" for "GPLv3-hostile markets" – but contributors are required to sign an agreement that "grants Canonical the right to relicense your contribution under their choice of license. This means that, despite not being the sole copyright holder, Canonical are free to relicense your code under a proprietary license". He concludes that this creates asymmetry where "you end up with a situation that looks awfully like Canonical wanting to squash competition by making it impossible for anyone else to sell modified versions of Canonical's software in the same market".[36][37][38][39] Garrett’s concerns were echoed by Bradley M. Kuhn,[40][41] Executive Director of the Software Freedom Conservancy.[42]

 

http://en.wikipedia.org/wiki/Wayland_%28display_server_protocol%29

 

Wayland is a protocol that specifies the communication between the display server (called Wayland compositor) and its clients.[3] It was initially designed by Kristian Høgsberg, who already led the development of AIGLX, as a replacement for the X Window System. In stark contrast to the latter, Wayland clients will render without detour directly into their own buffer located in the graphics memory, through the use of EGL with some additional Wayland-specific extensions to EGL.

The display manager is to do the compositing, hence it will incorporate a big chunk of the functionality of current-day compositing window managers. It will composite those buffers to form the on-screen display of application windows. The Wayland protocol is essentially only about input handling and buffer management. The handling of the input hardware relies on evdev in Linux, and similar components in other operating systems.

The initial implementation, chiefly libwayland-server, libwayland-client, libwayland-EGL and the reference implementation Weston are written in C and published under the MIT License. Weston is written for the Linux kernel. An initial port to FreeBSD is in progress.

 

Mir is based in part on Weston/Wayland and Andorid. Apart from the licensing issue I belive there has been some concern over the Android parts.

 

From my point of view it is in everyones interest that all the main components of linux are kept totaly open source and free.That should really be written in blood and chiseled in stone.

 

:breakfast:

Link to comment
Share on other sites

In many ways the Wayland vs. X or Mir vs. X debate reminds me of the UEFI vs, BIOS debate. In both cases you have a legacy bit of code that's 30 years old but still doing the job, and a new development that's supposed to be better but maybe isn't.

The parallel isn't totally equivalent though - as BIOS is a rather featureless entity, and X has many formerly useful features that are just excess baggage in today's graphics universe.

Link to comment
Share on other sites

Guest LilBambi

No comparison really raymac.

 

UEFI and BIOS are proprietary.

 

X is NOT proprietary, it is fully GPL'd.

 

It is really apples and oranges.

 

This is not about old code vs new code. This is about licensing.

 

IMHO.

Link to comment
Share on other sites

I suppose, but I can never really get all that excited about licensing. It's whatever works best. A lot of stuff we use in Linux is proprietary. Look at the Nvidia and AMD drivers for example.

Most of the licensing debate occurs in userspace as I see it. Linus and the boys keep a tight rein on standards and licensing in the kernel development.

For me the issue with Mir would be that it might block a user from using another desktop with Ubuntu.

Link to comment
Share on other sites

Guest LilBambi

Licensing is extremely important when it comes to GPL. If licensing changes for core things in Linux we could all end up back at the commandline again permanently for anything free as in beer AND more importantly free as in free speech in Linux.

 

Just means many of us will be moving away from these tainted linux distros if that happens.

 

I am not talking about driver software that you can go proprietary or not at your leisure. I am talking about changing the guts of the backbone of the GUIs here.

Link to comment
Share on other sites

OK I see your point but I don't think we'll be back to command line any time soon unless we choose to be. X is still available and stable and I'm sure the tried and true distros like Debian or Fedora will still have it as an option.

Link to comment
Share on other sites

Hedon James

I'm kinda with Ray on this one. I have no issues with proprietary software, per se, especially if it offers something that the FOSS community does not; the caveat obviously being that a FOSS version exists. My objections tend toward a lack of choice, or choosing between the lesser of two evils. The graphical server is invisible to the user, so most users will probably be indifferent. Myself included, I just want it to work, and to work well! I see this as a developer issue rather than a user issue, and I can see why developers would have a problem with this. I see Bambi's "slippery slope" argument too.

 

Assuming that the Mir versus Wayland versus Xserver issue really is about licensing, this should work itself out over time. For now, the change and the resultant competition is a good thing, as far as I'm concerned. But if I'm a developer (I'm not, btw!) who's driven by the OSS ethos, I'm not contributing ANYTHING to the Canonical project due to their creative licensing policy. And if enough developers feel as I do, and follow suit, then Canonical's project will stall and wither, while Wayland bursts ahead (or another future fork that does not yet exist). We've already seen this play out in the OpenOffice exodus that forked LibreOffice, resulting in LibreOffice mostly replacing OpenOffice en masse in linux distros.

 

In my opinion, all this noise is just developer squawk at this point in time. I believe it will get resolved, one way or the other, and it will be invisible to the user. If I am wrong, and it starts to affect me as a user, I'll find a new distro that solves the problem for me. Perhaps from the Debian branch, or perhaps Arch, as there seems to be an inordinate group of Arch users here, compared to other forums I have frequented before?! I'll switch to an LXDE/OpenBox desktop and do my best to make it look and behave like Unity. Because I can! :thumbup:

Link to comment
Share on other sites

So this is all a storm in a tea cup is it !

 

What difference does a small licensing issue matter , eh !

 

Nowt to worry about.

 

Bury your head in the sand at your own peril.

 

 

Microsoft had worked with Apple Computer to develop applications for Apple's new Macintosh computer, which featured a graphical user interface. As part of the related business negotiations, Microsoft had licensed certain aspects of the Macintosh user interface from Apple; in later litigation, a district court summarized these aspects as "screen displays". In the development of Windows 1.0, Microsoft intentionally limited its borrowing of certain GUI elements from the Macintosh user interface, to comply with its license.

For example, windows were only displayed "tiled" on the screen; that is, they could not overlap or overlie one another. There was no trash can icon with which to delete files, since Apple claimed ownership of the rights to that paradigm.

 

An so the saga began.

 

Remember it is not always the BIG things in life that are important.

 

If you do not fight for freedom do not complain when the man with the big stick comes calling.

 

:pirate:

Link to comment
Share on other sites

securitybreach

Well after using Linux for over a decade (11 years), I can tell you that licensing is extremely important to Linux and Open source. Linux would not be what it is today without the GPL and the enforcement of it. Now I do understand that some need proprietary software for things like video drivers but besides that, everything should adhere to the GPL without exception. If you want to use closed software, there are other OSs out there. Personally I do not think there would be Linux (distros) or open source without the GPL as Linux would of just stayed as a kernel, nothing more. Without being able to view the source code, development would basically halt and Linux would never have became what it is today. So yes, licenses are very important!!

 

https://gnu.org/licenses/gpl.html

Link to comment
Share on other sites

Hedon James

An excellent example of Bambi's slippery slope argument Barbarian! I get what you're saying. And I'm not a fan of the Mir license. If I was a developer, I would NOT be contributing to the Mir project. Perhaps I'm wrong, but I think that battle is being fought as we speak. But I'm not a developer with the power to affect an outcome. It's not practical for me to learn how to code so that I can NOT contribute code to Mir. I'm a user; what can an end-user do to have a voice in this outcome?

 

All I can think of is to vote for a choice. I'd prefer if Ubuntu made the Mir license compatible with FOSS licensing principles. If that's not possible, I'd like it if they'd use Wayland instead. If that's not possible, I'd like it if the proprietary Mir code in Ubuntu plays nice with the rest of my system, without interference. There is precedence for this in our available choices of proprietary video drivers versus OS drivers, as Ray pointed out. Similarly, proprietary codecs such as mp3 and mp4 are installed on many linux systems in order to enjoy music and video. Not quite the same, but close, another example might be the "shim" software developed by various distros (I'm not sure if there's a "standard" yet?) in order to "break" the proprietary UEFI software and install the choice of an OS linux distro in a dual-boot configuration, or outright replacement. We are still "free" because the choice of an OS alternative exists, and THAT is the important ingredient.

 

If I can't experience any of the above preferred outcomes, all I can do as an end-user is to abandon ship and find another distro. There is precedent in this also, as many former Ubuntu users jumped ship when Canonical started peddling the Unity desktop, and removed the choice of an alternative. I was almost one of them. The alternative choices have been re-introduced, but the damage has been done; those users aren't returning, as far as I can tell.

 

Which brings me to my final observation of precedent, the debate of init vs. upstart vs systemd. Summarized in this article:

http://www.zdnet.com/after-linux-civil-war-ubuntu-to-adopt-systemd-7000026373/

 

Init is the antiquated technology that could use some updating to take advantage of more modern coding on modern hardware. Ubuntu developed Upstart in order to accomplish this. Years later, Red Hat brought systemd to the dance. A big battle ensued. With Debian announcing they would switch from upstart to systemd, the battle was effectively over and Canonical conceded they would also use systemd. Much like the graphical X-server software, Init is another of those invisible-to-the-user, but CRITICAL piece of software that affects the entire user experience. Substitute X-server for Init; Mir for Upstart; and Wayland for Systemd; and the battle is nearly identical, with many of the same players. IMO, it is being resolved as we speak, but it's a process, not an event. And the process is doing exactly what it is designed to do. So far, so good.

 

As long as we have the choice of Mir vs Wayland vs X-server, we'll be fine. But I'll absolutely positively 100% agree that we need to "fight for freedom". I just see more than 1 potential path to get there and, for now at least, I'm relying on those with a compass to lead the way. For now...

  • Like 1
Link to comment
Share on other sites

Guest LilBambi

As a user, our voice, and if necessary later, our choice will speak for itself, and that helps us continue the fight for open operating systems, hardware, software, information.

Link to comment
Share on other sites

Unless we're actively developing software (I am not either) there's very little any of us can do aside from make choices as a user. And let's face it, right now if you want to go 100% FOSS, you'll live in a parallel universe to a lot of computer users:

  • You won't be playing MP3s or using proprietary codecs.
  • You won't use the proprietary video drivers
  • You'll probably be connecting your laptop by Ethernet because 3/4 of the wifi adapters out there won't work for you.
  • You'll be saving your Office files in Open Document Format.

It certainly can be done, and if you feel that strongly about it, more power to you. I did not claim that the licensing issue was trivial, but I do feel that the user experience will dictate what happens to Linux just as much as the licensing issue will.

For my part I'll always choose the open source solution if it works OK. I'm very happy to use the AMD open source video driver and keep the Linux graphics stack intact. I try to get Atheros or Intel wifi where I can.

Edited by raymac46
  • Like 1
Link to comment
Share on other sites

securitybreach

Well actually I do not use proprietary drivers (FLOSS radeon driver) and my laptops have Atheros chipsets so their open source. I do play mp3s though....

Link to comment
Share on other sites

Guest LilBambi

Unless we're actively developing software (I am not either) there's very little any of us can do aside from make choices as a user. And let's face it, right now if you want to go 100% FOSS, you'll live in a parallel universe to a lot of computer users:

  • You won't be playing MP3s or using proprietary codecs.
  • You won't use the proprietary video drivers
  • You'll probably be connecting your laptop by Ethernet because 3/4 of the wifi adapters out there won't work for you.
  • You'll be saving your Office files in Open Document Format.

It certainly can be done, and if you feel that strongly about it, more power to you. I did not claim that the licensing issue was trivial, but I do feel that the user experience will dictate what happens to Linux just as much as the licensing issue will.

For my part I'll always choose the open source solution if it works OK. I'm very happy to use the AMD open source video driver and keep the Linux graphics stack intact. I try to get Atheros or Intel wifi where I can.

 

We are not talking about user installed proprietary programs or drivers here. They have always been available.

 

We are talking about the foundations, the guts of the GUI, and licensing of said guts.

Edited by LilBambi
Link to comment
Share on other sites

Hedon James

OK I have nothing more to contribute to this discussion so I'll butt out.

 

You're contributing just fine Ray! And your opinion(s) are just as valid as mine, Bambi's, Breach's, Barbarian's, and anyone else posting here that I inadvertently left out.

 

I don't presume to speak for everyone, but I'd be willing to bet that EVERY LAST PERSON in this thread would be 100% in agreement that FOSS is a good thing and preferable, and that proprietary is NOT! I also think that less than 100% would agree that SOMETIMES proprietary is a necessary evil, and hopefully a temporary fix.

 

Lastly, I'd be willing to bet that the only thing we're disagreeing on is just exactly where does that line get drawn? Richard Stallman draws it at the front door; Windows and Apple both ask "what line...?" The rest of us are somewhere in between, to varying degrees. My position was already formulated in my mind, but you brought up the point about codecs and hardware drivers that I hadn't explicitly thought of; but when you said it, it resonated in my mind with what I was already thinking.

 

But Bambi has me thinking too. She draws a distinction between user-selected proprietary software and distro-provided proprietary software. Good point. She also distinguishes between system-critical software (I would place graphics drivers in there too, btw!) and esoteric codecs for entertainment. I don't know that I'm right, or she's wrong, but I do appreciate the exchange of information and ideas. I haven't changed my opinion yet, but my perspective has been broadened.

 

In a nutshell, I don't really disagree with ANYTHING that anyone has stated here. I guess I just believe that it's going to be resolved before it gets to my desktop. I've got 3 years left on the support period of my 12.04 LTS release, so I'm counting on a resolution before then. But if it doesn't happen in that timeframe, I HAVE CHOICES!!!! I can't imagine the entire Linux ecosystem becoming corrupted by a no-choice proprietary graphics stack, but if that happens, I've got no problems looking at BSD, or just installing a legacy distro and locking it down from updates. Again...choices!

  • Like 1
Link to comment
Share on other sites

Guest LilBambi

Yes! The discourse is what is important. Hashing out thinking on this is important.

 

I personally do not put graphics card software in there necessarily as mission critical because, for mission critical computing, the FOSS drivers work just fine. Even watching DVDs as long as you have the 'restricted/proprietary repos', the codecs, etc. that you must install 'generally' install on your own, rather than being part of the normal distro repos. This is how Debian does things. They separate out the proprietary stuff from the normal repos into a 'restricted' or proprietary repo. for software and drivers that will taint the kernel/installation.

 

if you wish to do other things like gaming, you may want to have an nVidia driver from them rather than the more generic one, as an example.

 

But as I say, all of that is user installations of software and drivers, not the backend so to speak as something that the actual graphical user interface rides on. That is critical stuff, stuff that gets between you and your computer. Stuff that can not only taint the kernel but be between you and everything you do in the GUI (not the same as kernel layer of course) but still. And because it's proprietary, you (meaning open source developers who can enlighten us users) can't see the code, and can't separate out what is in the pipeline between you and what that developer put in there. Just one more step really to bring it closer to Trusted Computing Platform's original mandates for Windows, and likely Mac too, and they will get back to that. They are just bringing it on more slowly, turning up the heat more slowly this time because it was rejected before.

 

I am just saying we need to really think about these things and be prepared as Hedon mentioned to make that choice if things are not going as they should for an open GPL'd operating system. That's all.

Edited by LilBambi
changed 4th paragraph quite a bit more!
Link to comment
Share on other sites

It's not that I feel I'm being shut out , it's just that I am not a lawyer and all the legal ins and outs just isn't my bag. I think it would be great if we lived in a totally FOSS universe with Linux but right now 95% of us don't so let's not kid ourselves.

With viable alternatives to Canonical's development, we'll be fine though. Nobody can convince me otherwise.

Even if Canonical took a future version of Mir private, there are lots of creative folks who could fork the last free version and carry on. If developers can come up with Nouveau with no knowledge whatsoever about Nvidia's code, I am certain they'll take good care of Mir - if it's even needed.

If Canonical/Ubuntu goes on its merry way to become a third proprietary O/S they'll have to butt heads with Apple and Microsoft, while we lose one distro of many.

Edited by raymac46
Link to comment
Share on other sites

V.T. Eric Layton

OK, let's take a nice deep breath...

 

Repeat after me...

 

AHHHHHHHHMMMM...

 

AHHHHHHHHMMMM...

 

AHHHHHHHHMMMM...

 

There we go. That's better. Now let's all remember that in the grand scheme of things none of this matters at all.

 

Have a nice day! :)

Link to comment
Share on other sites

Just to clear up a couple of points

- Most codecs are included in libav, no need for proprietary for mp3 etc

- Free graphics drivers have come a long way. I'm using nouveau for my GTX 560Ti with less problems now than I had with Nvidia proprietary blob. And Nvidia are helping with nouveau these days.

- Atheros firmware is proprietary. Included in the kernel but removed from Debian to comply with DFSG.

 

Non-Foss on my system:

roger@brain:~$ vrms
		   Non-free packages installed on brain

firmware-atheros				    Binary firmware for Atheros wireless cards
unrar							   Unarchiver for .rar files (non-free version)
virtualbox-guest-additions-iso	  guest additions iso image for VirtualBox

		    Contrib packages installed on brain

flashplugin-nonfree				 Adobe Flash Player - browser plugin
virtualbox						  x86 virtualization solution - base binaries
virtualbox-dkms					 x86 virtualization solution - kernel module sources fo
virtualbox-qt					   x86 virtualization solution - Qt based user interface
virtualbox-source				   x86 virtualization solution - kernel module source

 3 non-free packages, 0.2% of 1747 installed packages.
 5 contrib packages, 0.3% of 1747 installed packages.

Link to comment
Share on other sites

I use the ath9k driver which I believe is FOSS. Here's my vrms readout.

 

Non-free packages installed on ray-Basement-Desktop1

 

libfaac0 AAC audio encoder (library)

nemo-fileroller File Roller integration for Nemo

unrar Unarchiver for .rar files (non-free version) - binary

 

Contrib packages installed on ray-Basement-Desktop1

 

ttf-mscorefonts-installer Installer for Microsoft TrueType core fonts

virtualbox-4.3 Oracle VM VirtualBox

virtualbox-guest-dkms x86 virtualization solution - guest addition module so

virtualbox-guest-utils x86 virtualization solution - non-X11 guest utilities

virtualbox-guest-x11 x86 virtualization solution - X11 guest utilities

 

3 non-free packages, 0.1% of 2173 installed packages.

5 contrib packages, 0.2% of 2173 installed packages.

 

 

ETA: If you use the Atheros driver for USB based wifi cards you need firmware. If you have the PCI or PCi-e based card you don't. My netbook with an Atheros chipset started right up with the Debian net installer.

Edited by raymac46
  • Like 1
Link to comment
Share on other sites

I use the ath9k driver which I believe is FOSS.

Driver is FLOSS. Firmware is not, but is included in vanilla kernels but not in Debian. Mine uses ath9k_htc but needs firmware-atheros installed as well or it won't work.

Link to comment
Share on other sites

Exactly - ath9k_htc is the USB driver that needs firmware & ath9k (PCI/PCI-e) doesn't need it. Debian installer worked just fine without any firmware added on my netbook.

  • Like 2
Link to comment
Share on other sites

securitybreach

Yup, the wireless cards on my laptops are built in and use ath9k not ath9k_htc so they are using the floss driver. Here is the output from my netbook:

 comhack@Pluto ~ % lsmod | grep ath9
ath9k 98067 0
ath9k_common 1874 1 ath9k
ath9k_hw 398484 2 ath9k_common,ath9k
ath 18909 3 ath9k_common,ath9k,ath9k_hw
mac80211 477209 1 ath9k
cfg80211 409191 3 ath,ath9k,mac80211

 

ath9k is a completely FOSS wireless driver for all Atheros IEEE 802.11n PCI/PCI-Express and AHB WLAN based chipsets.

http://wireless.kern...s/Drivers/ath9k

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...