Mir's Abstraction Layer Now Has Cut & Paste Support (MirAL)

[Vistaus]

Linux/tech forums, non distribution specific, tend to have the largest number of *buntu users, when you see "Linux in the wild" photos from companies and institutions in 95% of cases it is Ubuntu, with Unity too, nearly every software vendor tends to support Ubuntu first, then other distributions, Steam still officially only supports SteamOS and Ubuntu, Ubuntu makes most of the headlines on tech sites, Ubuntu gets preinstalled on a bunch of laptops, where is Mint preinstalled? And these are just a few examples. So Mint still "much more popular" than Ubuntu? With the advent of Ubuntu MATE especially many Mint MATE users went back to Ubuntu, portion of them migrated to Arch Linux and its user friendly derivatives like Manjaro, Mint's popularity was high for a while until they chose to go LTS only, due to obvious lack of manpower to maintain short term releases, and even those LTS based versions are months late with point release updates compared to Ubuntu LTS.

Cinnamon desktop and their so called X applications, are an attempt to establish a brand for themselves and stop being just another Ubuntu clone with some paint slapped on it, that didnt work out too well so far. Their problem is the LTS base many users consider too old to use for 2 years and the lack of any real incentive to use Mint, Cinnamon desktop doesnt bring anything new to the table which we cant already have with other desktops, Unity at least offers a completely different UI paradigm, dash and scopes, it IS different, Cinnamon reinvents the wheel in many of its aspects, also what is the point in using Mint KDE, MATE or XFCE when they differ so little from Kubuntu, Ubuntu MATE and Xubuntu? Most people prefer going for the real thing, and *buntu has newer short term releases available too. When people say "Mint is more popular than Ubuntu" that is merely wishful thinking, not based on any real facts of any kind, pro Ubuntu facts I presented here are at least something it can be backed with some kind of evidence, though numbers are unknown because there is no counter in Linux distributions, I think they should employ those in order to end discussions like this.

To clarify: I wasn't advocating Mint or anything (I don't even like that distro!), I was just curious how you came up with the amount of users.

It's true that Ubuntu pops up more in companies, I frequently see Ubuntu on TV when journalists are filming in a place where there are computers (and one store chain nearby, Eye-Wish, has been using KDE on some Linux distro for years now). But I thought we were talking about end users, not companies (if so, isn't Red Hat popular either?).
The media reporting more about Ubuntu doesn't say much, nor does the amount of users on a forum. There are probably lots of Linux users who aren't joining any forums or have accounts that they aren't using anymore (I, for example, am still registered on Ask Ubuntu but it's been quite awhile since I've last used Ubuntu so that doesn't say much). The media doesn't really count either. For example, I frequently see iOS stuff making the headlines even though we all know Android marketshare has far surpassed iOS's. So again: that doesn't say much.

Thanks for clarifying though! I'm not sure what distro currently holds the crown, but I do think that Ubuntu is still somewhere in the top 5 (if not top 3).

[bregma]

As well, trolls aside, if they were to abandon Mir efforts, I would think it's more likely that they would change their display server to use Wayland than switch back to GNOME.

Completely rewriting a display server to use a different wire protocol, a different buffer management architecture, and different window management paradigms, would be faster than just switching to a different display server? I think your grasp of economics is a little odd.

Nope. I stand by my analysis that it's either finish Unity 8 or drop it. Re-architecting it and rewriting it from scratch at this point just to satisfy a few zealots who haven't sunk a dime or a line of code into anything related is a non-starter.

[Mystro256]

Completely rewriting a display server to use a different wire protocol, a different buffer management architecture, and different window management paradigms, would be faster than just switching to a different display server? I think your grasp of economics is a little odd.

Nope. I stand by my analysis that it's either finish Unity 8 or drop it. Re-architecting it and rewriting it from scratch at this point just to satisfy a few zealots who haven't sunk a dime or a line of code into anything related is a non-starter.

I don't think you understood what I was saying. It wouldn't make sense for Ubuntu to "switch to GNOME shell" as you originally said. Possibly you meant using the GNOME shell display server stack for Unity? Perhaps I should reword it to "I would think it's more likely that they would use a Wayland display server rather than switch back to GNOME". It doesn't seem likely they would give up on Unity given on how much they've put into it. If they were to give up on Mir, it's more likely they would move to a Wayland display server, home grown or taken from existing code.

[Mystro256]

It isn't more popular. Ubuntu dwarves pretty much every other distro in terms of the number of users.

I said "it seems like Mint is much more popular than Ubuntu these days", which would mean I'm not claiming Mint is more or less popular, but rather seems like it, at least to me and the articles I've read. You on the other hand are claiming with pretty strong confidence that Ubuntu is the most popular but I don't see evidence to back that up.

As well, I don't use or have stake in either (I only use Debian, Fedora and Centos, while I develop for Fedora and EPEL), so this isn't me being biased and being a fanboy, but rather saying that his use of "most typical users" is either short sighted or he misspoke and meant "most Ubuntu typical users" instead.

[bregma]

I don't think you understood what I was saying. It wouldn't make sense for Ubuntu to "switch to GNOME shell" as you originally said. Possibly you meant using the GNOME shell display server stack for Unity? Perhaps I should reword it to "I would think it's more likely that they would use a Wayland display server rather than switch back to GNOME". It doesn't seem likely they would give up on Unity given on how much they've put into it. If they were to give up on Mir, it's more likely they would move to a Wayland display server, home grown or taken from existing code.

No, I said what I meant. Unity 8 is a Mir server. To toss Mir, you have to toss Unity. Unity 8 can not be retrofitted onto Mutter, and the entire Unity 8 desktop environment can not be retrofitted on to Gnome. If you toss one key piece of the stack, you have to toss the entire stack.

Switch the wire protocol in Mir to use Wayland? Sure. A non-standard Wayland because of the difference in functionality and window management paradigms, so you'd lose the advantage of a single Wayland client library (but quite frankly you're already getting there with Wayland where everything is an optional extension, like OpenGL does but more bigly, and not every compositor using Wayland provides all of Wayland, with a few custom things thrown in on top for various implementations). Advantage? None. A whole lot of expensive development work running to stand still for zero gain to end users.

What I meant was if Canonical were to abandon Mir, it would mean abandoning Unity. Full stop. No carrots.

[duby229]

No, I said what I meant. Unity 8 is a Mir server. To toss Mir, you have to toss Unity. Unity 8 can not be retrofitted onto Mutter, and the entire Unity 8 desktop environment can not be retrofitted on to Gnome. If you toss one key piece of the stack, you have to toss the entire stack.

Switch the wire protocol in Mir to use Wayland? Sure. A non-standard Wayland because of the difference in functionality and window management paradigms, so you'd lose the advantage of a single Wayland client library (but quite frankly you're already getting there with Wayland where everything is an optional extension, like OpenGL does but more bigly, and not every compositor using Wayland provides all of Wayland, with a few custom things thrown in on top for various implementations). Advantage? None. A whole lot of expensive development work running to stand still for zero gain to end users.

What I meant was if Canonical were to abandon Mir, it would mean abandoning Unity. Full stop. No carrots.

It seems, and has seemed for many years now, like that is their best option. Of course they can choose to waste their time and effort all they want.

[paupav]

They have created display server on their own, new packaging system on their own and new desktop environment on their own. It is almost finished and it is shaping up pretty well. Packaging system is done. Desktop environment is almost done. And display server is almost done.

[TheBlackCat]

They have created display server on their own, new packaging system on their own and new desktop environment on their own. It is almost finished and it is shaping up pretty well. Packaging system is done. Desktop environment is almost done. And display server is almost done.

The display server has been "almost done" for four years now. It has been perpetually 1 year away for its entire existence. And it is still one year away now. So I will believe it is "almost done" when they get considerably closer to release than 1 year.

[boxie]

While libmiral is part of the desktop environment it is not involved in the cut & paste support: With Unity8 cut & paste is provided by the "content hub" service.

The miral-shell program is just an example of how to use libmiral, and is used to demonstrate its features. But it is not part of the production desktop environment and no-one has intentionally configured "content hub" to work with it.

That is actually good to know. Thank you for shedding some light into the ... mirky? depths of DE development :P