Lumina: PC-BSD's Own Desktop Environment 148
jones_supa (887896) writes "The PC-BSD project is developing a new open source (BSD license) desktop environment from scratch. The name of the project is Lumina and it will be based around the Qt toolkit. The ultimate goal is to replace KDE as the default desktop of PC-BSD. Lumina aims to be lightweight, stable, fast-running, and FreeDesktop.org/XDG compliant. Most of the Lumina work is being done by PC-BSD's Ken Moore. Even though Lumina is still in its early stages, it can be built and run successfully, and an alpha version can already be obtained from PC-BSD's ports/package repositories."
Why? (Score:1)
What is wrong with XFCE, LXDE etc? Why can't we just make 1 environment that is really good, rather than 110 mostly complete ones?
Re:Why? (Score:5, Insightful)
I'm not even going to link to the xkcd comic, we all know it.
Besides, one of the awesome things about open source is anyone can attempt to build a better mousetrap for any reason they damn please. Yes it leads to fragmentation and a lot of duplicate effort, but it also leads to people trying out new ideas and having fun. This guy wants to make yet another window manager, all the power to him. Maybe it'll be awesome. Maybe it'll have some clever thing that gets used elsewhere. Maybe he'll get bored in a month or so. It's his time to waste regardless.
Re:Why? (Score:5, Insightful)
When being asked, why re-invent the wheel, the best reply is because just maybe the wheel isn't good enough.
I can think of numerous times where people tore everything down and started over and found some flaws in designs that wouldn't have been seen otherwise.
Re: (Score:3, Interesting)
Okay, but when you've torn everything down and started over from scratch twenty-plus times already, maybe that stops being the right development methodology?
Re:Why? (Score:5, Interesting)
What does "development methodology" have to do with it? Sometimes you just want to start from scratch rather than hauling along someone else's baggage. I guess your complaint just falls into the category of "dissatisfaction with how others spend their own time."
Re:Why? (Score:5, Informative)
You realise that when Linux came about, there had already been more than twenty UNIX derivatives, right?
Re: (Score:1)
You realise that when Linux came about, there had already been more than twenty UNIX derivatives, right?
And you do realize that Linux was not better in anyway to any of them? It was the licensing that was different, right?
Re: (Score:1)
You realise that when Linux came about, there had already been more than twenty UNIX derivatives, right?
And you do realize that Linux was not better in anyway to any of them? It was the licensing that was different, right?
Ok. I'll bite. Perhaps when Linux was new (1.x timeframe) it was no better than the rest, however, by the time the kernel reached version 2.0, linux had already surpassed many other unix-like systems. How do I know? Well, let me tell ya about a time before google... In 1997, I had a quad-boot system including a BSD variant, Solaris X86, Slackware Linux, and Windows 95 (yeah I know). By that time, Linux had surpassed the rest for my uses. Maybe it wasn't faster or more stable, but it was eminently
Re: (Score:2)
Not better in ANY WAY to ANY of them? Really? Have you never had to relink your kernel to add a tape drive? Had a tar archive corrupt because you had files paths over a hundred characters? Hit tab to auto complete on a bourne shell? Shelled out a grand for a basic ANSI C compiler? Had to explain to a client that the OS doesn't include a TCP/IP stack?
Re: (Score:1)
Okay, but when you've torn everything down and started over from scratch twenty-plus times already, maybe that stops being the right development methodology?
Yes, computers should only run Windows 8, and nothing else. I've done a number of things with computers that others find odd - just last week, bought a Chromebook, installed Linux on it then run Windows programs on the Linux side.
Can you imagine just how frigging stupid that is? Why the heck would anyone do that when there are plenty computers that will run that Windows program? Tough titty. I do it because I can - I'm having fun, and I'm learning. Then again, learning today is bad, berry, berry bad.
Re: (Score:3)
Well excuse me! I merely read the fucking summary and somehow thought just because it said that "the ultimate goal is to replace KDE as the default desktop of PC-BSD" and that "Lumina aims to be lightweight, stable, fast-running, and FreeDesktop.org/XDG compliant," that meant the point was to make something better for the public, not merely to "have fun" or "learn" something.
Clearly, your reading comprehension skills are so far beyond mine that you were able to determine the "real" ultimate goal of the pro
Re: (Score:2)
I'm so goddamn sorry I deigned to participate in the conversation, when my ideas so pale in comparison to your obvious brilliance!
Holy cannoli, man! Sit back, relax, maybe have a brewski. This thread is not worth getting that worked up about.
This is Slashdot. This is what we do here.
Re: (Score:2)
Hey, you know what they say: if something's worth doing, it's worth overdoing!
(Also, in retrospect, the tongue-in-cheek didn't translate to text as successfully as the sarcasm.)
Re: (Score:2)
I think it really depends on whether or not the first twenty times go things wrong.
I'd agree, there's not much point in starting from scratch when things are working exactly how you'd want. However, if there's nothing that does things correctly, you might need to start over. Or even if everything else works well enough, it still might be that you have other needs, different from the intended audience for the previous solutions.
Re: (Score:2)
When being asked, why re-invent the wheel, the best reply is because maybe the wheel isn't good enough [kickstarter.com].
Then again, we also got this [youtube.com]
Re: (Score:2)
Then again, we also got this [youtube.com]
So that's where the inspiration for Windows 8 came from!
Re: (Score:3, Interesting)
But what are they going to change that will make the effort worth it? When I look at the variety of desktops, the majority (perhaps all) of them seem to be tinkering with the same basic concept. It would be much more interesting if this splitting was leading to a drastically different desktop concepts, but it's not.
Re:Why? (Score:5, Insightful)
We got tiling WMs.
I'm not a fan of them, but it's kinda different.
Re:Why? (Score:5, Insightful)
Probably: BSD license and guaranteed support for BSD unixes. The former occasionally matters to the people working on the BSDs, the latter definitely does. (And is notably lacking in many of the current desktop environments - even if they do work on BSDs, they are often missing features and poorly maintained, with no interest in providing better support.)
Re: (Score:2)
At some point you are simply drowning in wheels, and cant move.
Re: (Score:1, Interesting)
Why indeed. The BSD license will ensure that if it ever does gain traction (i.e. is adopted by an Apple or Samsung or Microsoft for some product) that the community will never see any of that effort.
GPL is true freedom for your code. BSD is a corporate handout.
Re: (Score:1)
Why would either Apple or Microsoft want it?
Have you looked at sales of Win8 lately? Whoever re-invents the wheel and comes up with something trendy in terms of how things work as a windowing gui will either make a tonne of cash, or in the case of MS cloning the interface become the replacement for "Windows" which Microsoft is desperate to find.
The difficulty is that finding a new paradigm other than a "windowing" gui that catches on with J Q Public is not that easy. Tiles and touchy feely screen win8 just ain't making it the customer is not getting
Re: (Score:2)
GPL code is why we can't have nice things like ffmpeg/libav producing good-quality AAC. You see, there are open source libraries but we can't link them against ffmpeg and redistribute a binary that's capable of using the shared object, because GPL says so.
Re: (Score:2)
That's because you're trying to reuse a library that has a proprietary license. What did you expect? I don't see how a BSD-licensed project could include a proprietary-licensed library either.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Not necessarily.
While parts will be vendor locked in, these companies can support back for other components.
In many ways the boring/non marketable stuff is sent back in help improving the product. As if they choose to fork, that means they will need to fully support that fork continuing on. Vs. supporting back changes and being able to keep your product consistent with the core project.
awesome? (Score:2)
Maybe it'll be awesome.
No, it won't be awesome [naquadah.org].
Re: (Score:3)
The flip side of that is the old adage "divide and conquer", the OSS community is almost self-defeating at times. Long before the mouse trap is the kind of smooth experience users want the core developers have moved to their new and even more grand mouse trap refactoring/redesign/remake that'll fix all the fundamental issues they discovered in the last design. Not that it's really different from proprietary software, at work it's exactly the same I'd love to get rid of the old and in with the new because ev
Re: (Score:1)
Maybe it'll be awesome.
Nah, awesome [wikipedia.org] has already been made.
Re: (Score:2)
Life doesn't have to be a sad place. There's happiness all around us if you know where to look.
Hope you feel better soon :)
Re: Why? (Score:2)
Re: (Score:2)
Re: (Score:2)
Wouldn't razor-qt meet their goals?
(Except it's GPL2 and so we're back to the holy wars)
Name? (Score:1, Funny)
Terrible name. Now I can associate it with junk cars:
http://en.wikipedia.org/wiki/Chevrolet_Lumina [wikipedia.org]
Re: (Score:2)
This is a good point. The only way they could have screwed up worse would have been to name it "Pinto", "Gremlin", or "Aztek". (The latter has been said to have been the worst car in all of history, responsible for the downfall of Pontiac.)
Re: (Score:2)
From what I understand the Aztek was actually a good car, It was just butt ugly so no one really wanted it.
Re: (Score:2)
I have no idea if the Lumina was mechanically sound or crashworthy, and for all I know, it was. I didn't matter, because it's also butt-ugly (though not nearly as bad as the Aztek).
Being butt-ugly is sufficient for a car to be considered "bad". No one wants an ugly car (at least, not enough to make the car profitable to sell). The Aztek's lesson is clear: make a car ugly enough, and you can kiss your company good-bye.
Re: (Score:2)
I've google it and find it not so ugly, and it has a smart form factor. Hatchback, that's good for putting stuff in esp. when you fold the rear seats down.
For a really ugly car look at BMW i3, it's vomit worthy.
BTW, nobody outside of the US has heard of Chevrolet Lumina.
Re: (Score:2)
Interestingly there's kind of a weird interest in the Aztek right now due to it being featured on Breaking Bad.
Re: (Score:1)
ok had to look it up -- but honestly It looks like a car to me. People are strange.
Re: (Score:2)
I didn't think the Aztek was butt ugly at all. I thought it was interesting.
It is funny but I still see a good number of them running around. They might have been pretty good cars/SUVs/CUVs after all.
Re: (Score:1)
This is a good point. The only way they could have screwed up worse would have been to name it "Pinto", "Gremlin", or "Aztek". (The latter has been said to have been the worst car in all of history, responsible for the downfall of Pontiac.)
Now now, those were at least profitable cars. How about an unprofitable car for a name: Project Edsel [wikipedia.org]. This car cost Ford millions of $$ in 1950s money. How's that for a bad project name?
Re: (Score:2)
Last I heard, the Aztek was not profitable at all, which is part of why Pontiac ended up on the chopping block.
Re: (Score:2)
No man that was a four door Monte Carlo. A real street machine.
Re: (Score:2)
Re: (Score:2)
This is going to be something like KDE, as that is what Lumina will be replacing.
On the other hand, LXDE is also being rewritten to be Qt-based, so that will probably be something along the same lines.
Several mistakes (Score:5, Informative)
The summary contains several mistakes.
1. Lumina is not yet available in the ports tree, searches for it do not return anything.
2. The project is not trying to become the PC-BSD desktop, at least not yet. Right now it is in the early/experimental stages to see if making a PC-BSD only desktop is feasible.
3. There is no default desktop on PC-BSD. KDE is one of the install-time options, which include MATE, LXDE, Cinnamon and many others.
Re:Several mistakes (Score:5, Interesting)
Re: (Score:3)
Once upon a time, in a Slashdot epoch of heroes and myths and CowboyNeal and editors who would actually edit, you could work with the editor that accepted your submission and get an update to TFS.
Alas, the time of the Gods is gone, and all we have is Beta and today's "editors".
But I still find myself wishing for heroes, as foolish as it seems.
Re: (Score:2)
Thank you for owning up to your mistakes in the article, that's more than the editors around here seem to do.
Re: (Score:2)
My apologies.
Let that be a lesson to you: never post Phoronix stuff here verbatim. They are typically only linking to the real content, while 95% of their own article is just water leading to the link.
X or Wayland? (Score:2)
Anyone knows how it would compare to razor-qt ? (Score:4, Interesting)
Re: (Score:2)
Razor-Qt and LXDE is supposed to become LXDE-Qt or something such.
But so far I haven't actually run it (now I can't say that I so far haven't seen any actual release of it because currently I'm running Fedora and they didn't seemed to have Razor-Qt packages or whatever (at least I don't have it installed, you don't have to tell me "but there is Fedora packages on this and that page" because it would make sense in some being available but I would kinda had expected it to be in the regular repositories (if th
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Razor-qt is GPL2. They can't fork it and make something that's BSD-licensed out of it. I imagine that's the motivation, whether you agree with the motivation or not. FreeBSD (and by extension PC-BSD) has gone to a lot of trouble to expunge GPL'ed pieces, including binning gcc in favor of clang for the build process and the shipped core system.
Re: (Score:2)
I'm all for lightweight (Score:1)
but I have to admit, my desktop is something I look at very very often and since I have only modern machines running as desktops, I'd like the option to put on all the bells and whistles I want, even if it's only for vanity and nothing to do with functionality.
It seems there is this interesting trend to keep targeting mobile/low powered PCs which I've found to be kind of funny considering how much progress we've made with our desktops. This course of "mobile first desktop second" is annoying.
Re: (Score:2)
I have a modern-ish machine but no SSD. So I won't use the heavy desktops like Gnome 3, Cinnamon, KDE, Windows 7 because waiting for a terminal or file manager to show up whereas it was instant on Windows 3.1 or 98SE is not fun. I'm all for a "bells and whistles" desktop that would stay lightweight on I/O. Software rendering has to stay as an option as well in case I have driver or GPU crashes.
Re: (Score:1)
I just opened up Windows Explorer on Windows 7 and it took less than 1 second. And this is on a laptop I bought 6 years ago. Maybe you need to ditch the Pentium 3 shitbox you're running?
Re: (Score:2)
Maybe not. You can transfer all you want, but you can't make your priorities and preferences everybody's.
I doubt his machine is as weak as you fantasize. Maybe he just wonders why we should accept less absolute performance of the desktop over what we had 15-20 years ago even though the hardware is incomparably faster and has incomparably more resources.
Had to be done (Score:5, Informative)
The problem with xfce, gnome, and most of the other desktop environments is that they tend to focus on Linux and most of them have actually removed *BSD compatibility recently in favor of the latest trends in the Linux community.
XFCE famously dropped FreeBSD support for some functions in their file manager for example. Gnome told us to FSCK off entirely.
We have to fight back.
Re: (Score:3)
LXDE is a great asset btw, it has a good and real philosophy of components independant from each other. Pcmanfm is impressive, a nautilus/thunar clone with the strengthes of both (ignoring the nautilus 3.x feature depletion).
You can use everything LXDE and a different file manager, or pcmanfm and everything different if you wish. Had a fluxbox + pcmanfm + audacious with xmms/winamp skin for music playback on a 1999 computer and it was brilliant.
Lumina could just use pcmanfm-qt, give some minimal input as we
Re: (Score:1)
Re: (Score:2)
> Gnome told us to FSCK off entirely.
Why not? They already tell Linux users to fuck off when feedback is provided to them. Welcome to the club, BSD!
Re: (Score:1)
XFCE famously dropped FreeBSD support for some functions in their file manager for example. Gnome told us to FSCK off entirely.
This is true.
Since then, I've been searching for window managers (I've already given up with full desktop environments, because they install much crap I never need and are slow). I used OpenBox quite long, because it was easy to manage.
One day being bored not trying out anything new, I installed Xmonad for fun. It was a good laugh because of the simplicity... I could not write Haskell code. I found some interesting configurations online and began to learn Haskell. After learning some basics about the f
Lightweight... (Score:3)
This is relivant:
http://blog.martin-graesslin.c... [martin-graesslin.com]
I can't add much to Martin's sage words, but basically the term doesn't have much meaning in and of itself. Its the tech equivilent of stamping a "Natural" label on a box. What does that mean? Almost anything.
Re: (Score:2)
That's just doublespeak to cloud the issue of KDE not being lightweight... He mentions all the right factors himself: low memory, low cpu use, fast, not too many features. His complaint seems to be that there is no definition of a cutoff point from where something ceases to be lightweight. That's like saying you can't call an airplane fast transportation, because there is no fixed definition of 'fast'. However, most people who say an airplane is fast are comparing it to cars, walking, boats, etc. They are n
Re: (Score:2)
But KDE isn't a single program or even a set group of programs, its a community of people that develops those things in a highly customizable way.
So *my* "KDE" might not use much resources at all, while yours might consume more than CERN. It depends.
Re: (Score:2)
Both of our KDE's use much more resources than XFCE or LXDE. And it's worth it. But that's still the facts.
And if you want to use name KDE SC for the set of programs everyone has always called KDE, be my guest. I will keep using the term KDE, thank you very much.
Re: (Score:2)
Well, we'll just have to disagree.
I try to use the names authors give to their products, just like I try to use the names people would be preferred to be called, out of respect for them. If Bradly want's to be called Chelsea, well, KDE can be called KDE Plasma Workspaces, Platform, and Applications.
Re: (Score:1)
He mentions all the right factors himself: low memory, low cpu use, fast, not too many features.
But all those things are highly subjective and vague. How low is low enough CPU usage and memory usage? What CPU are you basing this off of? On my octo-core desktop "low cpu usage" is going to be "high cpu usage" on something less powerful. What is the baseline minimum RAM? Again, on my desktop with 24 GB of RAM percentage-wise "low memory use" will obviously differ. How are you objectively benchmarking "fast"? What is the objectively-defined cut off point of "too many features"? What you may think has too
NIH syndrome (Score:2, Interesting)
This is great and all, but I really don't see any reason to reinvent the wheel besides anal-retentiveness about licenses and Not Invented Here. KDE's libs (and the Qt they're using) are LGPL.
KDE license requirements (a partial list)
LGPL version 2.1 as listed in kdelibs/COPYING.LIB or later
LGPL version 2.1, or version 3 or later versions approved by the membership of KDE e.V.
BSD license as listed below.
Ensure that the BSD license does not contain the so called 'advertisement clause'.
Qt license for free soft
Screenshots? (Score:2)
Circle of Life (Score:2)
I have no doubt that it will start out that way. And then after a few years of development to achieve feature-parity with other window managers, it will likely become just as bloated, buggy, and sluggish as the rest of the window managers.
"BSD-licensed"? (Score:2)
The PC-BSD project is developing a new open source (BSD license) desktop environment from scratch. The name of the project is Lumina and it will be based around the Qt toolkit
OK, so they're developing a BSD-licensed desktop environment atop a GPL v3/LGPL v2.1-licensed toolkit [qt-project.org], right?
Re: (Score:2)
Re: (Score:2)
I thought that Qt is available under a BSD license as well
Nope. [qt-project.org]
I like Motif. (Score:1)
nt
KDE Lite? (Score:2)
Has anybody actually tried to take the KDE and trim the rarely used and niche functions?
Knowing the KDE people, if somebody found a usable subset of functionality, sufficient for a larger chunk of applications, they would be probably happy to merge it as an option.
Re: (Score:3)
Has anybody actually tried to take the KDE and trim the rarely used and niche functions?
Yep, in fact one such effort was started by some KDE devs:
https://blogs.kde.org/2013/04/... [kde.org]
KDE is more than a 'desktop' (Score:2)
Sure an end user only sees that part, but something like KDE is far more than that and there is a whole lot of stuff going on underneath.
Good. (Score:2)
I'm concerned that iXsystems and the community is biting off a bit more than they can chew - Canonical's having issues getting Unity out the door and, though I don't have either of their fin
Re: (Score:2)
Re: (Score:3)
My goal will be to make it heavyweight, unstable, slow-running, and compliant with nothing!
I cringe how much that sounds like Unity. ;)
Re: (Score:2)
Beta Sucks (Score:1)
You're right. This is why Linux has been such a flop in the embedded market, and Windows has retained the same interface and driver model for twenty years.
Re: (Score:2)
I learned today why no sane company uses linux
Given that this is a desktop environment for FreeBSD, atop which PC-BSD is based, not Linux (it might also work on Linux, but I presume that's not the primary goal), that's not relevant to this article (and also involves labeling a large number of significant and successful companies "insane", but that's probably your intent).
Re: (Score:2)
First of all, this story is about BSD, not linux.
Second of all, Linux *does* include drivers for new and old stuff. It does not generally speaking, include closed source drivers. So, if you are using open source drivers that are in the kernel today, you're A-Ok.
I've had several cheap motherboards that never worked with windows XP or above due to sound issues, but all of them worked perfectly in Ubuntu Linux. They were set up back in 2006, and are still working with the most current version of Ubuntu. That's