In a recent interview with the Austrian newssite derStandard.at, GNOME designer Jon McCann talks about GNOME OS, the consequences of Canonical leaving GNOME behind, the purported removal of features and the future role of Linux distributions. “I think there was a time when GNOME had kind of a crisis, we didn’t know where we wanted to go, we were lacking goals and vision – that was the end of the GNOME2 cycle. So we pulled together and formed a vision where we want to go – and actually did something about it. And now we have been marching on that plan for quite some time.”
One of the architects of the current vision that Gnome is following feels that everything is hunky dory…
A lot of other people are feeling the exact opposite. With a more pessimistic outlook every day.
I wonder if gnome would be a lot stronger if his defined “crisis” extended till today?
The Gnome team keeps talking about a plan. I’d like to know what it is because it must be really really good if they think Gnome OS will attain 20% market share by 2020 when they first release of Gnome OS is expected in 2014 at the earliest. At least that’s the goal they have. They also set a 10% market share goal for 2010. Went great.
At GAUDEC they talked how they are gonna build their own buildbot/QA. Do what KDE did with cmake, reuse existing technology that accomplishes what you want and avoid reinventing the wheel? Nope. NIH.
Gnome was made to fight KDE, as it used Qt which at the time was not FOSS. Their goal was a fully FOSS desktop. After Qt went GPL, then later LGPL, Gnome’s goal was gone.
At Gaudec they talked about some bad things about the Gnome project:
“The “bad things” for GNOME they pointed out was their focus on the traditional desktop…”
From: http://www.phoronix.com/scan.php?page=news_item&px=MTE0ODg
1) They have a focus on desktops?
2) It’s a bad thing?
Gnome developers have gone mad.
The 20 by 20 slide was precisely a joke on the 10 by 10 goal. I’m assuming you weren’t at the talk and is just spouting bullshit you read on the internet.
They most certainly are using existing tools for that project, although buildbot has absolutely nothing to do with cmake. You’re confused and spouting bullshit.
Geez, really? I wonder why people kept using GNOME then.
This was the opinion of a specific speaker. If you actually read the interview posted in this news you’ll find a different opinion.
Did you have all this bullshit copy & pasted to post here? Because it has absolutely nothing to do with the interview posted. I wonder if you even read the summary, let alone the article.
It was a poor joke.
The project completig their goal does not immediately turn the software into trash. They didn’t have a new driection for some time, the developers not having a direction about the future does not magically make the software bad in the present.
Gnome’s original goal was to have a fully FOSS desktop at a time when Qt was not FOSS. Qt eventually did go FOSS and Gnome had a stable, functional full free desktop as well. Their goal at that time was met. As evident by years of gnome 2.x releases being essentialy the same to an end user.
No it’s a pretty dominant opinion in the Gnome prject that touch is the future and we need to accomodate for touch in our UX even though there is not one device the runs Gnome3 and touch. Hell there are barely devices running a true GNU/Linux stack and touch!
Even then interview says how they want gnome on tablets.
Side pane features were removed from nautilus as “it doesn’t work well on touch”.
Removing.
A feature.
Quite a popular one infact.
Not because the code was poorly maintanied or buggy.
Simply because it was not suited for touch.
The people for whom it was intended seemed to enjoy it.
Like I said, Gnome is not recreating a build system. It’s using existing technology.
You could say jhbuild was created within the Gnome community and found its way to many other projects, as a lot of other Gnome technologies did. I absolutely disagree with your point, but maybe you’d help it by providing some actually relevant examples.
Qt changing its license in 2009 has absolutely no effect on Gnome. It’d be like old school Unix systems being released as open source today having an effect on Linux.
They absolutely want to work on tablets and other computers with touch interfaces. Work is being done in Xorg and GTK+ to make that at least work. As the interview says, the focus of Gnome right now is notebooks and workstations, as that’s what most of its contributors use.
Some people definitely want to see Gnome and the freedesktop on smartphones and tablets as well. That idea is still in its infancy and as you said it’s not even possible for Gnome to be there when the components it currently relies on aren’t ready. As such there’s no such case of features being removed or otherwise impacting Gnome specifically to work on touch screen devices.
What you have seen in Gnome is the influence from modern UI’s that differ from the old WIMP model. You may very well dislike that but it has nothing to do with input methods.
You’re confused again, or straight up making stuff up. The extra pane functionality in Nautilus has absolutely nothing to do with touch.
Edited 2012-08-21 05:08 UTC
When they put the slides online, maybe they could have but (just kidding) in tiny letters below 20×20?
My point was, historically, Gnome uses technology from within it’s own community, and rarely goes outside of that. KDE often relies on 3rd party technology in major ways. I was citing cmake as an example, but a more obvious one would be Qt. I would not be surprised if they did the same with their buildbots.
As for jhbuild, it was originally written for building gnome.
Qt is not old, dated and barely maintained. I prefer it to GTK. But, neverhteless Gnome’s Goal was really to create a fully free desktop, with Qt license changes that goal. Gnome kept going due to corporate involvement from Ximian/Novell, Red Hat, Sun ect.
With KDE and Qt being fully free, they needed a new goal. Just “creating great software” wasn’t gonna last forever.
I also want Gnome on touch, but I don’t think they should screw over the desktop because of it.
KDE has a better approach. Plasma Active, a UI just for touch devices. KDE has actual hardware in the works too.
Again, Gnome 3 doesn’t really look like they are focused on workstations. Some people may like it. To me it was OK to use and looked great, but everything took longer to do then in Gnome 2.
If they were focused on workstations, they have gone batshit crazy if they think workstations do not need the option to power off. I heard they are reimplementing that due to how many people complained.
Bullshit.
Windows:Check
Icons: Check
Menus:Check
Pointer: Check
We aren’t moving away from WIMP. Android, iOS, WP7, Windows 8, OSX Mountain lion are _all_ WIMP.
On Android/iOS there are still icons. The windows are really just fullscreened. The pointer is your finger. Menus are still there. On android I hold down my finger to bring up a menu for things like “open in a new tab”, copy selected text, paste. I don’t use iOS. With metro (or w/e it’s called) tiles are pretty much Icons, menus are still there, Windows are still there (again, fullscreen everything, WTF?). The pointer is either your mose or finger. It may not be the WIMP we are used to but it is still WIMP.
You’re right, I don’t like it.
Why are seeing all these new UI trends right when the time tablets and smartphones are hugely popular and on the rise? It’s because of touch. Windows 8 was intended to be run on tablets from the start. All these new UI’s have everything to do with touch.
You sure?
http://git.gnome.org/browse/nautilus/commit/?id=b8d5b4a7bcf47ed34a6…
“Extra Pane mode was somewhat useful before GNOME 3 had side by side window mode. The combination of panes and tabs is just too much. It is inconsistent with the file chooser and doesn’t work well with touch.”
Wow/ Side by side windows eh? So to view 2 directories at once I must use the entire screen, instead of being able to do it with less then half.
Panes and tabs! Too many things! I can’t choose between the 2, my head hurts!
C’om, panes let you view 2 directories at once. Can’t be done with tabs, you need 2 windows.
I don’t know how it’s inconsistent, but the user has to
know about the pane feature and choose to enable it from the menu. It wasn’t on by default. It was nicely hidden.
The only thing true is that it doesn’t work well on touch!
They removed tree sidebar for the same reason.
http://git.gnome.org/browse/nautilus/commit/?id=253d4bf50b04a2ad3a0…
“It is inconsistent with the file chooser, doesn’t work well with touch, is really hard to use, and isn’t consistent with any other GNOME 3 apps.”
Again, last I checked, you had to manually enable tree sidebar, it was a option in a menu that the user could choose.
Hard to use? You had to enable it, which implies you want it, and it wasn’t hard to use at all.
Consistency? Again, you would never know it existed until you turned it on. It wasn’t something people were complaining about all the time.
The only valid reason was that it does not work well on touch!
Edited 2012-08-21 15:33 UTC
No, the excuse is that they like to keep the applications simple, an extra pane with tabs is not simple, the task can be acomplished with two instances side by side, some people may not understand the change like you and is ok, but is not the end of the world eather, some people just like to extrapolate their own issues.
Edited 2012-08-21 18:33 UTC
First, not on by defualt. You had to ask the application to show panes or tabs. There isn’t even a new tab button.
Yes you can view 2 directroies by having 2 windows, but that is inefficent compared to panes.
The point is, according to the git commit was it does not work well on touch. I really don’t care about the pane feature. I am just citing this a evidence that gnome cares more about touch then traditional input methods.
No, you fail to put your self in the developers shoes, their philosophy is to keep the application simple, not only the UI, but also the code, a side pane is just good to move/copy files by dragging them, it may looks easy, but for a developer it means to write more code and mantain that code just for a task that can be already be reashed with another simple medium, in this case to instances of the application side by side, if you put in the balance the number of user that use an extra pane with the code that needs to be changed and mantainined just to reach goal that is already acomplished with the minimun effort then you will undertand the essense of that desition.
Now, for the touch friendly part, of course, if it is part of their goal to make it touch friendly and that means get rid of some other no so touch friendly options is ok as long there is still a way to do the task, and btw, GNOME Shell is not only touch friendly, it is also keyboard friendly so you can use your keyboard in the case you are in a desktop computer or a laptop, I use it daily and I use the kb a lot, specially the win key.
Edited 2012-08-21 21:27 UTC
Simple? Nautilus is >90kloc. of just C code and >140kloc altogether. I’ve written a simple file browser before, it does not take more than 10kloc of C to do so. Mind you, menus/dialogs took up most of that code and Nautilus has significantly more menus and dialogs than I used (along with it being scriptable, which is not a property that a “simple” file browser should have.)
nautilus-3.5.90]$ cloc .
defined(@array) is deprecated at /usr/bin/cloc line 1054.
(Maybe you should just omit the defined()?)
defined(@array) is deprecated at /usr/bin/cloc line 1056.
(Maybe you should just omit the defined()?)
defined(@array) is deprecated at /usr/bin/cloc line 1379.
(Maybe you should just omit the defined()?)
defined(@array) is deprecated at /usr/bin/cloc line 1510.
(Maybe you should just omit the defined()?)
550 text files.
540 unique files.
201 files ignored.
http://cloc.sourceforge.net v 1.56 T=4.0 s (88.5 files/s, 47395.5 lines/s)
———————————————————————- ———
Language files blank comment code
———————————————————————- ———
C 146 22471 10110 94719
Bourne Shell 11 3925 4039 25629
m4 8 1084 138 9648
C/C++ Header 146 2439 3871 6251
HTML 17 92 3 3360
make 15 148 31 903
XML 10 16 30 409
CSS 1 18 31 217
———————————————————————- ———
SUM: 354 30193 18253 141136
———————————————————————- ———
Simple is not the oposite to extensive.
I never argued that it is. Simple is usually not complex. Nautilus is complex. I showed that and gave an example of additional features that make it so (menus, dialogs, and scripting.) None of those features are necessary to the goal at hand (an easy to use file browser.) It is dishonest to call nautilus simple. This is mostly true of everything that gnome does (gedit for instance is ~55kloc, and most of its features exist in the form of the gtksourceview widget.)
Simple of use is what I’m talking about, simplicity of code can be arguable, becuase nautilus needs to access network shares, dected mime tipes, show previews, etc.
Nautilus is certainly easy to use as a file browser, but so is thunar, pcmanfm, or dolphin/konquerer. There’s no reason to use nautilus if we’re just using basic functionality (which as I said fits into something a magnitude of order smaller.)
So if we aren’t using nautilus because it’s simple, then why would you remove features in the name of simplicity?
If gnome wants a small + simple file browser then they should start over. This prevents users noticing loss of features (since they can continue using nautilus) and it ensures that the structure of the new browser is simple (and thus easily maintained.)
I don’t know man, certaintly the philosophy is to keept it simple, the question is what is simple and what is not, and that’s a question only GNOME designers can answer, some times the developers cannot understand the designers desition, communication can be improved but is not something that bothers me at the point to calling it a “disaster”.
Disclaimer: I am a GNOME lover, either GNOME 1, GNOME 2 or GNOME 3. Recently I’ve forced myself to use GNOME 3, first with lots of extensions, but then without any extensions. At work I have to use GNOME 2 due to Ubuntu’s previous LTS, but I hope I can upgrade soon to get back to GNOME 3 for most of my day. Having said this…
Oh no, don’t give us that. GNOME might not have had a “vision”, but it had a goal: being a damn fine desktop environment. It was the desktop of choice for most GNU/Linux users, it was well respected, it was well backed up by corporations and it was very good.
Yes, and you also lost the majority of your users and corporate support. Ubuntu didn’t came up with Unity out of thin air, it was a clear response to GNOME’s new “vision”. Ubuntu knew it couldn’t count on GNOME 3 because it was too radical: to an outsider it felt like developers were having “visions” while completely ignoring the bad feedback received from users. The transition from GNOME 2 to GNOME 3 wasn’t good at all.
And why do you think Fedora 18 will have the MATE desktop (a GNOME 2 fork) on the official repositories? Hint: Red Hat Enterprise Linux 7 will be based on Fedora 18.
GNOME 3 technology is vastly superior to GNOME 2 technology. There is a huge potential there, but that potential has to be exploited, otherwise only the abyss awaits. People need to know they have that kind of flexibility with GNOME 3.
I see a lot of talk about “vision” and “design”, but you must know your target audience. I’m glad you have tablets in mind, but you are forgetting about your real users, people who use the desktop right now. A prime example was the removal/hiding of the “power off” button. Not everyone has a laptop that goes to sleep well. Not everyone is interested in wasting power during the night. I, for example, shut down the power socket completely during the night to save power.
This is not a failure of GNOME 3 technology, it’s a failure of “vision” and “design”. Another prime example is usability. I’ve done usability tests, both as a user and as a developer. I know they’re hard. I hear GNOME 3 developers talking about usability all the time, but I doubt they’ve either done usability tests or look at the results properly.
It’s not that GNOME 3’s “vision” is completely wrong. I mean, I was a huge critic of GNOME 3, but I’ve learned to love it. But the “vision” has to have some connection with reality and with users, otherwise it will definitely fail and leave us all poorer.
Unity is part of Ubuntu’s vision to differentiate themselves from the rest of the Linux desktop. That split came before GNOME 3, and Canonical was never a major contributor to upstream GNOME in any case.
Also MATE is going to be in Fedora for the same reason the other 20+ window managers and toolkits are there: a group of independent developers wanted it there and packaged and then submitted it. Red Hat has their entire desktop team working on GNOME 3 and related technologies, using something else for RHEL makes absolutely no goddamn sense.
“The Linux desktop”? You mean like GNOME is there to differentiate themselves from “the Linux desktop” (may it be XFCE, KDE or $your_favorite_linux_desktop.
I have news for you: Unity is a linux desktop and I found it to be closer to traditional desktop concepts then GNOME Shell.
It came when development on GNOME 3 started. There where critics from the beginning the direction GNOME 3 was going to. Ubuntu’s try to sort that out failed, Unity was born. FLOSS on work.
They contributed a GNOME 2 centric distribution focused on the desktop (unlike RH), became popular and increased the GNOME 2 user-base significantly. New developers joined, documentation was written, translation and bugreports came in and so on.
Now only Red Hat is left and they have no focus on the desktop. You can turn around like crazy but it does not change the fact that GNOME lost one (or two if you add Mint, threr if you add debian, …) of its most valuable distributions.
MATE is a Fedora 18 FEATURE and is as auch in the same role, position the GNOME Shell is. No difference.
No, I don’t mean “like GNOME”. I mean like every other Xorg based desktop environment that’s made available in a form friendly for packaging on many distributions.
I was talking about contribution of code and development work. Ubuntu was certainly important as a software distribution of Gnome, thanks in large part to key Gnome developers choosing to work on it and then quitting over the years when the orders from Canonical were to move to a different direction.
There’s just one difference… I’ll let you figure that one out when you download Fedora-18-Desktop.iso a few months down the line.
I really do not feel like any project I do not contribute time or money to should explain anything to me. If some guys have some good/bad/crazy idea and commit resources to see it done then it’s fine by me.
That being said, I have a sense that some of those devs live in their own little worlds. I still have no idea if that is good or bad.
It will be interesting to see what RedHat will do with Gnome 3 on workstations. Two or three iterations from now, Gnome might be a very different beast if they have to target those environments.
I think some people took the 3.x stable versions too seriously just as they did with KDE 4 when it was launched. Gnome 2.x was perfectly fine. KDE 3 was fine when version 4.0 was released. I used my Ubuntu 10.04 install (+ppa for some more recent software pieces) until Canonical actually managed to make Unity stay out of my way (and not crash and burn every time I looked at it) in 12.04. RHEL releases come with the old Gnome 2 code base too.
People that expect something rock-solid, tested, super stable every 6 months are smoking pretty good stuff. To them I say “pass the dutchie”
The thing with kde4 was that initially it was serverly lackin, however as time passed each release became saner and saner (at the same time as making improvements).
With Gnome 3 each successive design decision seems more removed from the practicalities of a desktop environment.
As a developer of a FOSS project I try to always explain why I focus on certain bugs, or skip some feature suggestions, or what vision I have for the near future.
Users may not contribute time or money but I sure do love them a whole lot. They make it so much more fun to continue with the project and their feedback (good or bad) always make me smile. It shows me that someone out there is taking an interest in what I do and what I create.
I sure don’t need to explain myself to my users, but the they are my most valuable asset. I sure want them to know that I always listen to their input.
🙂
In the end it’s the devs that decide where it goes so let them take it where they want.
IMO the 2.x to 3.x series marks a transition from an average/business user releases to testing/experimental releases. I say this for the following reasons:
1. Lack of migration plan
This simply shows GNOME is not ready for anyone that wants a consistently usable desktop. When they stopped supporting the 2.x series there was no transition plan to get users to 3.x AT ALL. It was a drop them in the desert and let them find a way out. For mom, pop, most average computer users this is huge problem.
2. Release before ready
The 3.x versions released before EOLing 2.x were not even stable. Before the 3.2 it was really barely even usable and IIRC 2.x was EOL (no more updates) before 3.2. I am going from memory here so forgive me if this is not correct.
3. Untested UI design
It looks a lot prettier but there are more keystrokes and steps to doing most of the common tasks (launch commonly used apps, access commonly used folders and documents, get basic information you want all the time, etc, etc) resulting in a noticeable loss in efficiency. It does do some things better (search) but most other common tasks take longer.
4. Lack of feedback loops
I’m not sure where this fell down but it was clear users were clamoring with input of which almost all was ignored. This was like the whole spatial browsing thing but on a much more fundamental level.
This is not to say all 2.x releases were good. There were definitely very bumpy spots but this shift seems considerably deeper and substantial than any of the 2.x issues. I’m grateful to the devs for giving us the 2.x series but I would have appreciated a heads up on the change from user oriented to experimental releases. For those of us pitching Linux as a viable alternative on the desktop this blindsided many of us and it quite clearly fragmented the landscape considerably.
I am just glad there is XFCE to pick up the slack. Even so it is a decreased user experience from GNOME 2.x IMO.
I like tablets a lot, I have a iOs and and an Android tablet. I think Gnome 3 is a high example of human stupidity, at least until they explain me how in the hell they think to put it on a tablet. In a desktop computer Gnome 3 is fully useless.
In a desktop computer Gnome 3 is fully useless
Speak for your self dude, I use it daily in my work station and I love it.
So who modded that post down to -1? (I modded it up to 0 again). Was it “inaccurate”, “troll” or “off-topic”? I don’t like gnome 3 much myself but I don’t see why people should be modded down just for saying that they do.
Sinofsky will be saying the same thing in a few months.
The media isn’t reporting on the people who like it!
That’s because the story is that most people don’t.
Windows 8 and Gnome 3 are evil twins that should be buried alive. Throw Unity in the pit as well.
Where did you get this from? RHEL 7 is schedule to land on third quarter of 2013. Why would that be based on Fedora 18 and not 19.
Like I said before, between these two ‘tales’ from different GNOME developers (one optimistic and other really doomed) I am more acquainted with the doomed version. People can’t use the Shell for productivity work and I’ve know many people who turned it down.
Fedora is going down on linux usage ranks hand-in-hand with Ubuntu. CentOS has actually gone up quite high, despite the known ego wars within its community. Red Hat is taking notice. The Shell can and will hurt their testbed distro in its current form, so now it needs a revamp again.
The numbers are not lying. People are turning GNOME down. Being optimistic is just a way of ignore the real problems with the platform.
http://www.h-online.com/open/features/Red-Hat-s-RHEL-7-roadmap-1631…
“RHEL7 will be bases on Fedora 18”
Not forget that RHEL7 has to go through a long QA process.
The numbers are not lying.
What numbers?
Here are some numbers:
GNOME received about 41000 changes (only counting the master branches in the GNOME Git repository, work can also happen in other branches plus some teams also use external infrastructure) by approximately 1275 people for version 3.4, and about 38500 changes by approximately 1270 people for 3.2.
http://blogs.gnome.org/aklapper/2012/08/03/guadec-2012-discussions/
And how do that numbers prove your point?
Those number are related to Git commits.
I find it a terrible irony that just when Linux appeared to become usable, it spends several years shooting itself in the foot, all up and down the front line.
KDE 4, Gnome 3, Unity, and for me personally, the problems that have come with overheating and fan control on many laptops since kernel 3.0 and which, as far as I can tell, have not been wholly addressed.
Just to remind you all, I use Snow Leopard and Linux Mint at work, and Windows 7 and until this weekend Linux at home. I have no axes to grind, and I just like different aspects of different operating systems.
In brief, there’s nothing I think intrinsically wrong with Linux technologically but I have discerned an evolution in developers’ mindsets that has been pretty regressive in the main, harking back to the first days of trying out Linux. Not exactly ‘RTFM’ any longer but certainly ‘my way or the highway’. The reason why I hang in there with Mint at all isn’t obviously because it is fundamentally ‘genetically’ different from other distros technically speaking but it has a very progressive, open, varied approach to users. I am not surprised it has headed up the charts of late.
The laptop overheating thing is prolly temporary (my Toshiba suffers but my back-up Vaio of the same vintage, and my MacBook Pro do not). However, I am not sure about how fleeting the increasing insularity of developers is (not limited to Linux, of course, but it feels particularly egregious there because often people who go on about ‘open’ are frequently pretty ‘closed’).
So, in many eyes I will have made a fool of myself; yes, users are idiots. The challenge for developers is to address that for them, not tell them how to spell the word in different target languages, and what its etymology is.
Orf.
I for one would really like to see one fully integrated linux-based OS where everything is done to serve the user. Like Apple has managed with their OS X where the UI manages to hide the underlying unix so well that you have to open up a terminal to see its existance. Unfortunately I don’t believe that with open source model such a thing is acheivable – too many disagreements which makes it impossible to move towards single perfect goal. Hence the unimagiable number of distros out there, because man can think about the same issue million different ways.
Still such a thing would be doable but single team must be able to manage the whole OS. Everything from the kernel up to beautiful buttons on the screen. Only this way they can achieve the architecture where everything is in harmony and no workarounds/quickhacks are neccessary. Just like Apple and OS X but with Linux kernel and userland and perhaps GNOME UI.
Edited 2012-08-21 12:17 UTC
I would argue that for linux users, that is precisely the opposite of serving the user.
You have to look at your userbase before deciding what features are serving them. This is exactly what Gnome is NOT doing ( in fact they’re doing the exact opposite).
Edited 2012-08-21 13:15 UTC
It might be the opposite of serving the current users, but if any Linux distribution expects to break through on to a large portion of the average user’s desktop, they’ll need to do something like an integrated OS X design. News flash: Most users these days don’t like to see plain text. It’s sad, but it’s true. For example, when a kernel update breaks X.org (which needs to be flushed completely before Linux on the desktop has any great chance), dropping the user to a text login prompt isn’t acceptable. In fact, updates breaking the system is not acceptable in the first place, and I think we’ll only solve such problems with a tightly integrated os designed by a dedicated, paid, team. As it stands now, even among one distribution, there are too many people pulling in too many directions not to mention upstream going off on their own, and the system feels more like a slap dash attempt at a cohesive environment than a true operating system.
Currently perhaps one of the few distros complying with above requirements is Slackware as it is basically one man show (actually a very small team of dedicated and knowledgeable people) and the new version will come out only if thoroughly tested and retested. Slackware has usually about 1.5 years release cycle and packages are upgraded with extreme care. If new version breaks something and old version still is OK, then package is not upgraded for the stable branch. It has always been a very solid distro. However the philosophy requires quite a lot of hands-on tinkering and certain attitude so it is not a distro for the masses or even for average linux user. The quality is there however the way it should be everywhere.
Edited 2012-08-21 14:35 UTC
Integration is not only quality packages, it is mor about the specific custom tools for that particular system. Users would really appreciate if there are GUI tools for changing all the aspects of the system, which are able to change specific lines of the specific config files in specific locations. This would be the total enemy of the universal software but it is the only way to win the users.
So the purpose built software for a single integrated linux distro is the key. Thats what I believe GNOME OS is about.
The situation is apparantly not as fluffy as the interviewee makes it out to be.
Especially in the nautilus case the actual author of the extra pane gets constantly ignored.
His questions to specify which code is a mess remain unanswered.
His enquiries how to realise certain user scenarios with two nautilus windows next to each other get no attention …
All this is documented here [1].
Also the links there are quite interesting.
In my opinion this is a disaster. How do you want to attract new developers and keep the old ones if you treat them and their work that way.
Instead of improving something that might not be ideal it gets removed. If this is really the way to go, then well, then gnome would need removal too …
[1] http://berndth.blogspot.de/2012/08/nautilus-extra-pane-removal-myth…
That is the answer from one guy to the removal of the extra pane, not the whole community.
Problem description from the GNOME foundation, no solution yet:
https://mail.gnome.org/archives/desktop-devel-list/2011-June/msg0001…
Failed try to establish rules for the problem from the GNOME father that made GNOME happen together with Miguel:
https://mail.gnome.org/archives/desktop-devel-list/2012-April/msg001…
Consequences:
http://blogs.gnome.org/otte/2012/07/27/staring-into-the-abys
The situation is far beyond a one guy problem, all the links above are from before his blog and there are many more voices from core developers, and there is still no solution on the horizont.
Edited 2012-08-22 06:57 UTC
AFAICT the answer from the community is that they want the extra pane.
If the developers stated grievances (ignoring requests for comments, non-communication). Then pattern of behaviour leads to what others call “the abyss”.
If somebody contributes a feature but requests comments on how to change,fix or improve said feature to fix the problems stated as the reasons for removal, then that isn’t going to inspire others to “waste” time developing for gnome.
There are numerous other examples where people provide reasoned (and correct) examples why the alteration/removal of a feature is a bad idea, they simply get ignored or shouted down as being unproductive.
They can fork the code and mantained them self, what is the problem?
The problem is the way of communicating things.
In the link I posted it is mentioned that it took quite a while with a lot of discussions, reviews etc. before the feature was taken in.
And I think this is a sensible way for introducing new stuff.
And now it is taken out without even discussing the issue with its author and even worse ignoring the author’s and the other’s questions, remarks etc.
“We hate it” is for sure the best attitute to improve something…
One of the links say:
> Design work requires conversation, closeness, and contact. You need to
> build a relationship with the problem, the users, the stakeholders,
> the developers, and ultimately with the product. These relationships
> must be durable.
I *absolutely* agree, and this is why I am concerned with the current
situation.
Saying that they don’t care has no base at all, since the contrary is expressed in the same links you provided and also goes beyond explained the flaws of real time communication:
Let me be clear: real-time communication has a role. But it fails in
three key points:
* There is no record of the culture for newcomers and other members of
the community to consult (and no, I don’t think logging IRC is
sufficient – can you imagine suggesting that someone read the IRC logs
for a month to get a feel for why certain decisions were made?)
* There is no clear way for a newcomer to contact the team, or to know
who the individuals in the team are (related to the lack of a record)
* If you’re not in the room, you completely miss out on any opportunity
to influence the conversation
So, don’t come here to tell me they don’t care, one thing is not to agree and other not to care.
Now, about the “Staring into the abysm” article, that one is aready debunked with the topic of this thread and with the number of attendes in the lates GUADEC.
Edited 2012-08-22 19:53 UTC
I fail to see what your reply has to do with my post.
The person who’s feature got taken out is not a newcomer to gnome at least to his own account — I am not familiar with the gnome devs.
I also did not say that none cares.
Obviously the person who wrote the blog entry I linked to cared, as well those who responded to his enquire. Though as far as I understood none involved in the decision to take the feature out answered.
Yet at least one of these persons still spreads in the public that the code was bad without any prove.
How does that fit together?
Edited 2012-08-22 20:02 UTC
That’s definitively understandable, I would be mad too.
It should be noted that the individual that made those points isn’t one of the developers heavily involved with the “new gnome vision”. He wrote that because the developers who *are* make decisions unilaterally, leave no records and don’t accurately justify their reasoning to the wider community…