Personally I found it a bit snarky. By the same token I find that UI wars are the most vitriolic, after all they are the way you talk to the machine day in and day out. Screwing with that, screws with everything.
I have found Ubuntu's strategy interesting because it seemed clear that while KDE was following general guidelines around Microsoft OSes to be more accessible, and Gnome was following general guidelines around Apple OSes for similar reasons, Canonical sort of 'turned left' and drove off the road to a new place.
I remember distinctly when I left Sun and had to give up my Suntools interface for what became Windows98 at the time. And it was hokey and painful and it crashed a lot, except that over the weeks and months it crashed less and less, all without a software update :-). And I realized it was not so subtley training me not to use features that failed. Of course if you use something long enough you become reasonably facile with it. When I switched my desktop to Linux I was always more comfortable with KDE for that reason, the whole 'start' menu on the lower left, the control panel abstraction, the way things laid out on the screen.
When I went to Google I got a Macbook as my laptop choice, it was different, and I struggled at first, but once I became reasonably good at navigating around I found that I was also less annoyed with Gnome.
I think the Unity strategy at Canonical will pay them big dividends. Mostly because the Linux desktop market has been such a small part of the whole desktop market as to barely merit a full pixel width in a pie chart of desktop OSes. I believe that part of the reason for that is that the strategy of being 'kinda like' MacOS or Windows in the GUI has failed Linux badly when it comes to non-technical users. It failed them because there was neither the cohesion of implementation, nor the quality of testing, in either KDE or Gnome which would ever cause a non-technical user to think the GUI was 'better' than the one they left behind. Unity breaks that cycle because it doesn't work like the GUI you used to use. and so I think users cut it some slack, they realize they are in a 'new' place and learn how to do the things that they want to do in the way that this gui does them. And there isn't a mental comparison to their previous gui because it wasn't like this at all.
Assuming, and its a big assumption, that Canonical can execute on the Unity strategy well, it will continue to be the dominant Linux distro. Further it will increasingly leave behind every other distro, because while others may trade off market share amongst the technical users, where programmers slosh from one to the next, Unity will be gaining non-technical users who won't go anywhere else in the Linux space. Ever.
Great comment. But allow me to disagree on comparing Gnome 2 to OSX. In my opinion old Gnome was exactly what you're talking about: "driving off the road". It wasn't like Windows or OSX, it was its own thing, the best desktop environment, in my opinion. The only element reminiscent of OSX (only visually, not functionally) perhaps was the menu at the top.
Unity, on the other hand, is the exact replica of Gnome 3, just not as polished graphically. They both have this weird "buaaa mode" where you're typing instead of clicking on a shortcut, they both destroy virtual desktop functionality and if anything, they're basically cloning OSX bad habits: instead of switching between windows on Alt+Tab they're now switching between apps (across virtual desktops!).
So I would suggest that the opposite is happening. Instead of staying in the "new place" Ubuntu is basically trying to get on the road and follow OSX.
Frankly, I'm not even sure it's possible to "drive off teh road to a new place" when it comes to desktop UX without a significant change in hardware: we're still using basically the same computers as we did in late 80s, even the screen real estate hasn't improved much.
> Frankly, I'm not even sure it's possible to "drive off teh road to a new place" when it comes to desktop UX
Sure it is.
3d file managers[1] never took off. With todays hardware they'd be a lot nicer to use.
Users have been limited to quite small (sorry about my incorrect terminology here, I'm going to go read some man pages) total virtual screen size within which their windows were located. Now graphics cards, and system memory, are huge; this leads to possible radical new interfaces.
Imagine a "Zoomable UI" - documents open in their own full size window; any document opens in its own window tiled next to it; the user can zoom in as far as they like to do detail work, or out as far as they like to organise all these documents. Programs would open toolbars in their own windows. Any toolbar would be able to work on any document (because it's Unix, so all input and output is text, right?) - but the result wouldn't necessarily make any sense.
That's not particularly radical; tiling WMs exist already.
3D file management never took off because no one has yet found a meaningful and intuitive use for three dimensions that couldn't be accomplished with two.
The screenshots of FSV on that site show me a 2D interface that's been extruded into cubes and displayed isometrically. Usable, but the 3D is superfluous. All the other 3D file managers appear utterly unusable.
To me, the clearest mandate for using three dimensions for file management is in creating a space that people can become as mentally familiar with as they can with a house.
We already know that one of the most powerful methods for creating a lasting memory of intangible items, the "memory palace", is basically walking around a three-dimensional space.
It seems clear that there is a possibility here to create a fully-functional three-dimensional metaphor. However, it is not an evolutionary step. It would be something so foreign that it would be better to teach it to people who never managed files before. And to be honest, with apologies to Dropbox, the numbers of such folks is back on the rise because file management itself is an increasingly irrelevant task, not just on iOS but on Chromebooks and just about any non-desktop computing device.
And so although there is a very interesting potential, I have to wonder if the quest for better file management is even worth fretting over anymore. Hasn't being "ready for the desktop" largely been an exercise in skating to where the puck has been? Maybe a revolutionary file manager would have been strategic ten years ago, but now?
So I welcome Canonical's focus on touch devices. Yet I'm concerned about the sheer inertia of maintaining an identity, in being Ubuntu, on a significantly different platform. We saw how well the Windows Tablet PC succeeded at being Windows, and in so doing lost the platform. I sure hope that's not what happens here.
But a file manager's principal function is to organize things. And people don't think three-dimensionally when they're organizing things. They don't even think two-dimensionally. It's always a single dimension, some kind of vector, and they compose things into one-dimensional vectors that may contain other one-dimensional vectors, but they're always looking for one dimension at a time.
If you're looking for a quotation from your favorite author (assuming Google wasn't helpful), you're not going to nevigate a three-dimensional conceptual space to find it. You're going to first iterate through your list of bookshelves, and find the appropriate shelf (assuming you have lots of books and sort them into shelves e.g. by topic or alphabetically). Then you'll sort through the books on the shelf and find the book you need. Then, you'll go through the chapters of the book, etc.
The mind organizes information into categories, and groups categories within categories. Visual mnemonics are great, but they help us to find the specific item we're looking for at the appropriate level of abstraction, and they work just as well in organized, two-dimensional spaces as well as three.
I agree that there may be a lot of potential for some major breakthrough, but having to remember that your Economics paper is stored in an inventory slot in a chest in the inn in that logging town outside Stormwind isn't it.
Yes, 3d file managers were ugly, and hard to use, and didn't really serve any purpose.
But with modern hardware, and with a better HID, and with a better analogy / paradigm / idea, they could come back and be useful.
I don't know what that meaningful use for the 3rd dimension would be, but I'm pretty sure there is one.
(The tactile screenshot is awful. It appears to show a bunch of stuff ordered by name. The TDFSB2 screenshot shows someone "walking" through their home/leo/jpeg directory, with all the jpegs being displayed. Remember that these are tinkering toys, and that things like mice or GUIs were initially weird and horrid. Maybe getting a bunch of smart people working on them would generate useful results?)
Exactly - someone will have to design a new file management functionality that adds real value and also happens to leverage 3D technology.
Window management certainly has benefited from using the GPU, e.g. with real-time thumbnails of applications and enhanced virtual desktops. But these are solutions that were enabled by the GPU, not features that were tacked on simply because GPUs were around to use.
If someone comes up with a new and revolutionary file management paradigm, or even just a useful innovation on the margin of existing file management, and it's one that 3D makes sense for, then great. Until that happens, it's a solution in search of a problem.
I agree, its way easy to drive off the road. Not as easy to navigate the resulting underbrush.
Aza Raskin's THE stuff is pretty interesting in that regard, there are varieties of menu systems, where do menus appear, how do they appear? (I've always thought the pie menus were cool) And of course how do you switch between manipulating the space (app selection, desktop, themes) and manipulation within the app itself.
Strangely (for me at least) is that the Windows 8 previews have some interesting stuff in them (after years of not changing much) more interesting in concept than some of mainstream UIs on Linux. Again not that a tiling WM is particularly new but the blending of the icon/status bar presence/window/launcher feels pretty fresh.
I do run Kubuntu on my desktop but I continually try out different schemes with virtual machines when I can to see where things are going. My netbook runs XFCE which reminds me in a strange way of Suntools.
>they're basically cloning OSX bad habits: instead of switching between windows on Alt+Tab they're now switching between apps (across virtual desktops!).
This is THE reason why I will not touch gnome-shell/unity. Too bad, because I could get used to most other stuff (except unity's menu bar on top) and I believe it has potential. Multiple displays and virtual deskops are a minor annoyance that can be improved.
The GNOME/KDE development model is pretty strange. They release a new major version that changes everything so they spend years fixing and stabilizing, and then they release another major version and repeat the process.
XFCE does a better job. "XFCE4" has been around for 7 years and they provide small improvements but don't make any radical changes.
Unity was pushed out before it was ready. I understand the need to get people using it but it's still not a passable desktop for many people and probably won't be for another couple of Ubuntu release cycles. Hopefully once it stabilizes they won't repeat the mistakes of GNOME/KDE and decide to change everything again.
It's a common scenario in the open source world. Projects get bogged down in personality or leadership issues, and it's a lot more fun to write something new rather than maintain some other guy's code.
It's funny that problems that have bedeviled the Linux Desktop haven't really changed over the last 10 years. Poor release QA, "almost there" hardware support, painful transitions from X to Y. Nothing ever really stabilizes.
They release a new major version that changes everything so they spend years fixing and stabilizing, and then they release another major version and repeat the process.
This is inevitable when you either set your sights too low (and after a while realize that your current architecture will never get you where you need to be) or develop too slowly (so by the time something is finished the market has moved on).
Personally, I think it has more to do with the fact that very few people want to fix old bugs in an old codebase in their free time when they could be writing a brand new version with brand new bugs. "This time, we'll make better bugs!"
Unity was first released in Ubuntu 10.10 Netbook Edition. And that was based on Ubuntu Netbook Remix, first released in 8.04 (they didn't necessarily share code, but the design was similar).
You would think that by 11.10, released a month ago, they would have something release-quality. I think it's just the design that people don't like. Canonical has the resources to fix bugs, but they don't necessarily have the desire to fix design defects.
It is hard for a dev to say a crash is a feature. It is easy for a designer to say an unusable UI is an intentional exploration of a new paradigm for modality in transhuman existence.
While I agree that Unity is the right step for Canonical to take with regards to Ubuntu, my main gripe is that they've taken away the old GNOME UI and has left us with no (easy) way to bring it back. This was my experience as of 11.04.
The old Gnome UI was about 2 clicks away on the login screen in 11.04 (unless you didn't have a password on your account, for some reason).
In 11.10, you can manually install gnome, and then you get gnome 3, which is certainly not what you wanted (and not necessarily an improvement over unity).
I have found Ubuntu's strategy interesting because it seemed clear that while KDE was following general guidelines around Microsoft OSes to be more accessible, and Gnome was following general guidelines around Apple OSes for similar reasons, Canonical sort of 'turned left' and drove off the road to a new place.
I remember distinctly when I left Sun and had to give up my Suntools interface for what became Windows98 at the time. And it was hokey and painful and it crashed a lot, except that over the weeks and months it crashed less and less, all without a software update :-). And I realized it was not so subtley training me not to use features that failed. Of course if you use something long enough you become reasonably facile with it. When I switched my desktop to Linux I was always more comfortable with KDE for that reason, the whole 'start' menu on the lower left, the control panel abstraction, the way things laid out on the screen.
When I went to Google I got a Macbook as my laptop choice, it was different, and I struggled at first, but once I became reasonably good at navigating around I found that I was also less annoyed with Gnome.
I think the Unity strategy at Canonical will pay them big dividends. Mostly because the Linux desktop market has been such a small part of the whole desktop market as to barely merit a full pixel width in a pie chart of desktop OSes. I believe that part of the reason for that is that the strategy of being 'kinda like' MacOS or Windows in the GUI has failed Linux badly when it comes to non-technical users. It failed them because there was neither the cohesion of implementation, nor the quality of testing, in either KDE or Gnome which would ever cause a non-technical user to think the GUI was 'better' than the one they left behind. Unity breaks that cycle because it doesn't work like the GUI you used to use. and so I think users cut it some slack, they realize they are in a 'new' place and learn how to do the things that they want to do in the way that this gui does them. And there isn't a mental comparison to their previous gui because it wasn't like this at all.
Assuming, and its a big assumption, that Canonical can execute on the Unity strategy well, it will continue to be the dominant Linux distro. Further it will increasingly leave behind every other distro, because while others may trade off market share amongst the technical users, where programmers slosh from one to the next, Unity will be gaining non-technical users who won't go anywhere else in the Linux space. Ever.