"Desktop" is too broad of a definition. Gnome/Nautilus already kick Finder's ass, but is it enough to recommend Ubuntu over OSX to my parents? No.
What Linux needs, is PC manufacturer's support. I am not talking about drivers here, I'm talking about quality pre-installs done at the factory. Once suppliers see this, they'll provide quality Linux drivers in order to compete. And for users stuff "will just work".
ThinkPad T61 running Ubuntu is in many ways superior to MBP/OSX (and 30% cheaper), but only because I deliberately configured it with only Linux-compatible components and spent a weekend tuning it upon arrival.
>What Linux needs, is PC manufacturer's support. I am not talking about drivers here, I'm talking about quality pre-installs done at the factory.
That would be wonderful. Ubuntu is making my laptop search very stressful. It is so much fun to use that I don't want to go back to Windows. However, if I buy a laptop that isn't compatible with Ubuntu then it will be hard or impossible to switch out the components for compatible ones like I did on my desktop. The only hassle-free solution is to buy Ubuntu laptops from Dell, but they currently charge about $150 more for the same hardware with Ubuntu installed compared to Windows (what happened to the "Windows tax"?).
Dell charges $1-2 per crapware install, but sometimes those are free if Dell feels like they add value. Vista Basic, OEM edition, is probably $50 (I am guessing about this one). This requires 20-25 cases of infection with useless shit to offset Microsoft tax.
This still can't explain how come Ubuntu-based machines are more expensive.
I think that Dell simply rides the demand train: they're the only big-name brand that sells Ubuntu-powered machines (very few know that Lenovo sells SUSE) and they're enjoying higher profit margins on them.
Could you link to such a Dell configuration? The last "comparison" I saw compared completely different notebook types. Are there notebooks where you can literally choose Win XP or Ubuntu and the latter costs 150$ more?
No, had to get the specs as close as possible. If you look on the Ubuntu notebook page, you'll find something with a name like "Dell 1800n". Then if you look at the normal page, you'll find a "Dell 1800". You can get the specs close, but not exactly alike. It was clear that they were offering more bang for the buck on the Windows side.
With laptops, make sure you buy one with a factory install of Ubuntu or at least another Linux. There's enough out there (at least in the US and UK) to vote with your wallet and make compatibility someone else's problem.
The point that most Linux users and developers seem to miss is not that a good user interface needs to be pretty -- it needs to be consistent. Sure, the stylish looks of OS X do a lot to woo people to the platform, but the experience is why people stay.
Consistency is the area where Linux fails miserably. Perhaps this is due to the lack of a single visionary who leads the way in Linux desktop development, but I'd say the problem lies more in the 'not built here' mentality of the developers. Which is kind of ironic, because the entire point of free software was that anybody could contribute to improving it.
Consider this: Think of your top ten Linux applications, and the GUI toolkits they use to handle their interaction with X11. I'll wager that, out of those ten applications, you can probably list four or five different GUI toolkits, each of which works slightly differently. Why isn't there just one really good GUI toolkit?
I think the answer lies in pride -- everybody wants to have 'their' project, so they start their own, rather than contributing to an existing codebase. For the same reason, open-source project maintainers often do an awful job of listening to their users and co-developers, because their pride of being at the head of the train is threatened by other people making key decisions.
This pride also extends out into assuming that the user of said application wants to be an expert in its use, which leads to overly complicated preference panes, nonsense confirmation dialogues, and so on.
This is why Linux applications often have the most appallingly inconsistent interfaces, and sadly, I can't think of a good solution to the problem. Frankly, I wish I could, because I would love to run Debian Linux on a Lenovo tablet, and have a GUI that is as useful for me as OS X.
You're overstating things. Apple itself has a solid handful of different toolkits and UI paradigms, some of which are better than others and many of which are legacy.
Take a look at an actual linux distro, and you'll find it uses only one desktop and is quite consistent. Slamming Ubuntu because Mandriva exists (etc...) isn't really a meaningful argument. Hell, most of that "linux" GUI software runs great onOSXtoo. So by your own logic the Mac is "failing miserably" at consistency and has the "most appalllingly inconsistent interfaces" (more so even then Linux, which after all can't run Cocoa apps).
Basically, this kind of broad-brush argument is mostly just flaming. If you have an actual complaint about an actual product (and I'm sure you do -- I have quite a few of my own), let's hear it.
Well, my nominal work is as a system administrator, and I've been working with Linux since about 1998. In fact, until 2005, Linux was my primary computing platform. So, I would claim to have tried at least one or two Linux distributions over the past ten years. There has been a very marked improvement in the user interface over that time, and of them all, I would say that Ubuntu is probably the best of the lot.
My point still stands.
Firefox is built on GTK, whereas KDE is built on Qt. Audacity uses wxWindows, and xterm is just raw xlib, if I remember correctly. OpenOffice has its own widget toolkit, as does VLC, and since I was a WindowMaker user, that's GNUStep, which is yet another widget toolkit.
All of these toolkits behave differently, and even though the situation has improved, they still don't interact terribly well. For example, cut-and-paste was, and still is as of the last time I tried Ubuntu, still not terribly functional. Sure, it worked with the X clipboard, but the behavior was subtly different between applications. This is something that even Windows has had working well for nearly fifteen years.
This is a very real problem.
International support isn't terribly good across the board, either. The situation has improved a lot in the past year, with Ubuntu and Debian now doing a decent enough job out of the box in terms of Japanese input and translation, but still nowhere near as good as it works on my Mac desktop.
Speaking of which, I can not think of a single piece of open-source cross-platform GUI software that I use at all. Of course, I use a ton of stuff on the command line, but not a single application from my Linux/X Windows days has made it on to my Mac.
Mostly because they don't use native widgets, and so even if they look like a Mac application, they behave differently in subtle yet highly annoying ways. For example, in any OS X application, I can highlight a word, hit a key, and then have that word fired over to the Dictionary. I use this regularly, yet even the latest build of Firefox doesn't support it. It's a built-in feature of the OS, and works just fine in any application that uses Carbon or Cocoa, yet not in Firefox, because they've bundled their own widgets with the application.
And this is my point. Even when Linux apps look like they're all part of the same platform, through appropriate use of themes and skins, they still don't feel as if they are integrated together.
OS X has only one set of widgets, although there are two primary APIs for developing applications -- Cocoa and Carbon. Cocoa is for applications written in Objective C, whereas Carbon is for C, C++, and Java. Carbon mostly exists to make porting applications easier, and to provide Java applications with access to native GUI widgets.
But, no matter which API a programmer choses to use, the application will behave like all other Mac applications. Copy and paste will work the same as it does everywhere else. So will all of the neat tricks under the 'Services' menu.
So, there's my complaint. GUI Applications on my Mac all act the same. GUI Applications on a Linux box don't, and coming close usually requires quite a bit of fighting, even for something as simple as entering text my second language.
Linux (and FreeBSD) are great server environments. I would be hard-pressed to even consider using OS X for anything other than a file and print server. But I won't use it as a desktop platform anymore, because it's simply too much of a pain in the ass, too much of the time.
I'm still waiting for an actual complaint about an actual linux product that shows an inconsistent UI. KDE distros ship Konqueror on the Desktop, not Firefox, for example. Open Office is actually very gnome-centric, even if the widgets themselves are implemented with different code.
Your point seems to be that you can assemble an inconsistent, frankenstein environment using "linux" stuff. Well, duh. But you can do the same thing with OS X or Windows. The stuff that ships as a product from actual vendors (be they Apple, Red Hat, Novell, Canonical...) don't have the kinds of problems you're positing, and haven't for quite a few years now.
This is exactly my point. Each individual piece of software, each product, nominally does work very well, and I'll even go as far as to say that they are highly consistent internally. But there is no consistency between even the most common set of applications that the average user is going to need to work with.
I'm not saying that there isn't good software. I'm saying that it doesn't do a good job of playing well together, and that's what it will take for Linux to be a better desktop platform than what Apple currently provides.
And, no, I'm sorry, but those problems do still exist, and are quite prominent.
Consistently functional copy and paste as an Ubuntu improvement has nearly 2500 votes across two requests, which would be enough to put it in their 'Most Popular Improvement Requests, Ever' section if they combined both of the threads: http://brainstorm.ubuntu.com/idea/3118/ , http://brainstorm.ubuntu.com/idea/4242/
A new user writes that, while they can use the clipboard in most applications, they have no idea how to do it in an xterm (http://ubuntuforums.org/showthread.php?t=772970). Why does it work differently there than it does in Firefox?
Japanese input is such a mess that there's an entire section of the Ubuntu site devoted to it (https://help.ubuntu.com/community/JapaneseInput). Why do I need to choose between two input methods? Why doesn't the input work across all applications? Even Windows, which despite Microsoft's largeness in the world has surprisingly bad internationalization, and which I do not like, at least handles this correctly.
One way to start to fix this problem is with marketing. Linux is just the kernel; GNOME is the OS. Ubuntu and Fedora are GNOME distros. You shouldn't expect to run KDE or Motif programs on GNOME; after all, they were written for a different operating system. Of course, this divides the former "Linux" market into even smaller pieces...
But this is the exact problem. Why should each window manager be its own platform? This is like having every car produced by GM or Toyota being totally custom-built, with parts that can't be interchanged, without great pain, with anything else in the range.
I'm just talking about the current state. It is a fact that interop between GNOME, KDE, and whatever else is so poor that the only way to achieve usability today is to pick one platform and stick with it completely. In some cases the differences between the platforms are arbitrary and they could interop or be merged, but in some cases the differences are philosophical and IMO irreconcilable (you might as well merge the Democratic and Republican parties). Given that situation, I don't see any way to create a single Linux desktop market. But then you probably shouldn't listen to me; I don't even use Linux on the desktop.
That is a situation where survival of the fittest eventually sets in. Eventually one will finally win just enough users to start eclipsing the others. My guess it will be gnome since it is the WM of choice for Ubuntu and Red Hat. And as more corporations get involved with development in the Linux-realm, there will be more incentive to standardize things and less incentive to fork.
Each window manager is not its own platform. These are platforms in their own right, a platform defined by the API, the theming, the standards, the libraries; the window manager is merely one part, out of many, that compose these platforms.
Kubuntu, Xubuntu, Fedora, OpenSuSE, SLES, RHEL, Yggdrasil etc. are a different OS, aimed at different people, with different interfaces, default programs, updates, and methods of doing things.
I don't like apple that much. I like the fact that they are good. Microsoft spoiled us, we thought that just because its open source its automatically better than anything else. Apple showed us that open source is just automatically better than windows, not better than anything else. Its good to finally have good competition.
>"The challenge for us is to figure out how to deliver something which is crisp and clean," without sacrificing the community process, he said.
There are multiple architectures for open source projects that run the spectrum from anarchic to authoritarian. If a founder has a particular artistic vision that he feels passionate about, nothing is stopping him from forcing it through and then releasing the source code.
Personally, I love using my Ubuntu desktop. The team has done a great job on meeting the usability challenge. The days and weeks spent troubleshooting hardware problems were ugly, horrible, and frustrating (I would love to have an Ubuntu "Installs for sure" hardware list somewhere), but the Ubuntu user experience is a delight compared to Windows.
A caveat: I've never used a mac for a significant period of time so I can't say how it compares to that competition.
If a founder has a particular artistic vision that he feels passionate about, nothing is stopping him from forcing it through and then releasing the source code.
Except the fact that he'll receive virtually no contributions from the community and thus he'll be limited to whatever can be done by himself and his employees. Also, community-controlled distros have set themselves up as gatekeepers, so most users never even discover software that the distros don't like.
I think Shuttleworth has identified the correct problem: We know how to get a large number of programmers to produce high-quality software (employ them) and we know how to get a large number of programmers to produce a large amount of software for free (the bazaar), but we don't know how to convince a large number of programmers to produce high-quality software for free.
I've been using Ubuntu for over a year now first on a T60p laptop, now on a custom built desktop. The only compatibility issue I've had is the lack of ati support for Linux, other than that smooth sailing. I use linux simply because of how powerful it is, it doesn't stand in your way at all. As far as usability goes I'm a poor judge, everything seems logically placed to me, although it lacks the unity of MacOs and even to some extent of Windows. The open source community seems to finally be concentrating on elegant design, so this might change soon.
Why does open source work better for code than for design (or does it)? I thought the reason for the good quality of many open source projects is that people fix the flaws for themselves, and all the fixes accumulate.
Why does it not work the same for design (or does it). Are designers less prone to fixing flaws in products? Are they just lazy and use Apple stuff?
Personally I can't really see a fundamental difference between an ugly widget (looks) and an ugly piece of code, they are both bugs?
Until you can bring in a visionary dictator like Jobs that everyone channels their energy through, it is impossible to bring Linux to the level of Apple.
"The challenge for us is to figure out how to deliver something which is crisp and clean."
Crisp and clean? That's all it takes to match Apple?
A great movie requires a visionary director. Shuttleworth has does some amazing stuff with Ubuntu, but I think attacking directly at the bow of Apple is a misguided battle.
Shuttleworth IS the Steve Jobs of Linux. Linus is the poster child for the technical guy, but Shuttleworth needs to lead the style. However I find him lacking in specific vision, he says these things are problems, we need to find solutions, but Ubuntu never provides a revolution. I feel like Ubuntu has found the annoyances of Linux and fixed many of them, but they aren't necessarily visionary changes.
Because Linux's problem isn't its bugs, or its glitchiness. Its problems revolve around the design theory that builds it.
That's where Apple wins big-time over the other two. It has a consistent theory of how to reduce clicks to get effects. From not having a lower bar to HAVING an upper menu to Spotlight to its Dock, everything is designed to make things faster for you.
Ubuntu won't ever have this. It can't. Nothing that's designed to be truly open can. You can't tweak things to the same level of control. You can't implement multitouch and make programs that RELY on multitouch. You can't put iTunes controls on the keyboard, or an Eject key. You can't make iChat integrate with Mail and your Address Book when not everybody will have each of those programs. It's not possible.
I've used both thoroughly. The level of polish on the one side is VASTLY superior. Look even at Adium versus Pidgin. They run the same library. But one takes up far less space, has better-designed themes, runs more smoothly... because Apple allows for that. It offers big screen resolutions and superior font support, so you can keep things tiny and out-of-the-way and still readable. It deals with a modern-day theme that means a focus on professionalism still works. It knows it can use Growl and the Dock for notifications, which keeps it much less in-your-face than anything FLASHING ever would.
You can't get keyboard shortcuts that are as good when you don't know the keyboard. You can't get software support when you don't know the hardware. You can't have a superior user interface when people use five hundred different displays and two thousand different methods of input. (Exaggeration.) Ubuntu can not win a battle when it can't control the tools used to fight. It can't control hardware, it can't control software, without betraying its design theories. Apple can, and so Apple will always win in a fight hands-down.
> You can't implement multitouch and make programs that RELY on multitouch.
Yes you can. Make the canvas widget support multitouch - bang, all apps support multitouch.
> You can't put iTunes controls on the keyboard, or an Eject key.
Multimedia keys and eject both work fine out of the box today.
> You can't make iChat integrate with Mail and your Address Book when not everybody will have each of those programs. It's not possible.
Yes it is, and it's used today. Pidgin uses the Evo address book, as do all Gnome programs. Ubuntu ensures that every desktop indeed has these programs, __because it controls the platform__.
Yeah, it has that, but at the same time this isn't a matter of pure functionality. This is a matter of getting a consistent polish. And having a hodge-podge set of applications doesn't do it.
I'm sorry if I misspoke earlier. I know that Gnome uses Evo to tie things together. It's just that it doesn't GIVE that impression when you use it. Evo seems, when you turn on the computer, to be a random program thrown in because it's free. Ditto Pidgin. Ditto OpenOffice and Firefox. It's because Linux has never had a consistent set of do-everything apps written by one person. And users notice this. I did. With iChat, I immediately understood the connection. I did because Address Book is a well-known application. Ditto iChat. Evo isn't. Pidgin is, but its integration with Evo isn't.
The eject key is supported, yes. But it's not the same as having a key on the keyboard that does it and NOT a button next to the CD slot. The Mac hardware is tight. And... no other hardware is. Linux can't copy that, because that introduces a matter of PAY into things.
You can implement multitouch, but you can't RELY on its being there. And for me, that's what made Leopard the OS that it is: powerful and tight multitouch.
No, but that means there's at LEAST a two-step program to beating Apple. First, you need to MAKE a set of applications that aren't hodge-podge; then, those apps have to be better than Apple's. And call me a fanboy (for some definitions that would be true), but Apple apps are top-notch and Apple is never stagnant.
There are already people preferring Linux over Apple, so at the most, it is a matter of taste. Not that some Linux apps could not be improved, but some Apple software sucks, too (what is up with the FileManager???).
"Linux, Shuttleworth said, must link up with Windows. He stressed his belief that 'Linux is the platform of the future. But I think it's essential that we learn how to work with Windows'."
I'm sure Linux developers would be more than willing to link up more with Windows if it didn't keep getting out of its way to prevent it.
it's hard when ubuntu role model is windows, implicitly by promoting it to convert windows user to linux, explicitly by making some apps look and feel similar to windows
the role model should be osx ... i'll wait the day when i see tutorials about 'how changing my osx desktop to feel & look like Ubuntu'
What Linux needs, is PC manufacturer's support. I am not talking about drivers here, I'm talking about quality pre-installs done at the factory. Once suppliers see this, they'll provide quality Linux drivers in order to compete. And for users stuff "will just work".
ThinkPad T61 running Ubuntu is in many ways superior to MBP/OSX (and 30% cheaper), but only because I deliberately configured it with only Linux-compatible components and spent a weekend tuning it upon arrival.