Shuttleworth challenges Linux to beat Apple

I currently develop on Linux and OSX as well as having used Linux for two years prior to switching to OSX for domestic duties too. The main issue, as I see it, is the "not development here" fragmentation - just how many variants of the same functionality do you really need. Focus that development power at a single target and it will reap the progress. Apple do one thing right - keep it simple and efficient by supporting the developer make the most of the platform itself. The technologies in OSX are straight forward, have clear purposes and boundaries.This makes it simple for developers to locate and utilise highly efficient/optimal components for quick development. With Linux I'm struggling to find and compare all the variants, only to find the one that fits still requires me to effectively pour energy into it to perform. Help make the user, the developer and the system productive..

More eye-candy for X. Swell. Let's all be delighted that Shuttleworth wants Linux to compete with Vista for a 'User Experience' with maximal useless crap. Would someone please get that man a C64 set up with GEOS?

I felt the title of this article was very misleading. I expected it to be a story about John Shuttleworth but it was not. I doubt John Shuttleworth has any involvment in Linux.

I think Shuttleworth (and at least one person responding to the article) miss the point entirely. The usability of OS X is not due to eye candy. The usability is due to things working well together. Things working in intuitive ways. Consistent actions getting consistent results. Universal ways of doing things. The problem with Linux (and Windows, for that matter) is not lack of eye candy - its poor design from a usability perspective (heck, look at Vista - plenty of eye candy, but still a lousy OS). Until Linux developers understand that fact, they'll never catch up.

@ Scott Jordan: I believe he compared it with OS/X, not Vista. And its quite obvious that 'eye-candy' doesnt just mean 'effects', in fact some of the most visually appealing components are just very subtle design choices. Unless you are a horribly inefficient software engineer you know that you don't have to sacrifice significant performance for a rich user experience.

Linux needs a version which is focused and aimed at the user. My own wishlist is for a version of linux OS that is easy to use for noobs, click driven without any need to learn any command line stuff, easy to keep secure and up to date, and pretty. Already there are easy to use freeware like Thunderbird for email, Opera/Firefox for browsing, and Openoffice for office apps, etc. The PITA with Linux is that apart from the ASUS eee laptop I've yet to see an easy to use version that doesn't need command line stuff. Even the Eee gets stuck with updates sometimes and needs a command line assistance. Good luck with making a sharp and focused Linux product with a clear design brief. Get the spec right and it will fall into place.

Dear whingers, whiners, and complainers of the world. If you aren't happy with the current state of Linux or other open source based solutions, then do something about it. All the development tools, APIs, and guides/people are there. Free, available, and ready for you to experiment and play with. Instead of "I wish" or "My wish list is" comments: DO SOMETHING. Be proactive. Create, innovate, develop, and most of all MAKE A DIFFERENCE! If not, there's really nothing to complain about it, is there? Is this what society has become? A bunch of complainers who don't want to do the work, but want to reap the benefits? Get off your butt and do something, if you want things to change! That's the beauty of open source solutions. If you don't like something, YOU can change it yourself! You ARE responsible for your future! As for Nick, a developer of Linux and OSX software: There's more than one way to do something in life. The point is to let humanity explore through experimentation and implementing something in different ways. Just because there's lots of ways to do something, does NOT mean you have to accept ALL of them! That's the old way of thinking. You pick the one that's best for YOUR needs. Trying to use Apple's development model (not all that different to Microsoft's in a sense) as an example, is merely re-hashing the old Industrial Age of control, efficiency and productivity. The "one size fits all" thinking. Open source isn't about "efficiency", "marketshare", "productivity", etc. (And its certainly not about coming up with excuses to get people to spend money). Its about creation of solutions through community. Its about solving problems. No, this process of development is NOT as "efficient" as those done by Microsoft or Apple. But it moves at a natural pace. Its own. Its not dictated by some clueless tech journalist trying to score ad money. Its dictated by technical merit. ie: Does it address our requirements? Can we do better? Think about this way: Do you see the developers of Firefox, Apache (web server), KDE (desktop GUI), SELinux, AppArmor, etc care for the variations of Linux distros out there? Of course not! They created something to fill a need. (In some cases, they've made their solutions workable on Linux, Mac, Windows, etc). Your role, as the software app developer, is to make software fulfill a requirement. (to solve a problem or do something better). It isn't to worry about trying to make it work with all those distros out there. Its the distro developer's role to make sure something works on their distro. NOT you. This is what I believe is the problem: Most people are conditioned into thinking the old ways are the only way to do something. While its "efficient", it isn't natural. Its rushed. It results in buggy products. Its half-arsed. It simply does NOT work in this age of Information. Look at the initial releases of Vista and OSX "Leopard". Both took their fair share of bug fixes because of a deadline that needed to be met. (If a product is not on the shelf, its not making money). Hell, look at Ubuntu 8.04...Shuttleworth was more worried about making released dates, that this release, supposedly labeled "LTS" was buggy. They had to release 8.04.1 to resolve issues that missed the original deadline of 8.04 To me, that's not right. Either do the job of a certain quality as promised, or don't promise any deadline you know you will not make. I'd rather wait for something that was properly designed and tested, then deal with some over promised, half baked, incomplete mess. This is why I: (1) mainly use Debian Linux. (2) wait for at least SP1 if I need to use Windows. and (3) wait for Intel's "tick" processor in their "tick-tock" model. (The "tick" is a refinement of the existing design, while the "tock" introduces a new design). As for the article... I believe Shuttleworth's goals are too high and too fast. He wants to push open source to meet his business goals or aspirations. He is, of course, an entrepreneur. While there's nothing wrong with that, the problem is, he has established a situation of comparison. You shouldn't ever compare yourself with someone else, and using them as a benchmark for you to strive to. You should be focusing on yourself such that you have conquered your own weaknesses and fears. The basic infrastructure of a typical Linux distro, as a usable solution, is there. It needs time to polish, and refine. (This may take several iterations). You can't expect change to happen over night. Despite our "I want it now!" culture, such quick-fix attitude is detrimental in the long term scheme of things. Another thing open source shouldn't aspire to, is excessive copying to the point of being an exact clone. It should establish an identity for itself. Making things easy doesn't necessarily mean do it Apple's or Microsoft's way. Its doing it in a way that helps people.

The whole "fix it yourself" idea is (in my opinion) one of the main problems with Linux. Yes, it's great that the source is there, and people have the freedom to fix things themselves if they wanted to. However, how does this help computer-illiterate people (the kind who have trouble with the Windows & Mac desktops)? If they were running Linux & something doesn't work properly, should I tell them to go out, learn all about how the different parts of the system function, learn C, C++, GTK, Qt, etc. then tackle the (probably quite complex) code of the specific program to figure out how to solve the problem? Most users aren't programmers. They don't care how things work, as long as they actually do what they're meant to. The Linux community wants more "regular" people using Linux on the desktop, but whenever they run into a problem (which is quite likely as things stand today) they're branded as "n00bs" and told to use Google & figure it out on their own. It's the equivalent of being told I should know how each individual part of a particular make car was made, performs & functions in order to be allowed to drive it. In that position, I'd just buy one from a different manufacturer.

While in a perfect, tree-hugging world your point is correct, in the real world where I work 50+ hours a week I don't want to be poring through tomes to learn how to program so I can join a small religion called Linux. Furthermore, google 'Con Kolivas' and see what doing what you suggested did to him - the Linux kernel Holy Guardians not only frustrated his efforts, but pissed him off enough that he gave up, THEN APPROPRIATED AND INCLUDED HIS WORK! The problem with the Linux community can be summarised like this : Picture the arrogant, fattened, smarmy comic book store character in the Simpsons. Now, picture a global community of those guys. Now, picture hell freezing over before they will humble themselves enough to compromise a little on their purist beliefs and work together for a greater good. That's Linux. I'm an extremely busy person. I've worked in and around IT my entire career. And for that reason, I've been converted to Mac by my flatmate, another cashed-up time-poor working professional. I just don't care, in fact, I have a care factor in the double-digit negatives for technology that does not obey me without question. Whose designers do not respect me by making technology that is seamless, consistent, fast, and as hassle free as possible. I dislike Steve Jobs personally. He's a smarmy, stuck-up egotist who is nasty to people. However, Apple is the *only* tech vendor who respects my time; as such, I respect their respect by paying the fat premium they command. I had high hopes for Ubuntu, however when the automatic patch process continually broke my Opera installation; when my ATI driver didn't install properly, and I trawled the net for a weekend trying to figure out how to fix it I binned it. Apple has made UNIX sing. OS-X beats the pants off anything MS ever made. I don't care if Apple is lauded by the Linux world as "Capitalist Satan-worshipping Communist Evil" <insert heavy metal demonic screaming here> the fact remains - the Linux 'movement' simply is not capable of getting their act together enough to respect 'n00bs' (and all the other elitist derogatory terms they've coined to refer to the 99% of the global population I belong to) and actually make a blazingly fast, simple, seamless, easy to use OS. I have needs. And in typical IT hubris style, the Linux community will dictate to me what I want / need / should do. When they wake up, get serious about de-throning Microsoft and Apple, and make something I can trust without having to learn scripting languages and arcane development languages, then I, and millions of others, will migrate. Until then, I'll pay whatever premium Apple asks for simple, seamless, clean, fast stable technology.

It's easy for Apple seeing as how their hardware range is locked down. Apple know in advance what hardware is used and the drivers are present accordingly. Windows and Linux both have the added problem of being likely to be installed on a huge range of hardware. Some of that hardware is obscure or doesn't work well with other types of hardware. All this must be allowed for because people these days don't, by and large, have much patience for "nearly works". For Linux distros there would need to be an agreement on which desktop was to be used. (Gnome or KDE? - That should keep them arguing for a year or two.) Then the developers of that desktop would have to agree to work towards a series of pre-agreed standards. The tendency for different distros to put some system files in different places (init.d or rc.d?) would need to be dealt with. It's not as simple as it first appears. And finally - Mr Shuttleworth, you already have a Linux distro, how about you go away and make it better than OSX and then come and tell us all about it?

You know why Linux will always fail to compete to any other closed-source OS? Because there are TONS of variants, and anyone is free to do it on his own. There is no standards -- actually, there are some -- but the way to do things varies too much to be considered a solid system. Yes, Linux *is* a solid system, but how you do simple admin tasks on a RH, on SuSE, on Fedora, on Ubuntu? Are them the exact very same? No way. So, how can you beat systems which has a very standard way to achieve such results, if anyone do it by his own? Linux brought us freedom to choose. This same freedom is slowly killing it.