Open Ballot: Too many geeks spoil the OS?

Open Source

'Scratch your own itch' is a popular mantra for open source developers. And the principal is a good one: programmers working on software they want tend to produce good code. However, the itches of most coders are very different from the itches of most ordinary users.

The end result is that we have the best selection of text editors of any OS, yet (according to Miguel de Icaza) an audio framework that breaks so frequently it's not worth setting up.

So, here at LXF towers we've been wondering, is Linux suffering from a surfeit of hackers scratching their itches? In order for Linux to become a leading desktop, do we need the geeks to step back and let non-techies (i.e. the majority of the population) take control of the OS?

Let us know your thoughts in the comments.

You should follow us on Identi.ca or Twitter


Your comments

Itching scratches...

...it's got at this far so why change a Good Thing, something that obviously works incredibly well?

If anyone is going to attempt to push a unified desktop towards the general public, I'm guessing it's going to be a commercial company, given the size of the task - any developer individual would have learned to use a particular desktop framework and its foibles, so the only reason to try and push a unified desktop would be money?

linus says

Best quote ever from your magazine is the one from Linus in your new edition.

Paraphrasing as at work:
Everything would be easier if there was no choice.

The man is a philosopher genius.

Absolutely

As for Linux itself, distributions are better than they have ever been and the desktop environments like Gnome & KDE are fine as they are, it's the applications that let the side down.

It's not about having a "unified desktop", it's about having applications that stand up well against proprietary alternatives, look good and are easy for us users to work with.

Having an application designed by a coder is not unlike having a house designed by a brick layer. You need architects and joiners in there too, or you'll end up with a collection of walls that don't fit together.

Don't create itches

I don't think there's a problem with scratching your own itch, but if the cure to your itch involves creating more itches for others, then there's a problem.

Speaking as a user, pulseaudio has resulted in more trouble than ALSA, not less (though perhaps it's easier to develop against, I'm not sure). Similarly I'm wary of the benefits of systemd over existing start-up solutions, but I confess I've not looked at it closely enough yet.

I think the slower and more painstaking approach of creating and getting adoption for open standards is the way forward, rather than trying to come up with unified frameworks in software that may restrict developers and disappoint users.

No

No to a unified desktop, I use several different ones on different installs & they all have their good & bad points, but I have a choice. Also no to trying to force coders to work on something they don't want to, because that will lead to bad code. If enough people don't like the present audio framework, someone will say "I can do better than this" and start to work on it. Also, I haven't had any problem with audio myself, so I'm not sure what he is talking about.

Itching Desktop

We've just seen the itch in action. Gnome bring out Gnome 3 with their shell and a lot of people get itching. Now we have MATE, Cinnamon and a few others under active development, plus xfce and LXDE are getting much more popular.

I just wonder how long it will take Gnome to realize they have lost a lot of people. Mind you, it doesn't mater, while Gnome heads for desktop obscurity, those of us in the Linux world will have plenty of alternatives!

mixed

What we should do is hold a month were we ask what the non-techies want, and every week dev's can do what they want and 'scratch your own itch'.

money

if you want to 'scratch your itch', buy a back-scratcher?

No

Too many geeks got us KDE, GNOME, Cinnamon, MATE, e17, fvwm, twm, WindowMaker, etc.

Too many geeks got us Ubuntu, CentOS, Fedora, SUSE, Arch, Gentoo, Minix, FreeBSD, OpenBSD, NetBSD, Haiku, etc.

Too many geeks got us the gopher, usenet, ftp, BBS, P2P, etc.

So no.
Too many geeks will not spoil the OS.
Too many geeks will spoil us [with choice].

take control ? how ?

What does non-geeks taking control of their operating system mean?

How is anything going to get done? The reason that Linux is an operating system is because hackers found an interest in a an hobbiest OS and continued to improve it.

Choice is there, because people deserve to have what they want when they develop the software.

Osx and windows are made by geeks also, the only difference being that the have a common goal, common framework.

Linux is a wonderful operating system. We should look into praising those developers who do all the work, from tiny bugs to complete projects.

We need to promote and make celebrities of the people who bring us the software that we love. Perhaps with this encouragement in place, we will see an even better community and one which is more united as a result.

Yes, no, maybe

The sort of hackers who give us yet another Ubuntu respin (with nothing new save a different desktop theme) would probably not be capable of hacking PulseAudio. People will only scratch the itches they can reach (hence the splendid selection of text editors).
Difficult Free Software projects that work tend to have someone or some organization behind them, such as Linus and the kernel maintainers, the Document Foundation with LibreOffice and any one of the major distributions with their founding company or stable community.
Who speaks for PulseAudio, provides it with cohesion and direction? While a quick Google search shows me a number of associated developers and a web site, I can't see much of a cohesive organization behind it.
The extent to which Linux and Free Software benefits from people scratching their own itch, depends on the talent of the person or people doing the scratching, to that extent it doesn't suffer from people doing largely pointless things as no one is forced to adopt their second-rate solutions. The idea that these developers could be corralled into dedicating their time to more useful projects is undoubtedly attractive, but isn't (a) feasible without some particular incentive or (b) desirable if they lack the necessary talent (I recall reading some articles in which lead developers wish that some people would not submit their dodgy patches).

the best...

will prevail. no matter how many times you successfully scratch your back, you will always ditch your hand for the back scratching diamond in the zone. good ideas last longer, ence, geekiness is not bad for any OS.

It's odd...

It's odd that the argument goes as follows:

User Friendly Features are missing from Linux
+
Geek Friendly Features are abundant in Linux
=>
There are too many geeks in the Linux-verse

Doesn't this just mean we need *more* geeks, so the User Friendly stuff gets taken care of as well as the niche stuff?

It's like bringing food to a potluck. If too many people bring exotic spicy desserts, you can just invite more people that are inclined to bring meatloaf and salad. Too many spicy desserts isn't your problem as much as your lack of meatloaf loving pals.

the best often dies fast

The best often dies fast in the face of strong marketing, good PR and a cheaply/quickly to made product.

I never heard the term PR geek before...

growing up

I think it's mostly about maturity. The Kernel had to come first, but now that's essentially finished (of course it gets updated and added to but there was a point at which it became 'complete').

The frameworks came next and they're sort of finished - they're at a point where people can quite quickly produce (whether from scratch of through a fork) a new desktop (Unity, Cinnamon, MATE) based on existing frameworks.

The foundation is really solid and very powerful. We all know that, that's why we use it. We just need to expose that to the average user, so next come the graphical apps.

As more people use Linux (and I do believe it's growing, if slowly (and the leakage to OSX has probably peaked)) then there'll be more demand for apps, more people will realise that's what's needed and work on them and etc. It's a slow process but that's the nature of the penguin.

So no, I don't think the geeks need to step back, I think they'll mobilise where they're needed, as they always have. But yes, we do need to let the 'non-technical' users take over and do their part too. We neeeeed designers and writers and UI/UX people.

And as far as I can tell pulse has been working fine for years (over various distros on various hardware I've had no big problems for at least a good couple of years). The automatic pulseaudio bashing is really getting old.

The slackware Model

I like the slackware model that gives you enough of a system to change to what ever you want it to be giving you control over your system. However not everyone is a power user a may require a distro that holds their hand and is based on the whims of the developer. The great thing about distro hopping is finding out which distro fits your needs.

As for the audio please don't get me started on pulse audio for me alsa works just great, howver some people love it and it works great for them.

I thing the way things work at the moment is great.

don't get the point

I do not get the entire point of this "Linux leading desktop discussion". Why do some people try so hard to make it replace other desktop OSes...

I use Linux exactly because that is what fits best to my work style. I don't care about an unified desktop experience, sound or video for working.

Using Linux means I have to get things done fast and efficiently. No time to admire the beauty of a desktop and icon set, or the splashy wobbling window animations, etc.

If I look for some unified beauty desktop OS, I get myself a Mac, if I want gaming I get myself a Windows box.

So DON'T SPOIL THAT by some silly attempt to make it more mass compatible.

The only annoying part from being not mainstream is the missing support from hardware manufactures. Albeit that got (thanks to the community) much better over the last decade.

Geeks forever

The geeks that create/maintain an OS are what makes an OS. People who have an interest in what they are using and have a need for it to exist. The reason we have an array of distros, desktop environments and text editors is that people have an array of needs.
Any "Linux for the masses" distro would have to compromise somewhere and would therefore alienate a section of current Linux users before its even begun trying to attract users from other OSs. Choice is the best and worst thing about Linux and remains why it probably wont ever be the next must have desktop OS.

I'm more than happy with

I'm more than happy with Linux in general but specifically audio is pants and printing photos is similarly pants.

I have one computer that runs Windows so that I can easily print photos, it would be great if all the different photo manipulation programs were unified and something to rival "easy photo print" came out of it.

Can never be too many geeks...

All those itches and all that scratching gets filtered and distilled into the end product - sure you can run the nightly builds, like Miguel, and have broken audio. You can also pick almost any stable distribution and listen to CDs.

Just imagine if you took all the geeks (=authors) out of fiction-writing, and let the readers choose what goes into the plot. We have that already, with 200+ channels of reality TV - do you want an OS for the lowest common denominator?

No wonder...

Linux has too many geeks and not enough artists.

Time to come out of the basement

Sorry, but yes!

Websites were once written by the techie in the basement, but are now organisation-wide efforts with 'under construction' animated GIFs a distant memory. And better for it.

Likewise, Linux needs people of all skills involved - entrepreneurs, business developers, accountants, marketers, writers, UX designers and so on - if it is to ever reach a wider audience.

Otherwise, Linux will continue to appeal mainly to hobbyists with the time and patience to evaluate countless desktops and who aren't put off by application names such as The GIMP.

Paid Labor, Free Software

+1 to Rhakios

I want to support those "mid-list" projects like PulseAudio that do not have corporate and/or foundation sponsorship, but are too difficult and essential for volunteers to maintain in their spare time.

I don't think we have too many geeks, I think we have too few mechanisms to *pay* geeks for their labor on Linux-related projects. By mechanism I don't just mean the monetary transaction itself, but a culture where donors have some idea that their contribution will make a difference (ala Kickstarter and Indiegogo).

I would love to see the nice aspects of "professionalization" – stability, more resources put into documentation and user experience – without turning completely over to the logic of markets and sacrificing FLOSS principles.

unified desktop NO - a highly customizable one - yes!, please!!!

As far as I am concerned, I would need a stable and secure desktop, which I can tweak/customize as per my needs without being too tech savvy, preferably based on the KISS principle. I'm a newbie. However I don't feel that too many geeks spoil the OS, I feel rather that too much of attention (i.e. perhaps in a wasteful manner) is what's the problem. Hey, as I said I'm a newbie and don't know the ins and outs of the OS tech what have you, but I feel maybe if there's like 10 geeks working on a project for 10 hours, while the project only needs like 9 geeks and only 9 hours, maybe that one geek and one hour that's being wasted could be put into a more useful area of maybe like another project that needs 10 geeks and 10 hours but only has got 9 geeks and 9.... well you get the idea.

Just Not enough Polishing/Finishing for "Prime Time"

THE REAL Problem is: With over 3,ooo+ VERSIONS of Linux Available; I do NOT know of Even One [1] that that can Compete With Windows or Mac, in being: quickly AND easily setup; with working email/printer/scanner and internet WITHOUT knowing/using/studying/copying/typing CODE!
I; personally, would prefer to NEVER Need to even know what CODE Is!! I just want an OS so that I can get onto the internet/recieve/write/send email; make a copy [printer] with my printer/copy with my scanner .....Just normal everyday tasks expected and accomplished with Either Windows or Macs.
NO Versions of Linux [That I am aware of] are enabled to do that as Easily or Quickly....without doing something with "Codes".. It may Surprise the Geeks; but, their interests are NOT MY INTERESTS. Please, Geeks, POLISH/FINISH ONE Version, and you would be really surprised at the response from the "General Public"...That Includes Me....
Thanks for reading, GS

Linux is like a "Kit Car"

I really do think of Linux as being like a "Kit Car" that needs much Expert Work, in order to "work/operate' as good as an "Off the Showroom Floor" New Car. With my knowledge and tools, I COULD Finish a Kit Car; but, I'm not interested in doing that much work. [I have other interests]
My Wife uses her Yaris for transportation; and that is all that I expect from my computer and printer/scanner: to operate as expected.
I consider this Mint 10 equipped computer operates better and quicker than my XP, for about 60%....The other parts need Finishing, before the public, and I, will be overjoyed.
I am a Virgo...meaning that I have Many Unfinished projects still awaiting my attention, tho, no one is expecting me to finish any project soon, as they are just hobby"s. I'll guess that all 3,000+ versions of Linux, are all just "Hobby's".
Are ALL Geeks Virgo's??

What does non-geeks taking

What does non-geeks taking control of their operating system mean?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Post new comment

CAPTCHA
We can't accept links (unless you obfuscate them). You also need to negotiate the following CAPTCHA...

Username:   Password:
Create Account | About TuxRadar