Quote:

Original post by Oluseyi Quote: Original post by seanw I don't understand how people can make statements like "all OSS have bad interfaces", "all OSS lack innovation" etc. It doesn't make sense. See generalization, under opinion, cross-referenced with use your brain. I'm sure that somewhere in your post was a meaningful, productive point. It just happened to go under my feet.

I'm sure you've got a point too, but do you have to be such a jerk about it? Please, read your post. Is that a polite way to respond to somebody who was neither addressing you directly or being insulting? I'm sure you're going to give some terrible reason why this is acceptable; you're just being rude and trying to look like a big-shot.

Quote:

Seriously, quit nitpicking over pointless semantics. We're talking about the average case. Read the ESR articles/rants I linked to. Maybe it'll hit you hard enough with a clue.

I'm not nitpicking over semantics, I'm trying to say that saying all OSS have bad interfaces is a stupid statement to make. Sun even did an HCI evaluation of the Gnome project and the project developers now strictly follow those guidelines. Novell and Suse are doing great work to improve the usability of the Linux desktop. I personally find KDE easier to use than Windows. My own family use Linux after using Windows with little problem except learning subtle differences. My Univeristy has Linux on all the computers (including those doing non-IT degrees) and people don't have trouble browsing the web and checking their email. There is countless non-OSS projects that have awful interfaces.

The point I'm trying to make is that the license choice of software does not make a difference to its HCI. For example, Mozilla is OSS and the interface is comparable to IE which is the most widely used internet browser.

It's just a pointless and gross generalisation to say that all OSS have bad user interface because it isn't true. You might say I'm nitpicking but the article is trying to tar all OSS and its developers with the same brush and it's not rational.

If I wrote an article entitled "All Windows software has bad user interfaces and lack innovation" you would flame me to pieces. It is an equally obsurd statement to make.

Quote:

Original post by seanw I'm not nitpicking over semantics, I'm trying to say that saying all OSS have bad interfaces is a stupid statement to make. Sun even did an HCI evaluation of the Gnome project and the project developers now strictly follow those guidelines. Novell and Suse are doing great work to improve the usability of the Linux desktop.

I've noticed this trend. It seems ironic (to me, at least) that most major HCI advances in Linux are being led by, er, private, commercial corporates. Both Sun and Novell were heavy (and rather sore) losers to Microsoft, who merely produced better products. It's not even as if nobody could have seen MS' victories coming; Novell made some incredibly stupid mistakes and they got the commercial hammering they deserved.

Solaris didn't get its "Slowaris" nickname for nothing. And Novell's NetWare was _way_ ahead of its rivals initially. Novell made the crass mistake of resting on their laurels, then blaming everyone but themselves when they realised they'd been overtaken and left way behind in the dust.

I don't see these HCI efforts as being an integral part of OSS, but merely of two corporates trying desperately to make up for lost market share using the cheapest means at their disposal. They are taking advantage of Linux because it effectively gives them a low-cost OS to port their existing codebase to. That neither company has ever been accused of "getting" usability isn't going to help, I suspect.

Quote:

If I wrote an article entitled "All Windows software has bad user interfaces and lack innovation" you would flame me to pieces. It is an equally obsurd statement to make.

But true, nevertheless. Windows is based on the same, tired, outdated concept as KDE and Gnome. (And, yes, even OS X.) Besides which, the original argument isn't so much that KDE and Gnome are 'bad' interfaces, but that they are merely indifferent clones of existing ones. KDE _is_ Windows, barring the extra doodads and gimmicks plastered all over the taskbar. Gnome is likewise all but identical.

Why in hell was this EVER allowed to happen? Why in blazes does Linux now boast TWO, all-but-identical, user interfaces, neither of which is particularly different from the existing Best Of Breed

? Since when did Earth actually need this massive duplication of time and effort to produce TWO mediocre clones?

At present, Linux has reached the point where it offers a sort-of-Windows-like-ish experience, but without the polish and coherency. It offers the choice of one sub-Windows-Lite-GUI or another sub-Windows-Lite-GUI, both of which are maintained by hordes of vicious fanboy monomaniacs.

Or there are the few nerd-fest GUIs maintained by some random loner in Latvia.

Mozilla? Firefox? Give me a break! What's so damned special about a fucking web browser? It certainly took long enough to write and when it was done, what did we get? A brilliant new take on the browsing concept? A wonderfully seamless experience to accessing the World Wide Web?

No.

We got Internet-poxy-Explorer.

Again.

Open Source Software has given us a clone of a 30-year-old operating system, built on 40-year-old concepts and ideas and two blatant rip-offs of existing GUIs. Whatever OSS claims to have given the world, it sure as hell isn't "innovation".

Hell's flames, even Apple has managed to produce some novel GUI concepts in the time it took for KDE to, er, copy a few more of Windows' existing features.

People sometimes ask me what's so great about Linux. The problem is that there's nothing remotely "great" about Linux. It's a missed opportunity.

I'm ranting. I would apologise, but I can't bring myself to do so. OSS could have been so much more. It could have truly revolutionised the way we work with computers. But it didn't. It hasn't. And, I suspect it never will, because the egotistical, self-important overblown arseholes at the top of the OSS food-chain have no interest in doing so. They believe "free as in speech" is actually enough of a validation in itself.

I contend that it is not. Freedom of speech is valuable only when you have something of importance to _say_. And GNU/Linux simply isn't important enough.

The only -- and I do mean ONLY -- reason GNU/Linux has had any success at all is because it is free. As in "beer". I believe it may well set back computing technology by years, if not decades.

I am really, really pissed off by this.


--
Sean Timarco Baggaley

That's "Windows XP", not "OS X". If OS X were so great, KDE and Gnome would be copying that.

Ok, enough complaining, let's actually start doing something about it.

Not. My point is, people have been complaining about Linux ever since its inception, but the zealous "monomaniacs" that you speak of control what gets in and out of Linux. So why don't we just branch off and spearhead our own project? I don't know why people don't do that. I'm sure if someone took the initiative to start new Open Source Projects to improve Linux, they would gain quite a number of followers.

Myself, I'm trying to learn as much as I can about Linux so that someday I can be competent enough to contribute my little share to some Linux component.

Quote:

Original post by seanw I'm not nitpicking over semantics, I'm trying to say that saying all OSS have bad interfaces is a stupid statement to make.

Nobody made that statement. Nobody said that "All OSS, categorically and without exception, have bad interfaces." Which was my point. You're getting agitated over nothing, largely because you feel that it offends your sensibilities.

Whatever.

Generalizations are a useful mechanism for examining aggregate or average behavior across a substantial sample of individuals. If you can't process that, your opinion is less than worthless.

Yes, I have a habit of being rude. To idiots. Isn't that the OSS way?

Hm...I've been reading this over and over and I've came to this conclusion.

Primarily, just because GNOME and KDE lack innovation (by "copying an antique interface"), doesn't mean that they're bad interfaces. I think what makes or breaks a desktop interface is not how innovative or new the interface is, but how it is implemented. I believe the traditional desktop interface that Windows, GNOME and KDE are based on still have life in them, because they're tried and true interfaces. Ever heard of the phrase, "an operating system is only as good as the applications written for it?" Well our problem lies in the lack of standardization in interfaces for applications.

"Linux is hard to get around" or "Linux is hard to use" comes from the fact that its applications are hard to use, or hard to get around, not Linux itself. If we go down to the definition of Linux, Linux is just the kernel, managing the low level processes. If we took Window's interface and fitted it around the Linux kernel, we would never know the difference. Again, our problem lies in the applications.

So I think most of our energy should be focused on making the user experience with GNOME and KDE coherent, streamlined, and consistent. Innovation comes second.

Quote:

Original post by Cipher3D I'm sure if someone took the initiative to start new Open Source Projects to improve Linux, they would gain quite a number of followers.

I suppose you skipped redmilamber's comments. Let's take a look at what "someone taking the initiative to start new Open Source projects to improve Linux" actually entails.

What is Linux? No, not the technically correct but largely useless answer of "a kernel." What does it connote to the average user?

"Well, there's obviously the UI..."

Okay, say we wanted to start with that, since that's the first thing a user sees, what does a Linux UI consist of?

X Windows, a window manager, an optional "operating environment' (GNOME can use multiple WMs, including that piece of bloatware known as Enlightenment) and various graphical applications.

So which do we tackle first?

Well, we can't tackle X first because it would break everything that runs on it. X might not even be the problem. We could tackle the window manager, but that's a rather broad project, and doesn't fundamentally affect usability outside of, say, changing the effective drag area per the IceWM example given by redmilamber. Okay, so it's gotta be the applications.

Which ones? And how do we get people to switch once we've implemented them? Or do we simply establish GUI guidelines, create a few example apps and hope that people follow the example?


My point is that this is hardly a simple thing that anyone should rush off half-cocked on. In addition, it's not just a cosmetic problem; user interfaces expose (and hide!) the underlying logical design. In many instances, the entire application stack will have to be - at the very least - revisited. Now multiply that by all the applications in common use.

(I'm using the term "application" here to refer to tasks that can be performed, not necessarily a program comprising its own top-level window, etc, etc. The validity of the application as such a fundamental unit has, rightly, IMO, been called into question.)

What is necessary here is a complete reevaluation of the nature and objective of computing, of the tasks that we engage in and of the means of completing them. In a sense, it is like writing software: don't do enough upfront requirements gathering or design and you end up with an unusable mess.

And that would just be ironic, given the context.

Just get rid of X entirely. Its completely arbitrary line of seperation between "server" and "client" is a stain on even mildly passable engineering standards.

Quote:

Original post by Cipher3D Primarily, just because GNOME and KDE lack innovation (by "copying an antique interface"), doesn't mean that they're bad interfaces.

Actually, they are. Windows is a bad interface, too, and OS X is just as dumb as the rest of them. Windows and OS X just happen to be a little more polished and refined. The fundamental approach is just as broken.

Here's the thing: all of our metaphors in computing - files, windows, icons, menus, folders, pointers/cursors, programs, applications - they're 30 to 40 years old! While there may still be value at the core of them, isn't it time we at least reevaluated them?

Read this: When Good Interfaces Go Crufty. The revelations aren't earth-shattering by any means, but they will make you reconsider what you had always accepted as the norm, and hopefully take that attitude to other parts of the UI/UE.

Quote:

Ever heard of the phrase, "an operating system is only as good as the applications written for it?" Well our problem lies in the lack of standardization in interfaces for applications.

Why are there applications at all? Why aren't there simply extensions to the operating environment? Why do I need specialized applications that all save data in incompatible formats, that duplicate code significantly (consider the number of applications that can read text files; tell me that isn't equivalent code - and effectively bloat)?

The entire point here isn't to bash OSS or to venerate Windows, the point is to villify OSS for the opportunity that it is failing to seize - a genuine chance to reexamine how things are done.

Agreed, we first had to do the familiar or the socio-cultural movement that is distributed open peer review would never have had an attainable, familiar challenge around which to coalesce. Now that Open Source is mature, though...

Michalson:
Is that a Wierd Al take on Ma$e/Puff Daddy's Mo' Money, Mo' Problems as your avatar?

Brilliant!

In other news, I completely agree about X, but then you always get some Unix traditionalist pointing out the obscure benefits of intrinsic network transparency. Given that Windows' Remote Desktop can achieve the same thing, however, I think the argument is finally dead.

Kill X.

Quote:

Original post by Michalson Just get rid of X entirely. Its completely arbitrary line of seperation between "server" and "client" is a stain on even mildly passable engineering standards.

Actually I think I'm on to something. OSS is far too dependent on this cult belief that Unix is the height of software engineering and usability. At the worst of it you actually hear arguments of "Linux is secure because it's based on Unix" (and yes, that is an exact argument I've seen used countless times).