GNOME Ignoring its Own Users? 735
Jonathan writes "Some editorials were posted on the web the last few days about GNOME and its apparent lack of interest on user feedback, especially when GNOME pitches itself to follow a 'users first philosophy' in their press releases. OSNews started with an editorial about market research or lack thereof, Expert-Zone posted another one on how OSS must learn to take responsibility on its great success."
Hot Button Topic (Score:5, Interesting)
Re:Environments vs. Simple WIndow managers (Score:5, Interesting)
Gnome is nice, but (atleast in this particular topic), Eugenia has a point. We keep hearing how Gnome focuses on usability and user-friendliness and then they come up with stuff like those awful file dialogs, or the damn bloat, which makes the system crawl running a few apps.
I haven't tried Gnome for a couple of version revisions now, but XFCE gives me what i want and does the job fine.
If you need a feature, buy the feature. (Score:5, Interesting)
Developers that are getting paid to work on GNOME are beholden to those that pay them. Yeah, they're working on an Open Source project, but by taking money for their time, the people paying them get to direct their coding. Unpaid developers are beholden to themselves and themselves alone. That's the way it should be. If you don't like it, you need to literally put your money where your mouth is. As has been said many times before, free software only costs nothing if the time spent developing it is worth nothing.
Re:I don't know... (Score:4, Interesting)
That being said, I dropped Gnome years ago because I felt it was focused more on programmers doing what they wanted and giving other programmers cool programming stuff, and KDE was much more focused on an easy to use experience.
And before I get the usual flames from someone with no life that thinks anyone who doesn't use a console is a a loser, I have a small business I run that is based on software I wrote. I was using command lines back in the late 1970s when I was lucky to get time on a paper terminal and excited when I could use a VDT.
I spent years in between teaching special ed and learned that people actually think in different styles, so many people will always do better with a GUI. My experience is that KDE has always been focused on creating a good GUI for the end user, wereas Gnome was more focused on a GUI with great APIs and everything programmers want, without a reall awareness of what helps end users.
Re:Something in the water? (Score:3, Interesting)
Why oh why.. (Score:1, Interesting)
Re:Obligatory comment (Score:3, Interesting)
I hear ya. GNOME is hellbent on cloning Windows internals while 'innovating' the look & feel (ignoring the whole spatial nautilus fiasco) while KDE is hellbent on cloning the look & feel of Windows while pushing new innovative internals, even if tied to C++ a little too tightly for my taste.
Why can't we get them to swap their bad halves with each other and have a desktop pushing innovative internals AND new ideas in appearance while the other effort focuses on cloning Windows inside and out atop a *NIX base.... Call it OS W. (ducking)
Unfixed bugs (Score:3, Interesting)
However, the biggest pain in Gnome has got to be Nautilus. It has always been and continues to be slow, buggy, crash proned, and a memory hog. Don't even get me started on the "spatial mode" crap, which is forced down our throats.
Exactly backwards (Score:5, Interesting)
In our article yesterday about "The Ten Worst Engineering Pitfalls" by Keith F. Kelly, on the No2 spot you will find this: "2. Basing the design on your own motives rather than on users' needs."
She uses this to argue that programmers should be user-driven -- but as Alan Cooper points out, this is exactly backwards. When a company is user-driven, they add a lot of little features and tweaks that each of their users asks for. Then they end up with a program that's intricate and complex and hard to use for *everybody*. (If it's a company, this is where their customers start leaving them for companies who take design seriously.)
No program (or system) can be perfect for all people. The successful ones are the ones that have a consistent design -- often this means doing one thing and doing it well. If you try to be all things to all people, you guarantee that you won't be much use to anybody. Attaching a shell to the bottom of every window is the ultimate in flexibility, but nobody would claim that it's the ultimate in usability.
The problem is that Eugenia seems to think "user-driven" is a good thing, whereas Cooper (who seems to have much more experience and success and believeable examples to back up his position) states quite emphatically that "user-driven" is a bad thing: you want to be *design-driven*.
OSNEWS & Eugenia (Score:3, Interesting)
Well, it was a link from
Of course it's a troll.
I also can't help but be annoyed by Eugenia claiming that "this is why Linux will never surpass Microsoft and apple". People like that think that by annoying people they can push them to work harder, and appear a kind of "hero" - in the "I gave them the push they needed" way.
Linux probably won't surpass MSFT any day soon, but when we get 20% desktop coverage I can safely say that Linux has kicked MSFT's ass
Re:The grand secret of spatial nautilus (Score:5, Interesting)
We SHOULD all change our habits to fit the GNOME paradigm, rather than the other way around.
I guess I should stop bitching about how, horrible, nonsensicle, slow, clunky, awkward, unintuative, difficult and inferior spatial browsing is and just brainwash myself into liking, no, adoring the 50+ open windows peppered across all my desktops.
Re:I don't know... (Score:4, Interesting)
IMO, Gnome is very focused on making things easy for the user, and very usability-test centric. The developers seem to want to stick to usability testing and the Sun-funded HIG almost to a fault at times. To the point that they *do* listen to usability tests more so than the users. There are times where this is bad, but at the same time, there are times that it's the only way to actually get a feature implemented.
Has KDE developed a comprehensive HIG and/or UI guidelines for its DE? I honestly don't know. I'd always assumed not because K apps tend to do things their own way and I've never been able to find much consistency between them.
Again, I don't think KDE is a bad enviro, I think it's largely a matter of taste. Gnome leans toward a very simple, sparse environment, whereas KDE these days leans towards a complex, but highly customizable enivornement. I personally find the former easier to use, but there are certainly arguments for the latter.
Regarding the open-source software model (Score:2, Interesting)
Frankly, I'm quite amazed that open-source projects don't use services such as these more often. Bugzilla could probably use more integration with these kinds services (say... as a web service). In addition to the "vote for this bug" feature, you could put an "add to the bounty" feature. That should solve most Gnome issues at blazing speed, with the proper integration. Moreover, Microsoft would likely have serious trouble competing with the model, and I might even give Linux another try and consider not calling it a sucky OS.
Re:Don't feed the troll (Score:3, Interesting)
She is the class of computer user who has just enough knowledge to be a pest but not enough to be useful.
Two schools of thought... (Score:4, Interesting)
1. STFU and Fork It - While I disagree with this (for reasons I will outline below), I agree that this is a valid point. For the most part, the people working on these projects are working entirely for free. As such, they have no real "customers" per se, because no one is paying them any money. Hence, they have no real obligation to care or even notice when someone suggests a new feature. The users, who are using software (for free) which was written on donated time, have no right to complain if it doesn't do exactly what they want.
2. Listen to your Users - Forking a project is fundamentally hard. You need, at bare minimum, a ton of extra time, skill in the language(s) the project was written in, and a working knowledge of the project's code base. Additionally, when a project is as widespread as GNOME, it's next to impossible to get any notable linux distributions to include your fork instead of the trunk. X.Org managed to pull this off, but only with the help of a large number of developers. When you tell someone to "STFU and Fork It", you're telling them to do the following:
That's a lot harder than just opening up a text editor, magically finding the right place to add your little snippit of code, and recompiling.
The spatial browsing controversy was what finally convinced me to give up GNOME for KDE. The straw that broke the camel's back was a very condescending article in favor of it that essentially claimed that anyone who didn't like the spatial file manager was using their computer wrong; however, since version 2.0, GNOME has had a history of removing configurability in favor of what the developers believed was simplicity, despite the vehement objections of their user base. The spatial file manager ordeal was just a stark example of a larger pattern.
For those of us who are trying to advocate Open Source, it would be really nice if certain developers were more willing to listen to their users. As a matter of policy, it would be a good idea to set apart a portion of the dev team whose specific duty it is to to proactively study and implement (with a how-can-we-make-your-experience-better attitude, as opposed to stfu-and-do-it-your-goddamn-self) feature requests. Why? Not because you necessarily owe people anything, but because people use your product, and it would be nice if you cared about them.
In the meantime, I've switched to KDE, which has shown itself to be far more responsive to the needs of its users. As things are going right now, GNOME will either adapt to the market or become obsolete, much like X did.
like the splash screen contest (Score:5, Interesting)
- didn't post the judging criteria
- didn't use any judges other than those that run the footnotes site
- didn't listen to the feedback they received regarding the winning choice
- shunned a good amount of popular opinion
i could go on but i won't. i was a proud GNOME user for years until it seemed the GNOME developers stopped hearing our cries. for example - the gpilotd project failed and when the users cried out - no one seemed to listen. all the while the KDE developers were busy taking in all the feedback from their user base and, when their next version was released, it was obvious they took that feedback to heart. i honestly don't know what the issue is. GNOME used to be (from my point of view) the desktop of the "common man" for the Linux community. not so any more. now it's become about as user-unfriendly as possible (i.e. spatial file managers and hard-to-create desktop icons). when is this going to change? or is it? is KDE going to become the defacto standard for more and more users while GNOME finds itself being used only by those the develop it? it seems to me that GNOME is now what Linux was nearly a decade ago - a project for the elitist/hobbiest/hacker and not the masses.Re:Obligatory comment (Score:3, Interesting)
Why do you say that? I think they work good for what they do.
I've been using computers since before there was a GUI and the taskbar/start menu is a good thing.
Yes, I use OS X every day, and I still think it sucks because it doesn't have a real taskbar or start menu.
Re:The grand secret of spatial nautilus (Score:3, Interesting)
The difference lies mainly in how you want your metafor to be, the browser metafor "browses around", looking at new things in the same view. The spatial metafor tries to model everything you want to look at as a separate object, and tries to model this by opening it separately. Some people say that this is easier to understand and relate to for beginners, but that is really hard to judge for someone like me, who is used to browser and commander interfaces.
Anyhow, the idea as I recall it was to lower the bar for everyone, especially people new to computers - which is a great goal. The bad thing was that they tried to tell us all how we wanted to do things.
I'm pretty new to Gnome, been using KDE until I switched to Ubuntu. I've been trying out the spatial way for quite some time now (as in months), just to give it a fair chance. Because, I've recently started to understand that especially we power users are missing out on a lot of great things just because we grumpily do things the way we've always done them, and so I've started to try and do things the designers way instead of spending hours on reconfiguring every computer I sit down at. I've discovered that several things I've frowned upon for years, for reasons that now escape me, actually make my life easier. Like the trash can, believe it or not!
Anyhow spatial navigation isn't nearly as bad as people say when you get used to it, but it still mostly clutters and slows things down. I'm probably gonna call the experiment in favour of the nay sayers soon and start browsing again.
Re:Don't feed the troll (Score:5, Interesting)
While you are at it Eugenia, your readers/users want
1). A better comment system for OSnews.
2). Registration based commenting.
3). Support for all XHTML tags (it's freaking 2005!)
4). A better moderation scheme.
5). A user friendly editor with spell checking and automatic tagging.
6). Ability to reply directly to comments with ugly @ in the reply field.
7). Ability to place certain trolls on an ignore list.
8). Ability to edit comments that have already been posted.
Oh and your users have been clamoring for these features for years. Why haven't you implemented them?
Re:Hellbent on cloning the look of windows???? (Score:2, Interesting)
Look != feel. It doesn't matter if I can make KDE look like anything I like, it still doesn't change the feel.
I have a constructive proposal (Score:2, Interesting)
My idea is to extend Bugzilla to allow users to donate money directed towards a precise feature (or bug).
To see how this is related with Eugenia's article, please read on.
The problem
In the article, Eugenia correctly points out that OSS projects tend to ignore the needs of ordinary users: currently the OSS model tends to favour the needs of corporate customers (because corporations such as IBM or Novell put the money for those features), and those of power users (because they program the features they need by themselves). But the needs of ordinary home users are not respected as much. They often tend to be ignored.
So Eugenia is right here. But, IMHO, her error is to assume that this lack of respect for the need of ordinary users is the developers' fault. Instead, I think it is the users fault; I think users are still not as responsible and self-aware to get together and directly finance the features they want.
Consumers are not yet a responsible community and they have no awareness of themeselves as a power. (I believe this is partly why the economy is dominated by corporations.)
The solution
The idea is that
I am sure many of us have a small amount money, but no time to contribute code. Some of us love OSS so much that we would happily donate some, *provided* we are guaranteed the feature to be implemented if the threshold is reached.
Example
Here is the process I am proposing:
Notice there are two times involved here:
- six months. This is the maximum time needed to implement the feature.
- september 6 2005. We can call this the "restitution date".
Both should be decided by the developer, I believe.Many more things should be discussed:
- Once the threshold is reached, how do we decide which programmer is to take care of the task?
- What does "guarantee"mean?
This is a difficult matter. Above I talked about "guarantees" as an incentive to donations.
Suppose the threashold is reached for a given feature: exactly, to what extent are you guaranteed the feature will be implemented? I mean: what if the developer fails to implement the feature in the given time? After all, 100% guarantees do not exist in the real world; that's why collaboration contracts exists, stating what happens when one of the parts fails to provide its service.
My best wishes to you all.I would like someone to expand this topic. Some ideas: if it is not possible to guarantee the feature is implemented, at least we can guarantee it will be worked on. We could also set up some rating system, where developers are rated according to how well they behaved in the past.
As an extreme measure, we could provide legal contracts between donators and implementors! I would not underestimate this possibility.