MAR
29
2005

KDE Desktop Usability Survey

The HASE (Human Aspects of Software Engineering) group at the University of Maryland Baltimore County is conducting two online surveys for KDE users. They expect to obtain results that will lead to useful discussions about the overall usability of KDE. The results of this research will be shared with the KDE community. The surveys take about 15 minutes to complete.

Comments

Who exactly are those people? What previous work did they do? What's their aim?
Quote: "will lead to useful discussions about the overall usability of KDE" -> how do they plan to do that? Are they subscribed to the Usability mailing list? Will they work with other usability groups such as OpenUsability.org which is deeply involved in making KDE better in usability? Will they work with the doc team to make KDE documentation better ( there's a question about that and no need to take a survey to know that docs have to be improved -> that needs people)?
How are the users targeted? Only through the Dot?

Currently, the KDE Usability Team (http://usability.kde.org/) is working with OpenUsability.org (http://openusability.org/). They do their best helping developers by testing programs and writing clear usability reports. I find this approach the best (from the base to global) versus the survey approach: from global to the base.

I disagree with that kind of news being on the Dot without further information (previous work, published results of previous work, timeframe for the survey, suggested help in writing a KDE HIG,...). I hope HASE will clearly answer those questions.


By Anne-Marie Mahfouf at Tue, 2005/03/29 - 6:00am

Why complain against this survey? It's probably just students trying to do an assigment. If you find this survey a wast of your time just don't do it


By chouimat at Tue, 2005/03/29 - 6:00am

Because its a waste of a perfectly good opportunity to do something useful.

As it stands, what they have here is not a usability study, its a popularity contest between KDE and Gnome (they applied there too), asking the enthusiast communities who likes their favourite desktop best.

If they really wanted to tell us someting we didn't already know, they'd sit down with *ordinary* people (the other 90% of people) and find out where the *real* problems lie.

At best, with this, all we're going to find out is who among power users subjectively (!!) prefers copying files most.

As a further sign of quality, they ask gnome users how much they enjoy using the "konqueror" file manager.


By koldpete at Tue, 2005/03/29 - 6:00am

Frankly I am shocked anyone else is interested in KDE.

It bears the thought that with the kind of corporate backing behind GNOME and Windows that 3rd parties would favor those systems exclusively.

KDE is just so insignificant it doesn't deserve the kind of outsider attention GNOME and Windows do. KDE already has its internal posse of snubs who are more than happy to use the desktop and keep it to themselves.

KDE don't need no stinking outsiders.


By ac at Tue, 2005/03/29 - 6:00am

"KDE already has its internal posse of snubs.."

I love this idea, maybe it could be KDE's killer feature - even better than, say DCOP. I can imagine snubs would be short, swarthy and rude - indeed a whole posse of them would be an impressive sight. They would lurk in the dark bowels of the desktop, and only come out at night when you weren't there..


By Richard Dale at Tue, 2005/03/29 - 6:00am

Please excuse, but what are snubs?

(I am not an English native speaker and can't find its meaning)


By Some KDE user at Wed, 2005/03/30 - 6:00am

It's a typo, the guy was actually trying to type 'snobs'.

But I find the typo is actually much funnier than what he intended to say :)


By Richard Dale at Wed, 2005/03/30 - 6:00am

Thanks


By Some KDE user at Wed, 2005/03/30 - 6:00am

The Dot is not the right place to advertise students assignments. Nor are KDE mailing lists.


By Anne-Marie Mahfouf at Tue, 2005/03/29 - 6:00am

Please do some minor research first. The HASE people are:

Dr. A. Ant Ozok, Assistant Professor
Dr. A. Gunes Koru, Assistant Professor

Do *you* have a doctorate?


By ac at Tue, 2005/03/29 - 6:00am

No, I've only got an M.A in comparative linguistics. But this survey is so widely spammed, so imcompetently setup and just so plainly unusable for any serious scientific purpose that I strongly suspect that the actual goal of the survey is not an analysis of the data entered, but an analysis of the reaction of users and developers of KDE and Gnome to yet another survey. I guess Dr. Ozok and Dr. Koru are busy tabulating all the responses on the Dot, OS News and the mailing lists. That, or they are really incompetent, despite their Phd's.

In any case, I've had my fill of surveys... There are two kinds of people: people who think they must fill out surveys because it's their civic duty, like voting, and who feel important because they're asked anyway: the people who offer all the necessary data to let people steal their identity without a blink because they're asked.

And then there are people who survey the surveys and think about it and decide only to cooperate if the survey looks scientific and useful. And otherwise they prefer to work on their projects.


By Boudewijn Rempt at Tue, 2005/03/29 - 6:00am

"All HASE surveys are reviewed and approved by the Institutional Review
Board of UMBC. HASE is not funded by any company, governmental agency,
profit or non-profit organization, except University of Maryland Baltimore
County. HASE has no vested interest in its research studies and findings."

This is as serious as it gets.


By anon at Tue, 2005/03/29 - 6:00am

Did you read the article? Click the HASE link to find out who the people are and what their references are. Their aim is explained in the article.

I understand you have a team doing the same work and so you have a vested interest in the matter, but why does your team have to have a monopoly?


By ac at Tue, 2005/03/29 - 6:00am

What you read is not what you get.
There will be no results, if we're lucky we'll get some kind of global comparison between the 3 desktops KDE, GNOME, Windows.

When I first reacted, I did not even know the same "survey" (I am not sure what a survey really is, scientifically speaking) was done for GNOME and Windows. Hence the KDE survey page is misleading as it implies there'll be some kind of results.

I read the page and I even mailed the HASE team (2 people). They don't know when they'll publish any results nor where.

I just say that we get tons of surveys demand like this and we never see any results. So people are free to do them, no problem, but don't think this can bring something to KDE. And if it does not bring anything to KDE, well, it's kinda useless on the DOt. Leave it for OSNews! Oh they already have it! They appeal to do the GNOME "survey"!


By Anne-Marie Mahfouf at Tue, 2005/03/29 - 6:00am

They said they *will* contribute the results back to KDE. They just don't have a specific date for when the project will be completed yet. You can say the same thing about many KDE projects. You don't have to attack these people like that unless you want to ruin their survey or bias it against KDE for some reason.


By ac at Tue, 2005/03/29 - 6:00am

Those surveys (there is the same for GNOME and Windows) have no scientific basis and no purpose at all. I just wrote the HASE people and they answered me they don't know when they will publish their results nor where.
In clear, those surveys are no help at all for any project as they cannot be called serious work.
I regret that the DOT posted such a misleading article without getting some idea about what they were dealing with. Political surveys for example are conducted on a panel of people. Lately, surveys about Open Source have been numerous as a new area to research. We cannot assess the seriousness of these surveys and usually we never see any results. By posting this on the Dot, at least it should be written that the results will be freely available, when and where. This is the minimum.

A quick survey to finish:
do you think surveys are useful?


By Anne-Marie Mahfouf at Tue, 2005/03/29 - 6:00am

> By posting this on the Dot, at least it should be written that the results will be freely available, when and where. This is the minimum.

why?

If you don't want to do the survey, then don't do it. I didn't do it. I also didn't raise a stink about it. jeez.


By smt at Tue, 2005/03/29 - 6:00am

"at least it should be written that the results will be freely available, when and where"

Yes and they should also fill out a form that allows kde e. V. to sue them if they fail to comply. Come on! It's a credible institution and credible people, who probably know a hell of a lot more about surveys than you'll ever know. I see no point in flaming them.

Another interesting survey would be:
are annma comments on this thread sensible?

(And by the way, off topic - I really respect and appreciate your work in KDE, particularly on KDE-Edu, so thanks!)


By John Usability Freak at Tue, 2005/03/29 - 6:00am

> do you think surveys are useful?
only if people take them.
it accumulates information, so it can't be bad. I took it, and I think it might bring up some interesting details on the user's perspective. So stop moaning please, it's much more unproductive then this survey. You go anywhere and do anything productive instead of whining here, please. Everyone who doesn't like surveys will skip the read; skipping your comments in this thread is much more difficult. So would you please STFU.


By thatguiser at Tue, 2005/03/29 - 6:00am

What I might find interesting is some unrelated-to-KDE results, such as, people who are programmers rarely use the Find functionality, etc. while non-programmers might weight the theming sections a hell of a lot higher.

They could be testing a very bad hypothesis that programmers absolutely need to arrange their icons on a daily basis.

Either way, I'm all for skewing the bell curve - anyone else?


By Troy Unrau at Wed, 2005/03/30 - 6:00am

bell curve skewing is key.


By illogic-al at Wed, 2005/03/30 - 6:00am

Always gotta love it when they repeat questions or ask the opposite of a previous question in surveys :)

However how can they talk about usability when their own survey page is that ugly? They know what to expect but not what to give or how to do it?


By Stephen Leaf at Tue, 2005/03/29 - 6:00am

I hope that you're joking.

It's standard policy of long surveys to ask redundant questions or similarly phrased questions to test for consistency and accuracy.


By frank dux at Tue, 2005/03/29 - 6:00am

I'm pretty confused me right now. blblblblblbbbblllb....11!!1!
ehh, yeah, second occurence (B3 & B6) of the same question I just N/Aed it. I feel guilty now. and, ehhrr .. confused. fffssmnnsssne...


By thatguiser at Tue, 2005/03/29 - 6:00am

You've already been answered, but I cannot help myself. Not only you've shown you know nothing about surveys but also that you didn't even stop to think about it. Yes redundant questions are common in surveys and they have a purpose.


By John Usability Freak at Tue, 2005/03/29 - 6:00am

As an expert, maybe you'd care to enligthen us about what is the purpose having the exact same copy/pasted question, spaced at 3 question interval (B3, B6)? Some references would be nice too.

I can perfectly understand "similar" questions being asked, maybe phrased differently, and/or having a different response pattern (rate good-to-bad vs. agree-disagree) to check consistency in the respondant's answers. "Triangulation" I think it's called. But having the exact same question, I don't see any purpose.

And also I don't think it shows Stephen didn't stop to think about it. In fact, the very fact he posted showed he did notice, thought about it and concluded it was annoying. And I did the same. I'm no survey expert, but I did think about consistenty. So what did I do? I scrolled back up, check my answers and made sure I made the same in the duplicate question. What's my feeling as a respondant: waste of my time. And can you tell me what the surveyer is going to learn from my duplicate answer to his duplicate question?


By Timothee Groleau at Wed, 2005/03/30 - 6:00am

There's a difference between redundant questions and copy & paste duplicated questions, isn't there? In the case of this "survey" the questions were just duplicated, spelling mistakes and all.


By Boudewijn Rempt at Tue, 2005/03/29 - 6:00am

All the time that i read about usability, i remember the only problem i see in KDE: the start menu . And then i remember this is a distribution problem. If you use a distribution that installs a small part of KDE, and provides a menu with less options, you dont have a "menu" problem.
Another problem: the toolbars. Again the distribution can deliver the KDE products with any buttons they want. The same applies to kicker.
The distributions can install just a small part of kcontrol, just some modules. Like kcontrol-light.
Another complain: why KDE has kedit and kwrite ? Complain to your distribution again.

There are many other complains that must be answered by the KDE community. But the distributions can work to deliver a better product to their costumers.

And i tried to answer the survey just to stop in the middle. It is just terrible. The tasks part is an example of non-usability.


By Henrique Marks at Tue, 2005/03/29 - 6:00am

No, it is KDE's problem. I use a distribution that uses vanilla KDE as shipped by the KDE-team. And the Kmenu is overcrowded, the toolbars are cluttered, control center is huge, there are too many menu's in the menubar with too many entries and context-menus are cluttered. Thinking that it's not KDE's problem but distributors problem is a bit wrongsighted IMO. By same token you could say that bugs in KDE are not KDE's problem, since the distributor can choose to fix them themselves.

Blaming the distributors is wrong, plain and simple. Of course the distributors COULD spend time trimming down KDE. And they could spend time fixing KDE's bugs, making it faster, making it look better etc. etc. So they are all distributors problems, and not KDE's problems. So what problems in KDE are KDE's problems? Aren't they all "distributors problems"?

With thinking like that, KDE will be doomed to fail.

I for one think that KDE-team should make releases that kick ass by default. Thinking that "let's ship something and let the distributors fix it" is wrong on so many levels. Note: I do NOT think that KDE-team thinks like that. I think that KDE keeps on getting better all the time. But there's still alot to be done.

"Another complain: why KDE has kedit and kwrite ? Complain to your distribution again."

I choose to complain to KDE which ships with three editors. Thinking that "oh it's not KDE's problem, it's distributors problem" is just plain WRONG. Can you say "pass the buck"? With your thinking, KDE could be as crappy as possible, and it wouldn't be KDE's problem, but distributors problem, since they do not spend their time fixing KDE.

"But the distributions can work to deliver a better product to their costumers."

I'm not a customer, I'm an user and a member of a community (both Gentoo-community and KDE-community). To whom should I complain? To Gentoo-folks? Why? Why not to KDE-folks? After all, Gentoo ships with vanilla-KDE. Why should Gentoo concern themselves with how KDE is designed, isn't that KDE-team's responsibility?

Instead of trying to fix the problems upstream (in the distributions), shouldn't they be fixed at the source instead? That way ALL users and distribution would benefit!


By Janne at Tue, 2005/03/29 - 6:00am

Well, if you're a Gentoo user, you should know that starting with 3.4, the KDE ebuilds have been split up into many packages, so that you only have to install what you want. In other words, you can install kate/kwrite, kedit, both or none, as you see fit. As for why both exist, as far as I know, kedit is there to support bidirectional text rendering, which kate doesn't have yet. Some people need that sort of thing, so getting rid of it would make KDE worse for some people, even if it made it a tiny bit better for you.

Since the split, I've had to do much less menu configuration. The only thing I really have to do is pull some things out of the 'More Applications' groups, because I don't like seeing those.

If, for example, KDE split their big packages into hundreds of small ones to allow for finer grained installation, it would have the following consequences:

1) More work for the KDE developers maintaining hundreds of little packages.
2) More time spent for 99% of Linux distributions who have to do a ./configure on each package, instead of building the monolithic package and creating small packages as necessary from there.
3) Gentoo users might have a little less clutter in their menus.

So I fail to see how we could make an argument for the KDE developers to solve this problem, considering it would cause more problems than it would solve. If you want to pick and mix individual applications from KDE, use a distribution that allows you to do so. Now Gentoo does, but Debian has been doing that for a while. The monolithic packages are primarily intended for distro packagers to use and split up as they see fit (as I understand). Choosing a distribution that ships vanilla, monolithic KDE doesn't grant an excuse to bash this.

I toolbar, menubar, context menu and other such issues should be solved at the KDE level, I agree. However, package management isn't really their area of concern.


By Dolio at Tue, 2005/03/29 - 6:00am

"Well, if you're a Gentoo user, you should know that starting with 3.4, the KDE ebuilds have been split up into many packages, so that you only have to install what you want"

you don't say? And how do split ebuilds (which I use) help solve the problem with cluttered toolbars, cluttered menubars, cluttered Control Center and cluttered context-menus? The split ebuild help you to reduce the number of apps that are installed, but the number of apps is NOT the main reason (it is one of the reasons, however) why KDE is cluttered.

"In other words, you can install kate/kwrite, kedit, both or none, as you see fit."

So it solves ONE problem. And the fact remains that by default KDE offers three text-editors. Maintaining three editors is pointless and wastes resoures. For how long have they planned to drop one of them? even two editors would be one too many IMO. I know I know. Plan is to have one full-featured editor and one simple editor. But we are talking about _text editor_ here! Just how complex can even "full featured" editor be? Hell, when user starts Kate, he should be presented with a blank document and he can start typing away. How is that "complex"? where is the need for "simple" editor?

"Since the split, I've had to do much less menu configuration."

For editing Kmenu, perhaps. But how about toolbars, menubars, contents of the menu's, the hysterical featuritis and the overall "busyness" of the UI? Split ebuilds do not help there at all.

"1) More work for the KDE developers maintaining hundreds of little packages."

They already maintain hundreds of packages, the only difference is that now those small packages are re-packaged in to larger packages. Or do you think that when someone hacks Kmail, he doesn't really hack Kmail, but "kdepim"? And besides. I haven't said that splitting the packages up solves the problem.

"So I fail to see how we could make an argument for the KDE developers to solve this problem, considering it would cause more problems than it would solve."

What on EARTH are you blabbering about? I talked about how KDE has cluttered toolbars, cluttered menu's, too many menus and the like. And you start talking about how KDE is packaged. Uh, the way KDE is packaged has nothing to do with how cluttered the UI is! It helps with the number of apps, but even there it doesn't solve the problem, it merely goes around it. Instead of having KDE fixing it, the task is left to the user.

"However, package management isn't really their area of concern."

Then why did you start talking about it? I sure as hell didn't start talking about it. I did mention that KDE ships with too many apps (in this case: text-editors), and it's true. But that's not really about how KDE is packaged.


By Janne at Tue, 2005/03/29 - 6:00am

Hi Janne,

The fact that you say that there is too many editors
just show that you are actually TROLLing.

Why people just don't stop working on vim, emacs, nano, pico, ed
and all starts working on SuperDuperTextEditor ?

Obviously, that's way too many editors ship by default from Linux distro!!!

--> because people have various divergent preferences!
and there's nothing wrong with that.
Nobody is eating the same food or drinking the same thing.
Developpers won't start dropping support for those editors
to work on YOUR prefered editor.
People want choices.
More importantly they want BETTER choices.
Obviously, the MOST POPULAR choice will prevail over less popular ones.

Even on Windows, you have at least three editors
by 'default' and nobody complains about:
Notepad, Wordpad and Microsoft Word

plus maybe two more:

FrontPage Express and MS-DOS Editor

plus maybe thousands more specialized one:

Visual Studio, Notepad+, EditPlus, CuteHTML, Understand, Eclipse, etc.

They simply serve different purposes.

Now, to go back to your argument, the best way to 'solve' this
would be to have one KVeryRichText Widget and using it
with various configuration on different kinds of editors,
[load or not syntax highlighting, etc],
so it can be easily reused in:
KWrite, Kate, KDevelop, Kalc, KMySuperEditor, etc.

So, that features can be *shared* around, while people can tweak
their menus/toolbars/status/shortcuts to their own taste.

Now, of course, usability can be increased for default KDE apps
and there's nothing wrong about that.

Currently, it's some sort of time consuming 'trial/error' process
with unfortunately a laggy/incomplete feedback from the user community.

We are getting there, it takes time...

The openusability.org project is there to help out
and the kde-usability mailing list is also there for just that,
please feel free to contribute with your own feedback.


By Stop trolling at Tue, 2005/03/29 - 6:00am

One nice feature of KDevelop is that it doesn't have its own text editor. By default it uses a Kate KPart, but you can replace that with a vim clone KPart if you prefer. One vim clone is about to be retired from the CVS, but I believe there is another one being actively developed. There was a story about it a few weeks ago here on kde dot news - I've forgotten the name of it.

Both KWrite and Kate use the same KPart, so they aren't really different editors. It allows uses with simpler non-programming requirements to use KWrite, and they don't have to see a load of programmer specific options in Kate.


By Richard Dale at Tue, 2005/03/29 - 6:00am

> There was a story about it a few weeks ago here on kde dot news - I've forgotten the name of it.

kyzis - http://www.yzis.org/


By Christian Loose at Tue, 2005/03/29 - 6:00am

"The fact that you say that there is too many editors
just show that you are actually TROLLing."

I said that KDE has cluttered toolbars, menubars, menus, Control Center and the like. I also mentioned that KDE shipping with three text-editors is rather pointless. I was then made to believe that the split ebuilds that Gentoo offers, solve the problems. Well, they _partially_ solve ONE of the problems, and it does so in only Gentoo! the other problems (the more important problems IMO) are left unsolved.

I find it really surprising that when I offer valid criticism (and I think my comments were valid), people like you decide that I must be trolling. So, users can't voice their opinions nor can they offer any possible solutions. If they do, they are "trolling"? Yes, I think KDE ships with too many text-editors. How exactly is that "trolling"?

"Why people just don't stop working on vim, emacs, nano, pico, ed
and all starts working on SuperDuperTextEditor?"

That has NOTHING to do with my arguments. I NEVER said that "developers should stop working on and work on instead!"

"because people have various divergent preferences!
and there's nothing wrong with that."

Then KDE should not ship with three editors, but with dozen editors? Seriously, your whole post misses my point completely, and it's just full of pointless rambling.

"Developpers won't start dropping support for those editors
to work on YOUR prefered editor."

And I never said that they should! Sheesh!

I'm not gung-ho about text-editors. I simply think that it's pointless to have a full-featured editor and a crippled-editor, just for the sake of "simplicity". Why not instead have the full-featured editor, that is also easy to use? Kate should be like so that user can easily start editing text with it (isn't that the whole idea of having "simple editor"?), but it would also be capable of doing more advanced stuff (like it does today). Having those possibilities in one editor does not mean that the editor is too complex for basic editing.

But, let me clarify on my point: Kedit, Kwrite and Kate are text-editors. Their purpose is to edit text. One of the has an important feature that the other two lack (BiDi-support), and that is why it's needed. But why three? Apparently so that we have full-featured editor and simple editor. But really: what's the point? Kate (the full-featured editor) can be simple editor as well. We are talking about editing text here! Just how complicated can it be? But hey, maybe KDE should ship with "dead-simple-editor", "mediocre editor", "full-featured-editor", "The cool-looking-editor", "Boring-looking-editor", mouse-controlled-editor" and "keyboard-controlled-editor"? I mean, we all have different preferences, right? Yes we do, but KDE itself should offer one editor. Do we REALLY NEED three (or two) _text editors_? Maybe KDE should ship with Krusader alongside Konqueror, since the two offer different approaches to filemanagement? And maybe KDE should ship with dozen mediaplayers, because someone might like the "Yet another mediaplyer 2".

One text-editor would be enough. If the user wants something different, he's free to install third-party editor. Hell, even the apps you listed do not ship together, no, they are offered separately. Yes, Windows ships with two editor (notepad and Wordpad), but does KDE have to do something because "Windows does it as well!"?

You seem to think that I'm advocating the "do not code any more text-editors! Let's focus on just one!". I'm not. I think the developers should be free to write dozens of text-editors if they so desire. What I'm saying that KDE should only _ship_ with one. And since Kate is the best of the editors (espesially when it gets BiDi), it would be the obvious choice.

NOTE: Just because KDE would only ship with one editor, the user would not be denied the right to choose. If he doesn't like the editor in KDE, he would be free to install any other editor he wishes to use. And there are plenty to choose from!

Really, KDE should have one app for one category. One text-editor, one music-player, one video-player, one filemanager, one calculator etc. etc. But, for some reason KDE ships with THREE text-editors! Yes, Bidi is important, but how long will it take to implement it in Kate?


By Janne at Tue, 2005/03/29 - 6:00am

> "Developpers won't start dropping support for those editors
to work on YOUR prefered editor."
>
> And I never said that they should! Sheesh!

Do you actually read what you're writing?

Quote: "Maintaining three editors is pointless and wastes resoures."
Quote: "Really, KDE should have one app for one category."

Seems to me like you have no clue how FOSS works.


By ac at Tue, 2005/03/29 - 6:00am

"Do you actually read what you're writing?

Quote: "Maintaining three editors is pointless and wastes resoures."
Quote: "Really, KDE should have one app for one category.""

Seriously: the developers can freely write zillion text-editors if they so choose. I merely question the reasons of shipping them in KDE. Should KDE include every possible app just because some user might find it usable?

Yes, maintaining three separate text-editors does waste resources. But if the developers want to do it, so be it. But I find it rather weird that same people (more or less) who maintain one text-editor also maintain a separate yet very similar text-editor. If the developer wants to do it, go right ahead! But that doesn't mean that KDE should ship with all of them.

"Seems to me like you have no clue how FOSS works."

I know very well how FOSS works, thankyouverymuch. I know all about "scratching an itch", and I support it wholeheartedly. But I do not think that they should ship in KDE by default.

Seems to me that you have no idea what I'm trying to say here.


By Janne at Tue, 2005/03/29 - 6:00am

> NOTE: Just because KDE would only ship with one editor, the user would not be denied the right to choose. If he doesn't like the editor in KDE, he would be free to install any other editor he wishes to use. And there are plenty to choose from!

If you don't like/need KEdit or KWrite, then don't install their packages.

I sometimes just need a simple text editor that starts instantly. Kate can not provide that because of its feature-richness and that's is okay. As a developer I do need those features and then I can wait a few more ms.

> But, for some reason KDE ships with THREE text-editors!

No, your distribution ships three text editors. KDE just provides source code packages.


By Christian Loose at Tue, 2005/03/29 - 6:00am

"If you don't like/need KEdit or KWrite, then don't install their packages. "

Again: that goes around the problem, it doesn't fix the problem. Maybe I'm just crazy when I would like to see the problem fixed, instead of simply avoided.

"I sometimes just need a simple text editor that starts instantly."

On my machine Kate does starts instantly.

"No, your distribution ships three text editors. KDE just provides source code packages."

No, KDE ships with three editors. My distributor offer me what KDE releases. I can try to go around the problem by specifying what apps I install, but the fact remains that by default KDE ships with three text-editors. That fact doesn't change no matter how much you try to twist around.

really: download KDE from kde.org. (rather: one of the mirrors). See how many editors are included. THIS is the KDE I'm talking about!


By Janne at Tue, 2005/03/29 - 6:00am

Kate and kwrite are not really two editors, they share the same editing component and offer different GUI around it, the kate one offering more features than the kwrite one.
AFAIK it is planned to remove kedit for KDE 4, since then katepart is planned to have all the features kedit has (AFAIK it's currently a limitation in Qt).
Then KDE will ship with let's say 1.5 editors :-)

Alex


By aleXXX at Tue, 2005/03/29 - 6:00am

"Kate and kwrite are not really two editors, they share the same editing component and offer different GUI around it, the kate one offering more features than the kwrite one."

I know that, and it's a good showcase for the underlying KDE-technology. But as far as end-user is concerned, the two are different apps.


By Janne at Tue, 2005/03/29 - 6:00am

>But as far as end-user is concerned, the two are different apps.
Exactly, they are different apps, but intended for different usage patterns. It's like a carpenters toolbox, he may have both a hammer and a nailgun. Two tools doing the same thing, but it is the nature of the job who decides which tool is the best to use.


By Morty at Tue, 2005/03/29 - 6:00am

"Exactly, they are different apps, but intended for different usage patterns."

But we are talking about _text editing_ here. Should KDE also have several filemanagers as well? Maybe Krusader alongside Konqueror, since it too is inteded for different usage-patterns? And maybe KDE should also include text-based web-browser besides Konqueror, since it too would be targeted at "different usage-pattern". Of course we would also need text-based mail-client, several different kind of media-players etc. etc. since they would all be targeted at "different usage-patterns".

"Two tools doing the same thing, but it is the nature of the job who decides which tool is the best to use."

So you do think that if there are more than one way of doing certain task, there should be separate app for each way? So we do then need several filemanagers, several mail-clients, several mediaplayers...

I for one think that KDE should offer one way of doing things (and text-editing isn't exactly rocket-science), and if the user really wants an alternative way of doing things, he could freely install a third-party app (and there are plenty of those). Many people prefer Krusader for filemanagement instead of konqueror, but KDE doesn't ship with Krusader. But now, all of a sudden there is a need for several _text editors_? And the only difference between the two is that one of them is crippled, whereas other one is not.

This is madness.


By Janne at Wed, 2005/03/30 - 6:00am

>talking about _text editing_ here.
We are not talking about the rather trivial task of changing and adding characters to a text based file, it's the process of working with those files which is the important part and the reason for different tools. It's a fairly well known concept and not hard to get at all.

In fact there are in reality 5 (4 if you don't need RTL languages) text editors in KDE, with your thinking you have to count KDevelop and Quanta too.

As for your filemangar example, you can configure Konqueror in many different ways depending on how you want to use it. And you always have the commandline. And there are already "one shot", "advanced with playlist" and "jukebox like" mediaplayers for the different ways to work with media. For your mail reading needs you can already chose between KMail standalone or the Kontact suite. It's all about the right tool for the job.


By Morty at Wed, 2005/03/30 - 6:00am

"We are not talking about the rather trivial task of changing and adding characters to a text based file, it's the process of working with those files which is the important part and the reason for different tools. It's a fairly well known concept and not hard to get at all."

In the end, text-editing is creating, replacing and deleting text. Nothing more, nothing less. And like I have repeatedly said, I REALLY fail to see the point in having several tools for that. I mean, we have three tools doing the exact same thing! Saying that "but they are meant for different styles of work" or something like that is IMO quite stupid. If we though like that, EVERY app in KDE should be redundant with several different apps each doing the exact same thing in slightly different way! Why is KDE afraid of saying "in this system, this thing is done like this by defaula. If you want to do it in some other way, feel free to install another app to do it".

What do you think KDE would look like if it had several mail-clients, several PIM's, several filemanagers, several browser, several media-players, several terminals, several different versions of same Solitaire, several IM's etc. etc. Crazy, right? then why is several text-editors an absolute-necessity, whereas those other things are not? Is it because KDE has traditionally shipped with one browser (for example), whereas there has been three editors for a while now? A case of "this is the way it has been done, and this is the way it will be done in the future as well!"? I don't see the difference between several editors and several IM's (for example).

"As for your filemangar example, you can configure Konqueror in many different ways depending on how you want to use it."

Krusader is still quite different thank Konqueror, maybe we whould add it there as well? Can't you see this is getting ridiculous? The "different styles of working" is just an excuse people use trying to justify the current system. If KDE shipped with just text-editor, and they now tried to introduce another text-editor, I bet you would be thinking "Huh? What's the point, I already have Kate?".

"And there are already "one shot", "advanced with playlist" and "jukebox like" mediaplayers for the different ways to work with media."

And do we need all of them? Really? Why not simplify the system? Again: people seem to be afraid of saying "this is what we offer. But feel free to install another thing instead, if you are not happy with these". Instead, we have a case where we are told "we have this. But if you don't like it, we also have this, this, this, this and this as well! They all do the exact same thing, but they do it slighly differently". Again: Madness.

"It's all about the right tool for the job."

And I REALLY fail to see how that applies to "full-featured editor" and "crippled editor". Really, Kate is not that difficult to use when compared to Kedit (or was it Kwrite), I really, REALLY fail to see the point in having them both available!

"For your mail reading needs you can already chose between KMail standalone or the Kontact suite."

If I were asked, Kontact would be the default client in KDE. But the individual apps could also be available, but maybe not displayed in the Kmenu (the advanced users who want Kmail, instead of Kontact can add it there by themselves). And no, having both Kontact and Kmail does not go against what I have told about redundant apps. Kontact is collection of different apps, one of which is Kmail.

"Right tool for the job" is a great euphenism for "we can't decide what to focus on and what to offer you, so we offer it all and let you work it our yourself!". It causes confusion and bloatness.

maybe some people don't like Kicker, but one of the Kicker-replacements instead? Maybe they should be shipped with KDE as well?


By Janne at Wed, 2005/03/30 - 6:00am

Well editing text files is not as straitforward as you might think.

For instance there is a need fo no nonsense "notepad" style text editr. It should be as simple as it can be, Just Open save exit. No font changing, MYBE, MAYBE some syntax highlighting etc(witch should be turned off by default).

Then you have Programmers text editor(Kate). You want syntax highliting you want line nubers, code folding, you want code completition you want context sensitive help, you want shell integration, debugger friendly etc. You also want good integration with build sytems code control systems, If we are taling about interpeted languages (php, perl, python, ruby) you want integration with interpreter etc. Plus you want plugins. You shouldnt even compare Kate to KWrite its more of KDeveloper type of programm.

This is not just another way of doing things, this is diferent task altogether. That might share a few similarities but thats it.

Its like saying hey i have bike licence why do I need truck licence as well, after all both are licenses to operate motor vehicles


By Luka at Wed, 2005/03/30 - 6:00am

Spot On(TM)!!!

>MAYBE, MAYBE some syntax highlighting etc(witch should be turned off by default).

No way! :-) lets keep KWrite as trimmed as at any way possible. It should have next to no features, and load in a microsecond ;-)

>You shouldnt even compare Kate to KWrite its more of KDeveloper type of program.

Full ACK.

The description "Advanced Text Editor" scares off people who dont need it, but is makes sense to have it as quite a large portion who uses UNIX are.. well.. advanced! This must be acknowledged and acted in accordance with. The day KDE becomes "moms-and-dads-only" kind of desktop, is the day where we will se a fork of the project!

And to you Janne: Which editor should be left behind? The one that you prefere?... or the one which is used the most hours a day in total? (kate for sure), or the one that the most can figure out? (KWrite fir sure)

Cheers!

~Macavity


By Macavity at Wed, 2005/03/30 - 6:00am

> Windows ships with two editor (notepad and Wordpad)

Wordpad is not really your plain text editor. Yes, Wordpad can write in plain text format, but so can Word. Wordpad is more like a light word processor.

By the way, I totally agree with your points.


By blacksheep at Tue, 2005/03/29 - 6:00am

> Then why did you start talking about it? I sure as hell didn't start talking about it.

Actually, you *did*. You said: "And the Kmenu is overcrowded,[..]" and Dolio explained that this is a problem of distributions that ship the original cvs modules as is.

> I talked about how KDE has cluttered toolbars, cluttered menu's, too many menus and the like.

You know that there are very different types of users. To some the toolbars are cluttered. Others want as much toolbar buttons as they can get. KDE can only try to ship sane defaults and make those customizable.

I'm not saying that the current defaults are perfect, but I agree with Henrique that the distributors could help with the situation. They know what users they are targeting. They are in direct contact with the users. So why don't they change the defaults, so that the settings fit the users?

IMHO the distributors are missing a chance here.


By ac at Tue, 2005/03/29 - 6:00am

"Actually, you *did*. You said: "And the Kmenu is overcrowded,[..]" and Dolio explained that this is a problem of distributions that ship the original cvs modules as is."

I did not complain about how KDE was packaged, I complained about Kmenu and number of redundant apps. Saying that since it's possible to install separate apps on Gentoo solves the problem is wrong IMO. It goes around the problem (on Gentoo), but it does not fix the core-issue. Instead of trying to cure the symptom, why not cure the reason?

Besides it is not distributors problem as such. They offer what KDE offers them. And besides, alot could be done to clean up the Kmenu. Why have separate entry for "Editors"? Why was Control Center moved from "Settings" to the root of the menu (creating yet another entry in the menu)? Besides re-organizign the menu, the number of apps COULD be limited. and no, splitting the packages up to individual apps is not a REAL solution to the problem. It merely passes the responsibility to others and tries to work around the problem. it doesn't solve the problem.

"You know that there are very different types of users. To some the toolbars are cluttered. Others want as much toolbar buttons as they can get. KDE can only try to ship sane defaults and make those customizable."

Are menubars customisable? Are the contents of the menus customisable? With Kiosk-tool, some of them are AFAIK. But that sort of editing is a bit too much to ask.

I do not think that there are any users who would like to have their UI cluttered with zillion buttons and menus. There ARE users who want lots of features at their disposal. But having lots of features doesn't have to mean that the UI must be cluttered. I'm not advocating for removal of features.

"IMHO the distributors are missing a chance here."

IMHO KDE is missing a chance here. Instead of trying to pass the responsibility to distrobutors, why not try to fix the issues right in KDE? Hell, if we pass the responsibility to distros, then the users would never see KDE in the way it's developers intented it to be seen. And that would IMO be rather sad, since KDE IS kick-ass desktop. Instead of having "KDE", there would be "SUSE Desktop" or something like that.

I for one do not want distribution-branded desktop, I like my KDE as KDE. That way I will have consistent desktop when I move between different distros (should that happen) and I get to enjoy KDE the way it's developers meant it to be. And it allows me to offer improvements to KDE-folks, since my KDE would be vanilla-KDE and not distro-modified KDE. Since the two are different, my suggestions might not apply to vanilla-KDE.


By Janne at Tue, 2005/03/29 - 6:00am

Pages