Quickies: amaroK, KDE-People Newspage, IP Telephony, Qt 3.3.2, KMail, Usability Blogging

Some time ago on kde-www there was a request
for volunteers
to help with the amaroK website.
The amaroK team is considering to switch to a CMS based site so that they
get more organized with their content.

When talking of websites, it looks
like the website about those people behind KDE is also
offering a newspage these days.

Are you a Qt programmer and looking for job? Well, it seems that
Skype, wants to build their P2P telephony program for
the Linux desktop when you look at their job openings.

We have some more Qt news: Trolltech
announced a new version of its
C++ multiplatform toolkit Qt 3.3.2 which is a bugfix release.

No programmer yourself but you want specific improvements in KDE?
I came across this nice idea. Here you can demand
improvements specifically for KMail with an amount of money you would like to pledge for it. It seems like there
are some recent pledges as well.


Jono Bacon
, a writer, consultant, web developer and musician, has written an entry in
his blog that discusses the direction of software projects, with a
particular reference to KDE. He is interested to hear any thoughts on these suggestions.

Dot Categories: 

Comments

by golan (not verified)

Something is wrong in the article! the title states Qt 3.3.2 while the article body Qt 3.2.2 ;-)

by Fabrice Mous (not verified)

Thanks ... I changed it.

:)

Fab

by Jono (not verified)

Hi all,

I am pleased that some discussion is occuring from this blog entry. As far as I am concerned the issue is very real and it should be discussed. The mere fact that it keeps getting discussed means there is an issue. Although some may feel my blog entry was a poke at KDE, it was intended this way - it is intended to identify the problem and try to resolve it if the community so decides. I don't subscribe to this KDE vs GNOME argument, I believe in free software and I would love to see both projects succeed.

From the responses I have seen so far, I have heard from a lot of people defending the current method of making everything configurable. Yes, it is cool in the fact that you can find things when you need to change them, but this configurability is (in my opinion) losing chunks of users. The problem is that people, such as my dad, who are not computer savvy will have problems with KDE. Richard Dale mentioned the fact that I have not actually put him in front of KDE, well no, I havent, but I have put countless others in front of KDE. this includes members of my family, my girlfriend, collegues and others. The unifying factor among these test candidates was a relative confusion when wanting to change certain things in the desktop. Things such as the background wallpaper were not an issue, but other issues were confusing. I remember when my workmate had a go with KDE he was a bit overcome with the amount of categories in the KDE Control Center. This is a fairly IT savvy person and he was a little confused; hence my example of my dad who would be totally lost.

The point of this blog entry was to not rib KDE but to see if there is an alternative method that pleases the dads of the world and the readers of this site. KDE is a very simply to use desktop; there is no doubt about that, but some options do seem rather advanced for a desktop that aims for a novice userbase. The key seems to be choosing a usability scheme where by someone can access their options easily and efficiently but not be confused. This is why the different userlevel idea initially sprung to mind. Other options such as Advanced buttons for further pages of options, GConf keys, and other options could be discussed.

I do feel that this issue is of paramount importance. It may not be important to those who love KDE for what it is now and are happy with it, but if KDE is to access a broader novice userbase (and they are getting more and more with the increased popularity of Linux), the current system needs to be better established.

I think that as new features are added to KDE and options are added to the GUI, the desktop can only take a certain amount of populated options before the new scheme needs to be considered. If Konq has 17 pages of options, what would be the considered maximum limit? In KDE 6.2 are we going to have 37 pages of options?

If we can discuss this and come up with a suitable plan early on, these should help further problems down the line.

by Richard Dale (not verified)

"The problem is that people, such as my dad, who are not computer savvy will have problems with KDE. Richard Dale mentioned the fact that I have not actually put him in front of KDE, well no, I havent"

Sorry about my harsh comments, but I really feel this is a setup/distribution packaging problem rather than a Qt/KDE usability issue per se.

But the burning question is "can some KDE expert set up a suitable configuration for the Jono Bacon's father audience?". Or is there something wrong in that we can't hide the complexity of KDE even though some people might need all of it? Is there something wrong with Kiosk mode for the minimalist users?

-- Richard

by Anon (not verified)

> "can some KDE expert set up a suitable configuration for the Jono Bacon's father audience?"

Yes, but shouldn't dads ideally be able to setup their environment without being computer experts? Think stupid user with notebook on the road: Constnt need to change network settings, proxy etc.

>Or is there something wrong in that we can't hide the complexity of KDE even though some people might need all of it?

I do think so.

>Is there something wrong with Kiosk mode for the minimalist users

Kiosk mode is great, but it is more suited for Jono to setup for his dad thaan for his dad to do it by himself as it might have even more options than the normal configuration. Remember: This is not about making it impossible to change settings for some users, this debate is about how to make it EASIER for them to adapt their environment to their needs.

Kiosk is more of a security framework.

by Datschge (not verified)

KDE is easy to adapt to specific needs, in contrary to other environments which either don't document specific features or lack them completely. And if KDE can be adapted to specific use cases why should Jono's dad not be able to find a service provider who offers KDE perfectly customized to his needs?

Kiosk being a security framework and thus ignoring it is a poor excuse.

by Jono (not verified)

"Sorry about my harsh comments, but I really feel this is a setup/distribution packaging problem rather than a Qt/KDE usability issue per se."

Don't apologise, it is good to chew the fat and discuss this. :)

I don't think it is a packaging issue. The reason I say this is that certain options cannot be easily taken out of KDE by the distributor. Yes, they can remove certain KControl modules if needed, but I don't see how this should be at the hands of the distributor. I see how they should not include certains software, and I also see how the distributor should ensure that X is working, but in terms of how KDE is configured, that is the responsibility of KDE.

I honestly believe that KDE can be made to work for the kind of person my dad his. He is not an unintelligent man, but he is a person who gets confused when too much is put in front of him on his computer. I think the key point is that a hacker who knows the ins and outs of KDE can make it do what they want; people who are newbies such as my dad and Eruic Raymonds Aunt Tillie are basically stuck with the choices decided by the KDE team and the distributor. If those choices are unsuitable, a bad usability case occurs.

The most difficult step is acknowledging that there is an issue. If people accept that this configuration mess will only get worse as more configuraton options are added, we can discuss alternatives.

Jono

by Richard Dale (not verified)

I'm sorry, you need to put a minimilist configuration in front of your father, and then comment. You can't base a usability study on hyphothetically thinking about someone "who gets confused", without trying it out in the real world. What about scientific method and peer review?

And Eric Raymond knows next to nothing about computer usability.

-- Richard

by fetchmailconf-fan (not verified)

@Richard:

### And Eric Raymond knows next to nothing about computer usability.###

You dare! Please refrain from such useless polemic. Eric wrote one of the greatest GUIs ever. It is in his famous "fetchmailconf" program. That is a classic. My aunt Tillie uses it every day, and she really really loves it. (Dunno about Eric's aunt....). See here for an example:

--> http://www.linux-mag.com/2003-08/img2/power_01.jpg <--

by Roberto Alsina (not verified)

Ok, now, that was mean. The UI, I mean, not your comment!

by . (not verified)

But nothing beats IC Station from Mentor Gparphics.
Have a look at:
http://www.ece.unh.edu/courses/ece715/assign/LAB4.html

As I remember it, the menue on the right hand side depends on the menu at the bottom. True horror.

Another smart thing is that when you want to print your circuit, you can specify how many layers you want to print, ranging from 1 to 99. If you choose 99, the print job will not be finished before the next morning. Of course 99 is the default. The true horror is that the widget shows only one 9. You have to click on the widget to see that it is actually 99.

Mark my words, never ever let an electronic engineer construct a GUI.

by will (not verified)

This is hardly an authorative statement, but I believe that use of minimalist design is well supported by studies, and is generally an accepted empirical fact. I don't know these studies (I don't work in usablity) but it is mentioned as a principle in Sun's Gnome usability study (with some relevant responses from the test subjects as well) and commented on by Joel's.

It makes sense that it is - if the options gets too plentyful, the user must with necessity use more time to read them and process information. The point of minimalism is not really minimalism for its own sake, but actually a derived feature of attempting to make user interaction efficient. The most efficient interaction is to offer the most relevant choices to the users so that he can do what he wants with the least amount of effort...

by Datschge (not verified)

To make user interaction efficient you need to know how the user will interact. When you do that the wrong way by considering only one kind of users you won't make user interaction efficient but far more inefficient for other users, for them not being able to find the way he expected, regardless of how illogical you might find his way of thought.

Minimalist design is well supported by studies in specific test cases with specific users. KDE has a broad range of uses and a broad range of different users. KDE can be adapted to every case, but you will fail crashing and burning if you seriously think you can adapt a minimalist design to the default KDE without decreasing the range of uses and thus alienating users.

by Aaron J. Seigo (not verified)

that does boil it down to a rather linear line of thought, and like most simplistic generalizations it fails to capture the whole picture.

yes, simplicity can be a very good thing.
so is consistency.
and utility.
and learnability.
and use-case based ordering of elements.
and ...

there's a lot more to "usable" than "simple". for instance, Raskin now regrets going minimalist with the single button mouse on the orginal Mac. he wishes he'd used (at least) a two button mouse and just labelled the two buttons: Select and Activate (or something along those lines... working from memory here ;)

so as you can see, minimalism is NOT the end-all, be-all solution to things. it's a tool like any other: good when used appropriately, and detrimental when missapplied.

discussions about software minimalism are often framed in absolutes, which is unfortunate, because that fails to do either minimalism or usability a service.

you don't nail screws in with a hammer, and similarly we shouldn't minimalize the desktop into restrictive discomfort and lessened utility.

by Richard Dale (not verified)

Aunt Tilly walks into the reception area of a legal firm. She opens her iBook and it wakes up from sleep mode straight away. It finds the WiFi network and uses rendevous to find the printer in the reception, and automatically adds it to the printer menu. She prints out the notes for the meeting without needing to even think about the technical issues involved. There were no so called 'friendly wizards' involved in doing this.

Or she plugs her digital camera into her computer, it does something useful such as start a photo album management program. Plug and play - that's useability.

Both the above examples are absolutely nothing to do with 'software minimalism' and are much more to do with sound software architecture and infrastructure.

by Datschge (not verified)

I'd personally be scared about the implied lack of privacy necessary for your first example to work without any additional need for authentication or fine grained configuration of what should be accessibly for who in what cases, both which will make your 'usable' vision not satisfying in one or another regard, but whatever.

by Richard Dale (not verified)

Well, if Aunt Tilley had visited before maybe she has the WiFi key on her keychain (like KWallet), and the network uses WEP. Someone would have to tell her the key, so she could add it with the keychain dialog.

Or perhaps they have an 'open access' WiFi area just for reception area visitors where you can print or browse the internet or send emails. However, they wouldn't be able to access the full corporate network.

This isn't a vision - it's something you can do today. If a printer can send it's configuration via rendevous, you obviously don't need a printer dialog. If every device can do that you can get rid of a lot of dialogs.

I'm really glad that the KDE project doesn't have Sun working on their UI guidlines. They have absolutely no aesthetic/design sense for UIs at all. Swing? Open Look?

by will (not verified)

To Aaron and Richard: minimalism is only a rule of thumb, and only works if done right. Nor is not the only part of usablity. But it is one part, and its a good rule of thumb, which is more than sufficient to take it seriously.

by em (not verified)

> From the responses I have seen so far, I have heard from a lot of people
> defending the current method of making everything configurable. Yes, it is
> cool in the fact that you can find things when you need to change them, but
> this configurability is (in my opinion) losing chunks of users.

Well, *I* don't think it's losing us chunks of users. The kind of user who would be confused by an abundance of configuration options is the kind of user who isn't interested in configuring anything and would do so very rarely. Anyway, my criticism is that you are advocating what seems to be a quite drastic change in KDE policy because of your impression of what a certain kind of user might think in an imaginary case. It seems to me you are succumbing to GNOME's propaganda of "configuration options are eeeeevil", specially when you link it with the comment that "the sense of direction in KDE seems to be flagging somewhat".

by Jono (not verified)

"It seems to me you are succumbing to GNOME's propaganda of "configuration options are eeeeevil", specially when you link it with the comment that "the sense of direction in KDE seems to be flagging somewhat"."

I have not succumbed to any kind of propoganda, less propoganda that I never been in contact with.

This has nothing to do with KDE/GNOME comparisons. GNOME do good stuff and KDE do good stuff. The point I was making is that KDE seems to have remained fairly consistent in its development and there is a lot in there that I feel is clutter. I am simply concerned that this clutter is going to be overwhelming at somepoint.

There is plenty wrong with GNOME, and I accept that. There is plenty wrong with lots of things in thge free software world, but there is plenty things that are right. I was merely pushing out my view in the hope that it causes some fresh discussion.

As far as I am concerned, if KDE choose that my suggestions are not in the interests of its users, then that is the decision and I am happy. What I would not be happy with though is if I shut up and did not make suggestions and expressed concerns with software that I care about. I care about KDE and want to see it become the dominant desktop platform for free software; I am just concerned about the amount of configuration clutter that is going to reach critical mass in a few versions if something is not done soon.

Jono

by Datschge (not verified)

Usability depends on the use cases and the kind of user. Whatever KDE installation you would have used for your dad you should have ensured that it's adapted to his needs, not to those of yourself or someone else. If your dad tells you to buy a car for him you won't buy him a truck after all. Now why do you expect that this adaption to your dad's use case needs to be done in KDE as a whole? KDE doesn't know who your dad is, what he would do with a computer, and why everyone should use a KDE setup adapted to your dad's use case. KDE however did and does collect and realize requests from many real users, and the KDE you see today is a result of that, highly adaptable, but you may not dare to complain that it's not adapted to a specific use case if you didn't (let) do that adaption before.

by Jono (not verified)

Oops, bad typo:

This:
Although some may feel my blog entry was a poke at KDE, it was intended this way - it is intended to identify the problem and try to resolve it if the community so decides.

Should be:
Although some may feel my blog entry was a poke at KDE, it was not intended this way - it is intended to identify the problem and try to resolve it if the community so decides.

It was NOT intended as a poke at KDE.

What a cock up!!

Jono

by reihal (not verified)

The next time this diskussion pops up, I won't read it.
I'll just close my eyes and scroll. Ha, take that!

As mosfet and others have said in the past; users don't want less options,
users want more options arranged in an orderly manner.
I am a user and thats what I want.
EOD.

by gerd (not verified)

The article was unnecessary.
KDe is very good in usability while Gnome still lacks features. In europe desktop Linux means KDE. Teh control center was bad, but not anymore. What the Author suggested will be implemented, it's already planned to provide a Kontrolcenter light.

When you want an "easy" desktop I would rather suggest to build a web based desktop system (who need windows and menus?), a kind of browser based operating system that is 100% configured via web pages. For special programs like terminals and editors and word processors tabs are opened and the rest looks like tabbed browsing. No WIMP Windows moving, one window-one document. And you can get every single option via its URL.

by David (not verified)

To get a desktop out into the world that can be used by many people is very difficult - impossible, quite frankly.

Try putting Gnome in front of a sys admin, a developer or a person who has used a computer within a business environment over many years. These people really do matter, but you don't necessarily hear about them. They will throw the damn thing straight out of the nearest window (no pun intended). Where I work, that's a long drop.

A desktop that Dad or Gran can use instantly is currently a complete myth. Sorry, but Jono Bacon has been at the acid on this one. He mentioned that his Dad never touched the Control Panel - and yet it needs to be there. Why? Windows, Mac OS, KDE, Gnome - none of them have managed this (and it isn't really the aim of either Windows or Mac OS), and they will never manage it until we get some unbelievable technology that can communicate with a user on a telepathic level. "Where we're going we don't need - desktops." This is why the Gnome direction is totally and utterly flawed. This is why people criticize Windows usability when compared with Mac OS. If Mac OS got used as widely within networks, by sys admins and developers as much as Windows did, it would look much, much different.

"I feel the best option in a situation like this is to make the desktop and configuration available in a series of schemas. You could have an Easy schema that has the most common options available and removes choices that may be too advanced or are not needed by someone such as my dad. You can then have a Medium schema that satisfies those with a tweaking finger."

In theory yes, but the best you can hope for here is an "Advanced..." tab or button, or to reorganize things. Having levels of easy, intermediate... etc. has actually been discussed on the KDE usability list - and dismissed for very good reasons. One is that in the search for simplicity (not usability) you make things much more complicated :).

"As an example, if you don't use Konqueror, you should be able to remove its options from your schema."

How is using a schema (an extra layer of complexity) making things more usable? It is also a nightmare when administering users as you then have to guess what schema they are in. Sorry - a non-starter.

As someone who currently has to design and maintain applications within a very sizeable banking environment I can tell you that Gnome's philosophy is total bollocks. These people have never designed a well-used application in their lives. "Look guys, it says this in the text book. It's so simple." Er, no - I don't think even an IT undergraduate is that naive. We have flame-wars about usability here all the time, but none of us believes that you can design a desktop for the absolute lowest common denominator - and get lower. Simplification is not usability, and Gnome seems to have got the two woefully mixed up. Remember that many, many people out there will need to actually get things done with their desktop.

"GNOME has had two important advantages: 1. That the most influential people in the project has the right vision and 2. The support of big corporations, like Sun, who knows this because live off selling their stuff."

Yer, whatever.

1. Gnome 'usability' vision is totally flawed, because they have lost sight of what usability actually is. What are the functional requirements of Gnome and the applications? You have to pay attention to this - you can't just decree that something will be 'usable'.
2. I read an article the other day about how Gnome was supported by Sun, HP, Novell and IBM. That was in the year 2000, and the perceived corporate support has got them absolutely nowhere. I'm still waiting. Sun do not live off selling this stuff, and it is quite clear to anyone that Gnome's usability drive did not emerge from common sense.

"One possiblity is to make a style/usablity fork which branches off as stable release and has the opportunity to remove, rename, redesign under a unified vision without compromise."

..."vich vill be obeyed visout quvestion" (Fawlty Towers). KDE is not a dictatorship, thank goodness. Gnome is an open source project, isn't it?

Enough.

by Patrick (not verified)

You write that "the usability vision of Gnome is totally flawed, because they have lost sight of what usability actually is".

I wonder whether you read the studies and documentation written by Sun and published by the Gnome Usability project (http://developer.gnome.org/projects/gup/). In the published study they clearly state which tasks users had to complete and what went wrong when users where performing these tasks with the Gnome tools.
1. To have a fruitful discussion please point out what is wrong with this study.

Besides that, the page lists several documents that show why it is important to have good usability.
2. Did you read the Introduction of http://developer.gnome.org/projects/gup/hig/? 3. Do you want to contest any of this?

Your post does not seem informed to me, please prove me wrong.

I can only concur with the idea that some advanced uses and less used options go to an 'advanced' page. Although I use KDE most of my Linux time, because it still provides me with more functionality (not necessarily options), the look and easy of use of Gnome really please me.

by Derek Kite (not verified)

> still provides me with more functionality (not necessarily options), the look and easy of use of Gnome really please me.

Maybe that is the point after all. One can have either, but not both.

Think about this: Apple gained a large group of users when they added a Unix shell.

Derek (studies are another word for 'believe me even if I don't make sense')

by Patrick (not verified)

"Maybe that is the point after all. One can have either, but not both."

I don't believe that. I think you can have a pleasing and productive environment and still have lots of options. Just look at Windows XP and work with it for a week (I don't know Mac OS X that well).

As far as Windows XP is concerned I believe Microsoft did an amazing job. Although I feel it could have been better and easier (1. configuration options for the OS are scattered through menu's and panels and available through many different menu's; 2. cryptic commands are needed to access often used information etc) they have a productive GUI that most users easily grasp and know how to adjust to their own preference.

Furthermore they have been able to make installing third party programs so easy that almost anybody can learn how to do it. Just try to explain to a new user how to install a download tgz and you know what I mean.

The latter is not something that is easy fixable by the KDE project. What *is* in our hands is to make the environment and it's native programs friendly enough that users can make them selves comfortable with them and feel they are in control.

This will make it possible for distributors to integrate the underlying OS with KDE (or Gnome) and have the same complete user experience that Mac OS X and Windows XP offer.

I therefore agree with Jono.

by David (not verified)

There's actually not too much wrong with Gnome's HIG. It is the way in which it has seemingly been implemented as some sort of holy grail. No where in the Gnome HIG does it say "We must simplify everything so we can get our Grans to use Gnome." Within Gnome, usability has become something altogether different.

"http://developer.gnome.org/projects/gup/hig/....http://developer.gnome.o..."

Yes it is all positive stuff, but the real world tends to be quite a bit different from studies.

"What a disappointment it should be when a user's ability to access one of the features we have coded is impaired or altogether halted because they don't understand how to manipulate the interface..."

That is a great statement, but it says nothing about trying to simplify something that cannot be simplified or slashing features.

The short answer is no - I don't want to contest any of this. I contest the way it has been implemented.

"Your post does not seem informed to me, please prove me wrong."

Err, real world experience?

"Although I use KDE most of my Linux time, because it still provides me with more functionality (not necessarily options), the look and easy of use of Gnome really please me."

That's great, but if Gnome or KDE are going to make it in the world they will need to be used for a multitude of tasks and be able to do them pretty well. That is an altogether different proposition than someone coming along and saying "Oh, it looks clean and polished and easy to use..." or "My Gran can use this..."

by Patrick (not verified)

"That's great, but if Gnome or KDE are going to make it in the world they will need to be used for a multitude of tasks and be able to do them pretty well."

I believe this is not a point of discussion but more a prerequisite for each of them to succeed . But my point is that just adding features and not making them easy to use will still not get you there. 'Simplification' and 'feature completeness' need to go hand in hand in some way.

I'm not saying that Jono's proposition of user levels is the way to go, because I really am not that much into user interface design. I do feel however that he has a valid point by saying that the user interface needs to be simplified.

To my opinion following the Gnome HIG is better than something KDE has now. Why not make it a vote on the Dot or start a more broad discussion about this on a mailing list?

by David (not verified)

What kind of new CMS site are the new amaroK people contemplating? Does the offer they have for fast, free web space dictate a particular technology? My preference on this would be a Zope based site, simply because Zope make this kind of thing incredibly easy and the security isn't bad either. Yes I know what people say about Zope and Plone, but they are good at what they do.

by Mark Kretschmann (not verified)

We're not set on any particular CMS. Most important would be a way to get our content in shape easily and reliably. Then we'd like to have:

1) Forum (a usable one)
2) Gallery (screenshots, developer's ugly mugshots and the like)
3) News

Also nice would be an additional wiki part? Dunno if that mixes well with a CMS, but I like the idea. You know, making the site a little dynamic, allowing users to participate. But what do I know, I'm not much of a webmaster :)

--
Mark.

by mitchy (not verified)

I love Plone, and think that they are the 'Jimi Hendrix of CMS' - incredible features waaaaay ahead of the competition.

The Zope environment was too heavy for my tastes however, and I switched to something much more lightweight, but still dead simple to use: Mambo (http://www.mamboserver.com). A healthy third-party community creating custom add-ons was also a big seller for me - templates, shopping carts, whatever I needed was either built-in or a download away.

I looked at others, but picked Mambo because my wife can use it with maybe 3 minutes of instruction. The only other CMS that I have worked with that was this easy was Plone.

Oh, I liked Mambo enough to get into the core development team, so I am now horribly biased. :) I would happily volunteer the time to setup a Mambo site for them and show them what Mambo can do for them.

BTW an awesome site for checking out PHP/MySQL-based CMS is:

http://www.opensourcecms.com

HTH,

-- Mitch

by Mark Kretschmann (not verified)

That's a great offer :) Mambo looks very powerful indeed.

Could you maybe subscribe to our mailing list and send a short note? I reckon it makes communication easier. You can find it here:

http://lists.sourceforge.net/lists/listinfo/amarok-devel

--
Mark.

by David (not verified)

Yer, I know what people say about Zope - and they're right. Mambo does look very good, and something lightweight so that could be a great option. It sort of depends on how much load your system will be taking. If it isn't great Zope/Plone is a good option.

I hope something great can be done here.

by CAPSLOCK2000 (not verified)

Let me plug www.tikiwiki.org . It's a great system, with lots of features and very easy to alter/extend. Amongst other modules it features an image gallery and a very nice Wiki.

by Aaron J. Seigo (not verified)

Jono: good to see your face around again... here's to seeing you on IRC again some day soon =) some thoughts on your blog, ranging the spectrum of randomosity:

- user levels don't work. they've been tried and they've failed. many times over. the reasons for this failure are rather well understood. google for them if you care to.

- KConfigXT is proving to be a path forward away from having to either put every option in the GUI or else have it hidden.

- we already have something resembling schemas in Konq: user profiles. they have now extended to include the XMLUI settings which allows for much greater definition of specific profiles, some of which will be in 3.3. i wouldn't be surprised if it eventually was extended also include the config dialog such that there was a "File Manager" version and a "Web Browsing" version (version as in list of panels) shown depending on the type of profile in use.

- you say that we won't get anywhere if we are only self-congratulatory, and i agree strongly. i don't think such attitudes are the "mode de e'mploy" in many developer circles where the work actually goes on, though. fan treatises, enthusiasm, straigh-up marketing, etc may skip over much of this hard-biting self-examination but it exists in all the successful projects i've seen, to one degree or another

- you say that constructive criticism is one of the biggest gifts you can give to a project. if only that were true we'd have all the problems solved. the world is awash with constructive criticism; it has become like water to a drowning man who needs not more of the stuff but a way to deal with it. right now we have a veritable Pacific Ocean of constructive criticism. i'd be overjoyed if there was even a Mediterranean of time and effort that accompanied it. ;-)

- i don't think that the direction in KDE has been flagging, unless "flagging" means something different in the UK than it does here. ;-) it's been evolving and growing, if anything. you are right that becoming feature rich creates new challenges, but i'm not sure i see that as a challenge related to KDE's direction.

- large numbers of different people will report on the same problem(s) in a relatively short time span (usually with remarkably few useful/well researched solutions proferred) each with same raging enthusiasm for the topic as the next. and yet when a satisfying conclusion is eventually arrived at, very few make any bones about the success: it's right on to the next Overwhelmingly Alarming Issue. silence when the war goes well, i suppose. at least this process keeps the trolls stocked with soundbites ;-)

- another observation one might make is that it's often damned if you do, damned if you don't. when there aren't features, it isn't ready for prime time, but when features abound there are too many of them. ;-) when you have lots of exposed options, you're unusable; when you don't expose them, you're inflexible. there's something to be said for ballance, and that ballance is a tricky thing to achieve. people on both sides of those fences often urge their "team" to head towards the other side of the fence post haste: most "what KDE should do" articles i read suggest aping some other project; indeed most "what GNOME should do" articles do the same. and this comes from people who purport not to like the projects that they may as well be describing in their prescriptive diatribes. greener grass? heh...

- it may be easy to dismiss such articles as being self-serving mouth wagging. but i do believe most people who offer their opinions on these things are well meaning and trying to encourage and help in their own way and as their time and energy permits.

- and through it all the developers largely remain positive and glean the useful points out of these things so as to improve the beautiful children of their minds.

- and through it all the cheerleaders strive to maintain the visibility of the positive things while attempting to discover "creative solutions" to the perceived problems.

- and through it all the users keep using the software. and writing about it. and even from time to time giving something back.

holy crap is this "comment" long. at least i found something to do with the last *looks at clock* 30 minutes @work ;-)

by huu (not verified)

http://news.netcraft.com/archives/2004/04/28/interview_with_miguel_de_ic...

CUT from the interview:

Q. Has Novell decided what form the desktop will take - will it be Gnome-based, or have other elements?

A. Gnome and KDE are basically the shells, but then there are higher-level applications like the office suite. We're making the decision it's going to be OpenOffice, the browser it's going to be Mozilla, the email client it's going to be Evolution, the IM client it's going to be Gaim. So we basically have to pick successful open source projects and put them together. There's a lot of work on integrating.

by huu (not verified)

Forgot to add a comment. Why does Novell let these idiots speak out in public?

by David (not verified)

Well it is amusing that he says nothing about Gnome and shifts the emphasis away from it. Until I hear a statement from Suse, or look at what is actually in their software (they lean very much towards KDE, use Konqueror by default, they do not have Evolution installed as the default mail client and the IM client in Suse 9.1 is definitely Kopete), I don't believe anything Miguel de Icaza or Nat Friedman says.

At the moment it has become clear that Migua de Icaza and Nat Friedman are doing some political shuffling because of what is happening at Novell/Suse. The soundbites are flowing out to convince us that things are heading in a certain direction, as they have so often done.

I don't have any time for what they say quite frankly.

Anybody know what's happened to the kdevelop.org website? I haven't been able to connect to it for three days now.