In the latest issue of KDE-CVS-Digest, read about new Kontact plugins for summary, notes and the newsticker,
KOffice gains improved import and export filters plus template loading from the command line, we also see an improvement in speed for Konqueror file and image viewing, and Dr Konqi gets hooked into KDevelop for debugging. Also, improvements to KDE Print,
KGhostview, user interface cleanups, and numerous bug fixes.
Dot Categories:
Comments
>HIG is meant to be followed by everyone, including developers who may think >otherwise.
Better do it the KDE way. If the uses wishes it the other way he has the right to do it so.
>GNOME apps actually look very consistent
RClick on the taskbar on some app. For most apps, nothing happens. For some apps, a context menu pops up. Very rare is the opportunity to move the taskbars like KDE users are used to. Gnome simply lacks the powerful framework of KDE and it is written in a wrong language for GUI stuff whilst this is the domain of OO paradigm. Viewing widgets and windows as objects is simply natural and leads to clean code and the derivation mechanism allows quick and safe extensions of existing work done by others. Looking at Gtk its quite amusing how desperate they try to map the OO paradigm in C.
> The goal of KDE is to empower the user.
There comes a point when this becomes ridiculous though. Right click on the desktop (KDE 3.1.0) and you are given the choice, to align the icons vertically or horizontally or to the grid? What the hell? Who needs this kind of flexibility? It just slows you down because you end up looking at the choices thinking, why would I possibly want to choose between the three? Is there something I don't understand about using the desktop where this could become useful?
What would be a killer feature would be a grid like system that could be drawn on the desktop with intersecting horizontal and vertical lines with movable nodes where the lines intersect. Exactly like what is used in morphing software. This could be another power-to-the-user KControl entry. Then the user could choose how many lines to draw and then move the horizontal lines up and down, with variable spacing between the lines, and move the vertical lines left and right, with variable spacing between the lines, and move the intersecting nodes wherever they like. The user could even have different layouts for different desktops, or different purposes. Now only keep the option that reads Align to grid, and drop the other two options (align vertically / horizontally). By default, a newbie will not know about this funtionality and that is fine because the icons will simply snap to the square grid that he is used to in Windows. On the other hand, the power user will also be happy as the icons now snap to whatever shape he has created for himself.
In one move, you have both given power to the user and rid KDE of "bloat". This is what I understand by people complaining of there are too many options in KDE. Simplify the many options into less options while giving power to the user at the same time.
Imagine a circular, or a wavy grid that runs across the screen. No-one offers this. This would put KDE ahead of everyone in terms of this particular feature.
> Little to no impact on improving KDE can come from Gnome as Gnome is not or not openly developed as KDE.
I disagree. We can learn how Ximian has learnt what works from MS land. There are many carbon-copied features of XP that show up in XD2. This in my opinion is a good thing if the features are good features. Was that not the idea behind KDE, to take the best features from the different OS's, incorporate them and then improve them? Ximian is doing this right now.
One thing that KDE implemented from MS land which they shouldn't of is the dotted-line bounding box around selected icons, as well as default buttons. This "feature" appeared in 95/98/(Me?) but was removed from 2K and XP.
MS land: If I want a file on the desktop to lose focus, I can click on the desktop. Now the desktop has focus, so the file that was selected appears completely unselected. That makes perfect sense, doesn't it?
Now go to KDE or 95/98/Me land: If you do the same thing the bounding box remains, making the file look like it is still somehow selected. This is poor UI design, which MS thankfully fixed with 2K. Another side effect of the bounding box is the total lag you feel when you move an icon quickly on the desktop. The bounding box tries to play catch-up with the moving icon and is left behind. If the bounding box was removed, this lag would also disappear.
The bounding box on default buttons, IMHO is simply ugly. It should be removed for pure aesthetic reasons. Using a lighter colour on the default button, for instance, would be great as it could instantly be standard across all the different available styles.
> CAPS is not appreciated either and never has been the indication of person capable or willing to help
I don't think that this is a fair comment. The more you love something, the more you express emotion about it.
I feel that he was simply trying to get that level of emotion across.
I prefer KDE to Gnome, and I want KDE to outshine any DE available, regardless of the platform. To do this, we need people with passion. Without passion, how can you get exited about anything?
VL User
Regarding the choices for icon placement on the desktop: For me in KDE 3.1.2 it's all in a neatly sorted submenu called "icons" so it doesn't feel like bloat at all.
Your idea to let the power user define his own arbitrary order in which the icons are aligned is nice, but I'm sure quite some people will scream about how KDE's setting library Control Center just got even more "bloated" due to that. In any case the one who really needs it is welcomed to implement it (or pay someone to do so).
KDE never has been about copying from other systems, in many cases the developers use only KDE and implement whatever useful feature they need or other people request. It's funny how many people come again and again comparing KDE to other systems expecting everyone eg. to install XD2 or buy Windows Server 2003, asking someone to do this is simply ridiculous. Instead describe what's needed and why it's needed and how it fits into KDE.
The bounding box which apears around one icon on the desktop or one button in your current application is the cursor which you can control with your keyboard in case you prefer that over a mouse. Removing it removes a necessary cursor indicator for using KDE's GUI with a keyboard. (Are you sure that got removed from 2K and XP?)
As for caps: Caps are commonly considered as screaming, which again is considered as highly impolite. So if you want to make a really bad impression regardless of what you want to say write everything in caps. Also emotions and passion are nice and dandy, but please refrain from inflicting that on others too much, you might just annoy everyone with it when always taken in overdoses.
I think he thing about arbitrary alignment was meant as an extreme example of how customizability can become stupid. But then again, I think you got that. Just to be sure.
I don't think this is a very good example however. Practically every desktop, from classic MacOS (since around ~6.0-ish) to Windows (since 95) has had this set of features. Even GNOME 1.4 had it. Only GNOME 2.0-2.2 are missing the align to grid feature. There was a patch for Nautilus floating around that added align to grid. I think it's there in GNOME 2.3.2, though.
It's hardly an extreme example. In fact, it's a good example of why gnome2.x pisses me off so much compared to practically everything else in existance :)
No i meant the part where he said there should be an option for a power user to be able to define his own grid and arbitrary angles and stuff to that effect. Align to grid is perfect by me.
I think you misunderstood my post :)
I was using this as a serious example of how we can simultaneously remove options and increase power at the same time.
Where would we remove options? Remove Align Vertically and Align Horizontally from the desktop context-menu.
Where would we add power? Add a completely customisable Grid KControl entry.
How does this achieve both objectives simultaneously? Less options on the desktop context-menu, and Align to Grid now has two meanings:
1. Align to a square grid - which is what the newbie is expecting.
2. Align to -any- shape grid - which is what the power user is expecting.
This will never be an issue for a newbie as it will work as he expects. He can then later "upgrade" his skills by defining a grid shape when becomes more proficient with KDE.
Here is another example. When I move an icon on the desktop, such as a file and drop it onto a folder a pop-up menu asks me, do I want to copy or move the file? I have to answer this choice every time I do this. This makes it painful to move files in this manner and annoying that KDE tries to hold my hand.
Why can't we rather have default actions such as the following:
Standard drag-and-drop = move file to target folder
Alt+ drag-and-drop = copy file to target folder
Ctrl + drag-and-drop = create a shortcut in the target folder
shift + drag-and-drop = another action.
Window-key + drag-and-drop = another action.
... you get the picture.
Then have a separate list of default actions for other types of icons:
For example: Drop an Ogg file onto a Konq. link:
Standard drag-and-drop = Launch Konq. and play in the media side-bar.
Alt+ drag-and-drop = Launch Konq. with gg:filetype.
trl + drag-and-drop = Launch Konq. with Konq. help on file type.
shift + drag-and-drop = Launch Konq. with another action.
Window-key + drag-and-drop = Launch Konq. with another action.
... I'm just sowing the seed, please modify this into a workable solution :)
MS deals with this by letting you hold an icon and letting you press the Alt key, and a corresponding overlay icon will appear to show you what action is going to be taken. If you are still holding the icon and press the Shift key, followed by the Alt key, the overlay icon will change. In a few seconds you have now learnt which keys are associated with which action.
We can learn these little tips from the other OS's.
VL User
>Regarding the choices for icon placement ...
A submenu just for icon placement? Sounds like bloat to me :)
As long as Align to Grid is toggleable, then I am okay with that. In KDE 3.1.0, the icons only align once. If you subsequently move an icon, you have to align to grid again.
> Your idea to let the power user define his own ...
I understand you concern about having another KControl entry, it could simply be added to an existing entry.
> KDE never has been about copying ...
I disagree with you here. Many KDE features are carbon copies of MS, such as the find dialog. The MS find dialog was horrid, but KDE copied it exactly.
> The bounding box ...
My statement was correct, MS did remove the bounding box from icons in 2K/XP. It did not remove the bounding box from buttons and html links for that matter. This was a separate point that I made pointing out that I thought it was an ugly looking solution in general. In 2K land, if a desktop icon is highlit and you click on the desktop the icon loses focus and -looks- unselected. If you then press up or down on the arrow keys, the correct upper or lower icon is selected, if it exists. The last selected icon is remembered, even though it looks totally unselected when it loses focus. This is a sane way to deal with the nasty looking bounding box.
My point about bounding box lag when icons are moved is another reason to kill this behaviour.
> As for caps: ...
Agreed.
VL User
> My point about bounding box lag when icons are moved is another reason to kill this behaviour.
Please check out whether the following patch helps http://lists.kde.org/?l=kde-optimize&m=104463123606728
> One thing that KDE implemented from MS land which they shouldn't of is the dotted-line bounding box around selected icons, as well as default buttons. This "feature" appeared in 95/98/(Me?) but was removed from 2K and XP. If I want a file on the desktop to lose focus, I can click on the desktop. Now the desktop has focus, so the file that was selected appears completely unselected. That makes perfect sense, doesn't it?
I think that this feature should be removed personally. In reality, what KDE is doing is drawing the focus rect on the icon with the current keyboard focus. You can press the up key or whatever to go to the next icon. However, you're right. It does appear like a leftover selection, and is confusing to users. I myself was confused the first time I saw this-- the only reason I knew what it was because I've made QStyles and I recognized the focus rect.
> The bounding box on default buttons, IMHO is simply ugly. It should be removed for pure aesthetic reasons. Using a lighter colour on the default button, for instance, would be great as it could instantly be standard across all the different available styles.
This really depends on the widget style in use. Widget styles can pretty much do whatever they want in differentating the default button from regular buttons, including what you mentioned. I do think that it really depends on the widget style whether the default button box looks good or not. In some, like kde 2.x's web style, I love it, and in other styles, I don't.
> Imagine a circular, or a wavy grid that runs across the screen. No-one offers this. This would put KDE ahead of everyone in terms of this particular feature.
Ah yes, the famous staggered grid from classic MacOS. Thanks for reminding me- now on my ever growing todo list.
Please stand in your seat once KDE 3.2 arrives. That little submenu is going to get extended by one.
> Ah yes, the famous staggered grid from classic MacOS
I have never owned a Mac, so I wasn't trying to compare this to a Mac :)
Does the Mac really offer the flexibility I was talking about, eg. variable grid lines with movable intersecting nodes?
Thank you for adding it to your todo list :)
VL User
>> One thing that KDE implemented from MS land which they shouldn't of is the dotted-line bounding box around selected icons, as well as default buttons. This "feature" appeared in 95/98/(Me?) but was removed from 2K and XP.
Actually, it isn't removed from 2K and XP..
It's just hidden away and can be enabled or disabled somewhere in that right-click-on-the-desktop-dialog (you know, with the desktop settings in it).
When you use the mouse to select things (ie. widgets, icons..) the focus rectangles (dotted lines) will stay away (not only from widgets, but also from those icons, including icons on the desktop).
When you start using the keyboard (I believe it happens after pressing ALT) you'll see the dotted lines everywhere again.
Argh, you are right. Sometimes the icon is highlit without the bounding box, which looks great and other times if you tab through to the desktop it draws the bounding box.
My apologies for the misleading statements.
I still would like it scraped from KDE though.
VL User
I don't think it should be removed.
While I don't like the focus rectangles because they're terrible to look at, they're great for usability.
KDE just needs (yet another) option to hide these things ;)
(possibly in Windows style using ALT activation)
What is XD2? I did a google search but it took me to a web design company. Is it something to do with enlightenment?
http://www.ximian.com/products/desktop/ - is there something like usability for marketing? :-)
What is XD2? I did a google search but it took me to a web design company. Is it something to do with enlightenment?
Not to mention XD2 breaks KDE when you install it.
You shall become the third person I correct about menu editing in XD2 today. You can do it in two ways. The first, somewhat akin to Windows, is to use Nautilus. You can add and remove launchers from the directory heirarchy and the menu will be upated. The second is to right click in the menu and you have an option ot add/remove/edit menu entries very directly. Enough for me. Maybe not for you though. It seems some people still prefer the separate app for menu editing.
I think KDE does deserve a polish job too. You should see some of the stuff on gnomesupport.org (Do not worry, they do not flame KDE there). Just read for interests sake and see what they are talking about. I thik it makes sense. There is no need for KDE to be less powerful to get good defaults and get rid of a lot of maybe unnecesary stuff.
And agreed with you on XD2. They should actually look at it to get an idea on how to polish KDE. Not that XD2 is perfect though. I have already noticed a number of issues, and some were pointed out to me. But it is GOOD.
"I can not edit the menu easily" yes, it can be edited, I already knew that, but it is not as easy or as powerful as in KDE.
Also I agree a KDE company like Ximian would rock I guess TheKompany is kind of like that but they are so much smaller, don't make things free =p and are focused and doinga great job for the zaurus.
Though, if they get our support they will grow. But, when you have Quanta Plus, why would you want to downgrade to quanta gold apart from the better support for more platforms (yes quanta can run on other platforms too, but it is not as easy or integrated)
TheKompany is quite different from Ximian...TheKompany focuses on cross platform products -- that's where the money is.
TheKompany used to focus mostly on KDE applications. However, they were tied less to KDE than Ximian is to GNOME. This is because no KDE core developer or founding (as in, no developer who started working on KDE before 1997 or so), was ever employed by tkc. Ximian, on the other hand, was founded by gnome-core hackers and founding developers.
Tkc pretty much went where the money is-- very few companies can make commercial success from ONLY developing for the Linux desktop. I think the rapid bankruptcy of Eazel proved this. Ximian has pretty much stayed in the Linux desktop market by being smart. They've secured a lot of partnerships, and focused on primarily enterprise applications (where bling bling $$$ is)
I'm not sure if KDE needs a Ximian though. It'd of course be nice to have some core KDE developers employed by a company that cares only about KDE, but on the other hand, KDE is developed in a much more of a Bazaar-environment than GNOME historically has. It's worked for nearly six years folks-- sometimes it's better just not to change.
Good website. A couple of reviews of XD2:
"The impression I get so far from the reviews of XD2 is that its kind of rushed and buggy in places. I'll be sticking with stock debian gnome, I think - theres no way I'm installing Ximian GNOME after the way it trashed my system when I installed the previous version."
"Actually compared to a review I read just before XD2 came out, this is a pretty fair review. I've always loved Ximian stuff, but there really are more problems here than I expected. I use apt-rpm and synaptic, and my dependencies were just destroyed by the Ximian install. As the reviewer pointed out, most of the icons in your menus are gone, which gives the desktop a bad feel. Without a doubt, there's a lot of nice additions, but there's so many things I'm having problems with, I'm not sure it's worth it. (One other poster on the review's website noted how much OpenOffice Writer Ximian version crashes -- I can't even use it.) This is disappointing to me because I've waited eagerly for this for a year and half. And now I'm considering reinstalling RedHat 9 because everything works better, the dependences aren't all messed up. To put up with some of the problems I'm having, I feel like the upside should be MUCH higher than it's turning out to be. I think Ximian is a victim of their own hype.
While this is certainly just more "opinion", when my friends who look to me as their Linux guinea pig ask me if they should install Ximian Desktop 2, I really can't say yes without adding all kinds of qualifications. :-( I'm used to saying: "Hell Yes!""
You know, this has been on my mind all week...is it perhaps time for KDE members to form a Ximian-like company to charge forward into the marketplace? The efforts of SuSE, Connectiva, etc. are wonderful (I am a long-time SuSE user), but as Linus himself said, from now on the exciting stuff will happen on the desktop, and the distros still aren't there, focusing on the desktop like Ximian is. I've been working on a document outlining my thoughts on this issue (and considering if perhaps we could even do it). Anybody interested in discussing this drop me a line. I think the KDE community is stronger than just about any other, but *it is my opinion* that for some reason we are still too passive about gaining marketshare. I made a laundry-list of KDE's strengths, and it is amazing. It is hard to understand why people would want to use anything else (some areas need work, yes, but the potential is there!!).
KDE truly is an end-to-end solution, covering everything from desktop utilities, Internet applications, office suite components, to groupware (and all native!) I'm not yet confident to contribute code (I'm still a student...maybe soon), however I'm excited about getting involved in other ways (documentation, usability studies, training, promotion, financial contributions). Yet I wonder if the KDE project can continue to scale as it is as a volunteer-driven meritocracy...while I'm sure some people would disagree, I think most would support a commercial venture of KDE developers, designers, and advocates incorporating. Many people I talk to like the idea of KDE, but are afraid to take the plunge because there is no business behind it, providing corporate support, training, boxed packages & documentation, etc. Look how much more success you can have giving your boss a boxed copy of Redhat 9 with a 90-day support certificate vs. a burned CD-ROM. This isn't to downplay the contributions of the Linux distributions, but to ask instead "Can KDE stand on its own?" Can it's community continue to scale with it's success?
I'd say your approach is a little bit naive. Ximian came through on the trailing edge of the dotcom boom. It was quite a feather in Nat Friedman's cap. Also it was $30 million in venture capital. That changes everything, and in both good and bad ways. So right up front we need to say that a relatively equal entity needs a rather large pile of cash. All other things being equal, do you have that much money laying around?
Currently there are several small companies like Matthias Kalle Dalheimer's Klarälvdalens Datakonsult AB which is involved in major enhancements to KDE's email capabilities and also recently hired David Faure. My company sponsors Andras Mantia and recently two individuals stepped up for some notable sponsorship levels of their own. There are also organized groups associated with KDE with the specific mission to to get more people exposed.
The example of handing your boss a CD vs a copy of RH with 90 days of service seems like a good example of public misconception. One of the first things they teach you in customer service schools is that the way to develop customer loyalty is to have something small broken that you can handle effectively for a customer. This exposure creates more loyalty than not having a problem. Sometimes I think RH has learned this too well with the odd bugs that always seem to be only on RH. Recently my partner asked me if we might try to offer enhanced service to web shops running Quanta. We actually had a contributor make a feature request for code shortcuts which we delivered to him in less than four hours. The funny thing is that right now users can email our user list and have an answer anywhere from minutes to hours, which usually substantially beats not only commercial software but my experience with Linux vendors. Problems I've used my free customer support with have taken days and been less than satsifactory in my mind. Direct to developer contact has consistently been better and more satsifying.
As for a KDE distribution I use one... it's also a GNOME distribution. Gentoo rocks! There are lots of great things about it... it leaves KDE as the developers intended, it allows me to get speed by optimizing for my hardware and it gets me off the upgrade treadmill with CDs and RPMs. Gentoo now has sponsors supporting it's development, as does Quanta.
The reality is that other models besides the Ximian model are currently producing many good things. Many small consultants will spring up to service the marketplace with KDE. There are a lot of strategic avenues to work with. Unless you have deep pockets and years of solid business experience you are looking for piles of trouble trying to emulate the Ximian model. Frankly I don't see a need for it.
That is an interesting perspective. And quite frankly, if I was going to produce such an effort, you would be recruited for PR, as you do very well at promoting Quanta and communicating with the users. Still, I'm not sure that the question is necessarily naive. Do you think it has to start on such a large scale as Ximian? Do you think such a large project such as KDE *can* sustain itself indefinately without some kind of direct commercial backing? I don't know of too many other projects that are (especially ones that have such a direct impact on the experience of the end-user). Look at our own Qt! It is much more successful than gtk+, partly because it has commercial motivations and obligations. The markets help to drive it's success.
Not every business venture starts with venture capital! There are definately way to start from the roots up. I'm not directly arguing your point, but I am a believer that people who come together with a goal in mind can achieve those results with the right effort and dedication. Should we be more aggressive as a communtiy? Should we instead just let things happen as they will, and praise success brought to us by contracts won by SuSE, Klarälvdalens Datakonsult AB, etc.? The problem I see with relying on Redhat, Mandrake, SuSE, etc. to distribute KDE is that outside of the few paid developers, server space, conference underwriting, etc (and yes, this all is very important and appreciated), KDE is just another set of packages, like Apache, GNOME, and Perl. I suppose I hope for a more synergistic relationship, versus a symbiotic one. Building on your example of having a simpler product to produce solutions with and provide quicker support on, chances are a commercial KDE distributor can produce enhancements and fixes faster and more effectively than the big distros because the desktop is *all they focus on*. Right?
As far as Gentoo, yeah, it definately is an excellent project. Their focus is not on deeply enhancing the desktop however. Again, KDE is just another set of provided binaries. They ultimately rely on the changes we make. Perhaps other models can be explored, but I don't see how Ximian is necessarily unique...every company that packages free software, adds value, provides support/training/etc., is doing essentially the same thing. I'm all for examining other avenues, but I think the dialog is healthy for the growth of our community.
P.S. - that Quanta fix works great!
Eric mentioned it, but I want to emphasize the small consultants role here. This is the business model I understand, and like.
Ximian, Suse, Redhat etc. are software houses in the old tradition. Package a solution, put it in a box. They have their place, and I think we will see more and more from Suse with support for kde. Their expertise is generally towards a retail product. This perspective is valuable and has brought the desktop to where it is.
Consultants focus is on solutions. The german ministry needed a groupware solution, it was put together for them using existing kde technology plus a bunch of new stuff that will be integrated. Another consultant will take that work, extend it further, solve another problem, etc. etc. They don't depend on a package for their income, they depend on fees, 'this much to solve that problem'. I suspect much of the linux server side stuff is produced by and for that market, and probably the desktop will be the same. And it will accelerate as the desktop becomes more and more viable.
One thing I know is it takes serious guts to take on a project like that. Some necks are sticking way out. A successful outcome will embolden others to try the same.
Derek
Replying to myself, but there is a commit today from Zack saying that AutomatiX GmbH is funding him to develop Koffice, particularly Kexi, a database type app.
Check AutomatiX on google, then translate. They seem to have vertical apps based on QT, and are consultants.
Obviously they see value in KDE, and are willing and able to help finish some apps that are lacking.
And good for Zack. Eating is somewhat habit forming.
Derek
i want a magazin.plz send me in this my address.
rathindra das
vill+p/s astagram
dist;kishoregonj
bangladesh
> That is an interesting perspective. And quite frankly, if I was going to produce such an effort, you would be recruited for PR
Assuming you could get me. ;-) I'm extremely independent and don't take direction unless your credentials are so impressive I just want to be in the same room with you. Besides... who says I don't have my own ideas about extending my current efforts? Adding programs we maintain and getting corporate sponsorships and contracts for application development are natural steps for our current position which we have already established a solid ability to perform in.
> Still, I'm not sure that the question is necessarily naive.
Well, consider that aside from programming I've also worked in marketing, done some consulting and owned a number of businesses in the last 25 years. To my way of thinking anything that isn't a tightly defined pro forma that details not only how it works but how it limits my exposure and keeps me out of the soup line it sort of a ticking time bomb. You should feel okay because my lack of naivety has resulted from being vigorously beat about the head and wallet more than a few times over the years. ;-)
> Do you think it has to start on such a large scale as Ximian?
Absolutely not. It's just that was your reference so It's difficult to imagine you meant something other than what you referred to. If it's that vague it's very difficult to debate the particulars. My policy is to avoid borrowing money... However my financial planner has told me that even a botched venture capital start on that level could easily leave the recipient set for life with some creative bookkeepping. ;-)
> Do you think such a large project such as KDE *can* sustain itself indefinately without some kind of direct commercial backing?
Well, let's look. It started in 1996 and over the years there have been a few developers sponsored to work on it and some equipment and hosting services as well as some financial donations. To my knowlege I sponsor one of three full time active developers, at least referencing a comment by David Faure. Currently KDE has the majority of Linux desktops and is highly thought of. I'd say depending on your definition of "direct commercial backing" you could make arguments either way. So far we're going on 7 years with limited support. Clearly as opportunities become more available there are a number of people with the skills to work the model.
As Derek pointed out small consultants can make a big difference. Also a solid proposition to commercial entities to contribute to extend F/OSS. The advancement beyond what is already being done, and in my mind shoudl not be overestimated, will be by small shops as it should be. Also KDE established KDE eV and the KDE League for the very reason of promoting KDE. If you've read much of what I've written you know that my plan involves giving web developers as a group of early adopters the incentive to move en masse to KDE because of Quanta. Patience can be a virtue. If anyone has learned that it's me.
FWIW I think that so far Ximian seems to be doing okay with a very difficult business model, but over all they run huge risks because any model that can use software for a product can be a liabliity. To me software is a tool, not a product. Software retailers have to make decisions based on retailing their product where as contractors produce software for specific fees, usually for vertical markets but possibly for more general use and to be released free. Anyone who retails packaged software can make a lot of money off of it but they have certain overheads so they can lose it. If they encounter decisions between quality and cash flow you may not like the product. So by definition contracting software must be done on a smaller scale because it is more personal. Any large scale company producing software has to have a means of sustaining it's self. Ximian was fortunate to have millions of dollars that allow it to use a conservative plan where by they can have the bills paid for years before any revenue.
Please note also that there is no Linux Kernel Inc. KDE is an exceptional model. even though it has not made any millionaires it's still an excellent and sustainable model. Trust me on that. ;-)
> P.S. - that Quanta fix works great!
I'm glad to hear that. We are working very hard to make Quanta the best tool anywhere for web development. We're also not planning on stopping there. ;-)
Ximian is a very different beast from what you seem to think. Ximian did not make evolution and Red-carpet out of goodwill. They were made to provide a basis for services that Ximian wants to offer. It took them a long time and a lot of work to get where they are now.
Ximian is not focussing on the desktop per se, but rather on he enterprise. Without enterprise, they would not have made Evolution or Red-carpet or even started the Mono project.
To make such a project succeed, you need more than just KDE. Ximian adds value in a way that is hard to destroy. Look at Mandrake. They are going out of business even though they were the most popular distro. Redhat does NOT make money (or much off it) from selling boxed CD sets. I think the desktop is actually becoming less relevant, rather than what add to it. Ximian is not just packaging GNOME. They are doing a lot more besides that. The desktop is probably just an avenue for providing services which is where they make the money.
I agree. I was a Ximian customer for the first year they were in business. It's easy to see that Evolution and Red Carpet were strategic moves, ones that I applaud. Could KDE not do the same, however? Are people afraid of targeting enterprise customers, and developing pay-for-use services and add-ons? Red Carpet-like upgrading could very well be added to KDE, and Kontact looks to overshadow anything Evolution offers. Time and hard work will definately be necessary. Isn't that what's happening already here?
Ximian's focus is on *bringing* the desktop *to* the enterprise. Of course the core desktop is just a commodity, the the powerful framework it provides to develop solutions upon is much more important. Isn't this our goal, too? How are we going to get there? I've seen many great products and ideas never thrive not because they weren't great, but because the right marketing and business decisions weren't made. Just because you build it doesn't mean they will just show up.
KDE already is a successful project on it's own. Don't you think significant value can be added to attract enterprise customers to KDE? We've already got our foot in the door. If it wouldn't have been for Active Directory, my whole organization might have been able to move to KDE. Things like KOffice, Kontact, and Kiosk already show great potential for targeting enterprise solutions. Commercial association could catalyze this. Would you rather have GNOME's desktop being the avenue for building viable business solutions? We need a more concrete plan for "konquering the desktop" (since that is why we're doing this, right?)
I guess what I am saying is you have to 'forget' how much you like KDE and try to think of why corporations would like KDE instead of Ximian. Ximian control what goes into Ximian Desktop, and therefore offer a product that fits in with their goals. They do not work on stock GNOME so much now so its not so much a question of contrlling GNOME, or controlling KDE for that matter to be able to make a good solution from it. Rather, they add value in their own way to make businesses see them as a viable choice.
I do not care which desktop is an avenue for building viable solutions. If it is viable it is good. KDE looks viable, but no one seems to have made an effort to bring it to the enterprise. GNOME has the backing of some of the heaviest hitters behind it, and some of the biggest Linux companies like Redhat and Sun, and I think the cause of this is that they can influence much of the direction that it takes. KDE's independence could actually be one of the reason it might be difficult for big companies to come in. The KDE community understandably does not want companies controlling it. Maybe the community may have to contend with companies exerting more control over the KDE project. They cannot take it away from you, so it should not be a worry.
Also do not forget how long it took Ximian to get where they are now. It will not happen easy, but it is possible.
SuSE is offering a desktop (not server) distro using KDE for enterprises since this month. This obviously happened due to the demand from their customers who were asking for desktop solutions to work together with their servers. I'm sure the more successful SuSE's desktop offerings are the more they'll improve the desktop basing on the feedback from their customers and contribute the changes back to KDE.
SuSE already mentions KDE's existing advantages: "Proven security functions -- the KDE desktop security system, Kiosk, as well as e-mail and file encryption - provide protected availability of all data on local hosts, laptops, and in corporate networks." and "KDE "Desktop Sharing" for remote control of the screens of several network hosts across hardware borders."
http://www.suse.com/us/company/press/press_releases/archive03/sld.html
Do we really need a Ximian for something like this? =P
> XD2 looks amazing, the entire desktop is integrated and overall pretty polished.
Just look at the desktop trashcan's right mouse button menu: It offers 16 entries in XD2! I fail to see where it's more polished than a vanilla Gnome 2.2. Even a complete KDE 3.1 installation shows less entries, not talking about HEAD with only 3 entries.
Yes, a desktop's trashcan context menu is a great index for its usability.
apparently such details are good enough to judge KDE ;-)
Gnome is so far behind KDE (look at nautilus in GNOME 2.x - it looks like and is featured like Konqueror in KDE1.x!)
Sorry, i can't seem to find emblems for Konqueror and I also ca't seem to find the zoom feature nautilus has, or cool looking previews and selections. A more usable list view has also escaped me. And so has an integrated cd burning mode and many other things.
Fortunately everything I mentioned is already a wish list.
http://bugs.kde.org/show_bug.cgi?id=37300
http://bugs.kde.org/show_bug.cgi?id=58944
http://bugs.kde.org/show_bug.cgi?id=58943
http://bugs.kde.org/show_bug.cgi?id=59791
> And so has an integrated cd burning mode
Which absolutely sucks in Nautilus. If you want to do cd burning in a file manager, at least do it right. Wait until linux gets packet writing UDF support.
> find the zoom feature nautilus has
Konqueror has zooming too. Nautilus just zooms icon and text while Konqueror only zooms icons.
> cool looking previews and selections
KDE's previews and selections are way better looking than the laggy as hell selection that Nautilus has. Even going a short distance, it lags behind the lag pointer. XP at least implemented this right with no lagging.
> i can't seem to find emblems for Konqueror
Sorry, but I've really tried to find a use fo emblems in Nautilus, back when I was using GNOME 1.2/1.4, but I couldn't simply find any. Very useless feature.
And if you install k3b (the best CD burning app available for Linux) you have the action "Burn data CD with k3b" in the rmouse context menu.
It seems to me that the Gnome guys want to overpaint the lacks of their (if you can call it so) framework by making a lot of noise. Funny, abstruse rules and definitions like the HIG are typical for people which haven't any real thinks left and try to persuade itself that ("the less is more"). With enough definitions of this type you can make the hell a "paradise" (commonly used in "real existent communist" systems).
Let'em cry, let us use our good codebase.
Excuse, but Gnome looks weird&ugly compared to this
http://www.kde.org/screenshots/kde310shots.php
Look at the design possibilities offered in Gnome2.0 - KDE 1.x level!
And Nautilus - it seems that they've tried to implement Konqueror - but the KDE1.x version!
Whilst KDE has Maintosh-grade beautieness, Gnome2.0 remembers me too much on Win95. And from the technical point of view this comparison gets even better.
No thanx, gnomes. We're on our way ahead and can't wait for you.
you will never aknowledge that the competition has done anything right!
Konqueror's zoom isuseless, at least have it implemented right like in nautilus.
The previews with shadows, you may consider slow, but I have not noticed a difference and I'm sure KDE developers could implement it faster, in addition nautilus even w th the "slow" previews loads image directories faster than konqueror on my computer anyway.
GNOME also actually has a decent theme manager and many of its themes rock, like Ximian Industrial and Redhat Blue Curve or Mandrake Galaxy. art.gnome.org
Emblems may be a useless feature to you if your unorganized, but I like to label my stuff to make it stand out. You obviously have not tried to use it fi you really couldn't find a use. It is essential to me.
the cd burning in nautilus is indeed simple, but it works, I don't know what you mean.
> you will never aknowledge that the competition has done anything right!
The problem is that people are growing sick and tired of people like you. In every single CVS digest, you post a list of (the same mundane) things that are found in Nautilus but not in Konqueror. Also, you consistantly (in every article) complain about usability and documentation. Not only this, but complain in other websites as well, such as OSNews and kde-look.
You do nothing to help whatsoever any of these situations however. Try doing that instead of blindly spouting things about how GNOME rocks. It's like Windows users spouting off how Windows rocks in Mac-only news and discussion forums back in the mid 90's, or Mac users spouting off how Macs rock in PC-only news and dicussion groups. It's ultra-unbeleivably fucking lame. It's understandable if you can't code, but you can always write documentation.
The sad thing is that you genuniely do seem to care. However, you do nothing to help. Please try diverting your energy there or else be considerate of others and stop spouting off the same things over and over in every article.
Thanks.
I've submitted over 20 bug reports, I think that's doing something. I also feel its important for KDE to have developers know what I think si wrong with KDE so they can fix it. I also think its a good diea to be aware of the competition and where you stand.
I remeber an old but very interesting OSNEWS interview with the KDE and Gnome UI/Usability Developers and I found out that almost all of the interviewed people were not at all familiar with OS x and Windows. Without knowing your competition inside and out its very hard to win and many innovations taht your competition has may never appear in your product.
Check out the article here: http://osnews.com/story.php?news_id=2997&page=1
In addition, I don't just complain, I mention specific things that I like in GNOME which i would like to have in KDE when I do make comparrisons.
I think my criticism has been generally well received,a list of complaints I had have gotten 90%+ on KDE-LOOK.org.
I really don't have the permissions required to write documentation or do any coding, let alone the time.
Even though I have spent more and more time learning about KDE, its community etc. lately, and now I am submitting bug reports and wishlists too.
http://tinyurl.com/ef3m
You don't have to read my posts, and oyu don't have to agree with me, nobody is forcing you to. I do care about KDE, I only criticize it so that it may be improved. The first step to solving a problem is afterall acknowledging it.
> I think my criticism has been generally well received,a list of complaints I had have gotten 90%+ on KDE-LOOK.org.
Any wish entry titled "KDE Improvement" gets such vote there. And it's still the wrong forum and off-topic there.
> many of its themes rock, like Ximian Industrial and Redhat Blue Curve or Mandrake Galaxy.
If you want to say that the default, non-commercial und freely under its name distributable theme sucks, I agree.
Nautilus faster than Konqui? WUHAHAHA! Redhat Blue Curve? Is a eye illness compared to Keramik or Aqua. And is there _any_ CD burning app which is so powerful and nicely integrated in the desktop like k3b? Any IDE comparable to gideon? A model for a modular desktop which doesn't looks like Suns RPC's in its early days?
And gnome libs? A heap of unstructurized functions (what should you expect from an implementation using the wrong implementation language....).
Network transparency? A framework which enhances older programs too? Powerful and easy extensible dialogs&widgets?
Here at the TU Berlin Gtk is dead - the courses teaches either Qt or Java/Swing.
All in all: die SILENT.
I never said Nautilus was faster, jsut that it appears to load image firectories faster.
In addition, i didn't say GNOME's cd burner is better, I think theirs sucks, I just liked the idea of building it into the file manager. http://bugs.kde.org/show_bug.cgi?id=58866
In addition, Anjuta compares to Gideon, but I don't know which is really better.
All I'm saying is that GNOME has some cool features too and we should try to have them too. I'm not saying its better overall.
In thinking of ways to begin contributing, I took a look at Konqui's interface (3.1.2). Sometimes adding polish means the littlest things. I select the "What's this?" item, and randomly click on buttons, fields, etc. Nearly everything produces no result, yet some do. It seems that this could be a good area to get a feel for making contributions to the codebase. After learning Designer, it becomes easy to make modifications to an app without touching the logic code. Therefore, one could go through the KDE code, and improve the UI files, adding tooltips, extended descriptions, etc. After becoming comfortable enough, you could then move on to improve the visual layout of the UI files, tightening their design, making sure they adhere to the style guide (of course, working with the authors). Does this sound like something that would be welcomed? I'd offer to coordinate this, and hopefully begin working. I was inspired by the quick turnaround on the clock UI.
There is a fellow, Malcolm Hunter that has gone through all the messages in the codebase fixing minor grammar, spelling, capitalization stuff. Making things consistent.
Yes, your contributions would be appreciated. Here's how to go about it.
Get the cvs version running on your box. Fix a few things, and send the patch to one of the maintainers, or send to the appropriate list. Start simple. Learn how to submit patches that are easy to apply. Accept suggestions and any feedback. Figure out how the i18n stuff works. Repeat a few times. Build a reputation and get accepted for quality work. Remember, most things aren't complete due to lack of manpower, not because nobody cares.
It's not very hard. There are many many little things that need fixing, and mostly need noticing by someone.
Derek