In this week's KDE Commit-Digest: Resurgent development work on KDevelop 4, with work on code parsing, code completion and the user interface. Support for converting the KVTML XML-based format to HTML in KDE-Edu. Support for the much-wanted feature of multiple album root paths in Digikam. Various continued developments in Amarok 2. Multiple additional comic sources for the Plasma Comic applet. Support for Kopete plugins written in Python, Ruby, JavaScript and other supported languages through the Kross scripting framework. A simple command-line application for playing media supported by Phonon. WavPack, TrueAudio and Speex format support added to the TagLib support library used by JuK and Amarok. Audio device work (utilising Solid) in KMix. Work begins on KsCD by a team of French students. Various optimisations in Plasma and Dolphin, amongst other applications. okular moves to a shared FreeDesktop.org library for PostScript format support. KGhostView finally removed in favour of okular for KDE 4.0. Code for supporting Apple OS X Dashboard applets via WebKit imported into playground (not for KDE 4.0!)
Comments
Thanks, thanks, thanks !!!!!!!!!!
Thank you Danny. Your work is very appreciated !
Many thanks, Danny!
Thanks!!
Thanks!!!!!!!!!!!!!!!!!!!!!!!
You reminded me of the mask salesman in Zelda: Majora's Mask when you said that :)
Thanks for the Digest Danny but why do you always put the sunday date and publish it on Wednesday? I guess you don't have time to release it on Sunday and that's ok but then you should put the date of Wednesday no? :)
It is about the commits until Sunday, isn't it?
Thanks for Digest.
I could be mistaken, but I think the digests used to come out on Monday, and eventually Tuesday. I'm assuming Danny does these as an unpaid volunteer, and I certainly look forward to them each week. Sure, I get a bit impatient when they aren't here, but I'm just grateful when they do show up and I get to read them.
I'm assuming Danny (like many of us) is just busy.
I thought the reason was obvious: the Digest is weekly publication which covers KDE commits in a set time period of one week (actually between Saturday and Saturday). For that reason, the Digest can never be more than a week "late".
As for releasing on Wednesday, circumstances change, and I can no longer be as rigid with my release schedule as I once was. Also, i'm sure you want a good read, so if the content for the introduction section is not ready, i'll wait for that.
But i'm still committed to producing the Digest, and I haven't missed a week out of the last 84...
Danny
So do you need at least some help from the community?
which part of producing the Digest takes more time? tell us, maybe we could help?
Thanks for so much effort and so many nice reading hours. You keep us (simple mortals) up to date every week. For me, each week is like when-I-was-a-child Christmas eve, -and you like Santa-. Thank you for making me feeling that way. Take you time, but -please- continue the good work.
I have to agree with that.
I remeber my first commit digest beeing far less detailed as the ones we get today (I may have a bad memroy thought ;)
But if I have to wait 3 days more to get these extra, I'm more than willing to wait ( he, nice to see Kstars used in real life, or explanations about Raptor...)
It sometimes feel the commit digest is more than just the digest, but a couple of What's new in SVN and a What's new in the Komunity. And that's great.
So, I've tried the new KDE4 beta, and I had the opportunity to use Kickoff for the first time. First impression: it looks pretty. But how does it fare when you actually try to use it? Well, I think that many of its critics have been a bit unfair. It's quite obvious that Kickoff is still a beta product, and it still has quite a few rough edges to soften. Nevertheless, people seem to pounce on it and tear it to pieces as if it were a finished product -- give it a chance for goodness sake!
One example: the navigation between the tree of applications is still very awkward and confusing. But don't you think the Kickoff guys know about it? Give them some time, and I'm sure they'll improve it to the point where it is usable. Also, I like the idea of the "clickless" switching of tabs. At the moment it still causes a few problems -- let your mouse wander a bit too much and the tab switches unexpectedly -- but with some tweaks it can become really cool and usable.
My conclusion? I think that Kickoff has quite a number of neat ideas and is showing a lot of promise. My advice to its developers would be to get some of the usability people involved, because I'm sure they can give you lots of good advice (for instance, the navigation through the application tree seriously needs to be revisited; I am sure those usability folks will have a heart-attack when they try that one...).
Anyway, keep up the good work!
Kickoff has been in openSUSE KDE 3 since 10.2 as default and it has the same problems. It didn't change at all between 10.2 and 10.3 so it seems that developers (at least at openSUSE) think it's perfect. This menu in KDE 4 is a full rewrite but looks and works almost exactly the same.
Interesting. I didn't know that. But it doesn't invalidate what I just said: Kickoff has a lot of neat ideas and potential, but it's quite obvious that it needs the eye of someone with a background in usability. It's very tempting to implement all sorts of new ideas (such as Kickoff's method of navigating through the application tree) just because they are cool; however, you also need someone to tell you "slow down; that may look cool but it's unusable by your target audience".
They did a lot of usability testing and they found that it was good. (More info: http://en.opensuse.org/Kickoff .), while many people (including me) find it unusable. The difference might be that in their test they used Kickoff for everything, including things that were not possible with the traditional menu, while I do these things with other tools and mainly use the menu for application browsing. I think the main problem with Kickoff (both the KDE 3 and KDE 4 versions) is that nothing can be configured: not even the default tab or whether there should be an animation, which is not usual in KDE.
there will always be people that don't like something, you can't please them all. Personally I love kickoff and have no problems with it.
The only thing I didn't like about it in KDE4, and it is a beta so will change, was that I couldn't add things from the kickoff to the desktop. But that will change, it's just a beta.
I did not say that it is a problem that some people don't like Kickoff, as long as there is an alternative.
I think I read that someone is reprogramming the old KDE3 menu as a plasmoid.
I do think that's because Plasma has no proper support for "desktop icons" plasmoids yet.
kickoff is quite ok (and i say that as an user of many console apps).
i am using it on opensuse installs i have, but there are some problems with it, obviously.
most of those could easily be solved by adding more kde-style configuration (this is where ui config nazis clash on each other :) ).
1. already mentioned animations - there should be a way to disable those (on some setups applications meny looks veeeery bad when animated).
2. in suse-styled old kde menu there was a filter field which would disable meanus and entries not matching a filter - this allowed an easy way to find where the shortcut is located in the menu. in some cases, this is not shown in kickoff.
3. tabs switch on mouseover, not click. this probably is the only thing i really hate about kickoff. if i could make them switch on click, that would make me much, mcu happier.
i also understand that some usability study might suggest switch on mouseover is better for some groups - which i can happily accept. but this has gotten in my way too many times and has really pissed me off, thus i nominate this as the biggest headache.
first two items are minor annoyances, third is ANNOYANCE ;)
> and mainly use the menu for application browsing.
How often do you browse through your 'all application' list? Are you really saying that you very regularly use an application that is not in the 'Favourites' tab? If so, why isn't it?
> or whether there should be an animation
What animation? The green geeko? Of course it is configurable.
I use other methods to start regularly used applications: I put icons on the kicker or on the desktop. Additionally, I have the Desktop directory as a menu on the kicker. For what I need a main menu is browsing all applications.
I agree. For starting my favorites I use Katapult (I hope that will still be in place in KDE4.0). If I need an application which I do not know by name or which is not found by Katapult, because it is newly installed, then I use Kmenu. I tried Kickoff the few months I had opensuse installed and also had a look on it using current beta-liveCDs. Application-browsing is rather a pain, the menu is not configurable. I really hope that KDE4.0 will also have Kmenu and Katapult.
The animation when I switch to a submenu.
http://www.kdedevelopers.org/node/2620 and http://roderick-greening.blogspot.com/2007/02/kcontrol-updates-for-kicko... may provide a hint how to configure it. Through I don't know by myself about the state of a GUI to change such options.
You think I complain about it?
You should hear my wife scream. She thinks it is the worst thing ever invented.
I can't understand how something so painfully slow is good for usability. This has never once been explained. This particular menu is so bad, it has me completely sworn off menus, and just using Alt+F2 all the time.
"They did a lot of usability testing and they found that it was good."
The problem with this "usability testing" done by the developers themselves it's that it's not testing usability at all! If I developed something, of course that I would think it was great and intuitive and all that. But usability testing must be done with REAL users, meaning people who didn't develop the thing!
I bet that if INDEPENDENT usability experts would take a look at Kickoff they would immediately come to the same conclusions as many posters in this thread: a) that the typical KDE user would find the application menu too slow to interact with; and b) though you think you might be helping users by hiding the tree structure, in fact you're only making people feel lost.
Now, I have to insist on this usability thing because it's a common pattern that I see in many KDE apps: the developers themselves assume that if something feels intuitive to them it will feel the same to everyone else.
I really like some of the ideas behind Kickoff, and I would love to see this project thrive. I would really hate to see Kickoff fail simply because their menu structure is unusable and the developers were to proud to reconsider it.
So, Kickoff guys: please get some usability people involved ASAP! We love your ideas, but you must also listen to REAL users.
RTFA
"Central to this research is a usability study where about 30 users of different types..."
But redding is teh hard!!11 I'd rather spout irrelevant drivel on the dot!
Lulz.
It's not really relevant if he read it or not. As explained before by some other people: it tries to solve problems in the menu field that should'nt be in a menu. It might be the most optimised menu-solution to create the fastest way to do these things, when they all need to be done using a menu... which they don't: favorites in a menu? come on, one simple "favorites" plasma on the bar that behaves like the "favorites" list in kickoff would be a lot faster (you don't even need to press the kickoff menu... what a speed improvement). Put too manu things in a menu and you'll get some slow action.
Replacing problems to another area and then screaming: we've found the best solution to these problems in this area without even considering that moving the thing was wrong to start with is not quite a good study.
>>But redding is teh hard!!11 I'd rather spout irrelevant drivel on the dot!
Irrelevant drivel ??
Following your 'logic', people would have to read this 'study' in order to submit to their very basic ongoing frustrations.
It seems valid points are pointed out in this thread. Irrelevant drivel ??
do you consider this number of guinea pigs to be sufficient for such testing?
if someone is doing public opinions poll within our small 10 million country, mathematicians laugh at him if he has less than thousands of respondents (carefully selected to cover all social groups)
30? - it is far less than how many professions are listed in our bureaucratic forms ... could you imagine that one menu approach would fit the same to CNC programmer as to old-school secretary still missing her purely mechanical typewriter? how many other contrasting examples you can think about?
... not talking about the fact, that the study does not consider what we all need the application menu for - starting an application (no, I really do not want to find my documents via "start" menu etc.)
I just don't believe the study.
It hides options, it forces you to click 5 times to get to one thing, and it is the slowest menu I've ever used.
How is that good for usability? No one has ever answered that.
If it makes me CONSIDERABLY less productive, how is that good for usability?
Sorry, but using 30 random people hardly constitutes a proper usability test. Besides, using "different types" is a fine way of trying to please everyone and ending up being hated unanimously. And how many of those 30 were actually typical KDE users?
I still maintain that Kickoff was never subjected to a proper usability test with members of its target audience. That means typical KDE users, who on average are more technically minded than your regular folk. Once someone picks, say, at least 30 such users and tests Kickoff with them, then you can claim that Kickoff has been properly tested.
Why should the target audience be current KDE users? And to your "how many?" question, how about actually reading the study/presententations done about it?
> Why should the target audience be current KDE users?
since he considers a bad_thing(tm) betraying large user base of "unix lovers" just to attract a few Windows deserters who did not end up with Gnome for some obscure reason?
> And to your "how many?" question, how about actually reading
> the study/presententations done about it?
well, how about actually reading the question? - it is NOT answered within the study
> it is NOT answered within the study
Sure it is, eg Akademy 2006 slide #20.
> Once someone picks, say, at least 30 [technical] users and tests Kickoff
> with them, then you can claim that Kickoff has been properly tested.
So, you don't disagree the 30 is too low and you don't disagree that the test being done was bad. Just that the people the test was run on doesn't fit what you think the user audience is.
Well, I think thats a good thing then :)
I mean; I actually do think the user audience being a random set of computer-users is perfectly ok. If you don't think so, then we disagree on that. But I'm very happy we agree that the test for the intended audience was good.
Cheers!
> The problem with this "usability testing" done by the developers
Who says that the usability was done by developers or with developers? Clear, get your facts straight.
Where does one get the source for KickOff? I use Gentoo:
$ qsearch kickoff
$
There's nothing in portage for it (unless it's called something else). The page you link to doesn't have a link to the sources. A quick check on Source Forge (just for grins) shows a Soccer Game and some Windows Logon Script something-or-other.
All I can find is an SVN link, but I honestly don't see anything list there for kickoff. Am I just missing something, or is SuSE/Novel holing on to this one for themselves?
Learn to use Google before raising accusations because your distro being lame?
learn to think a little more before calling a distro lame. did he offend yours? btw it used to be in the xeffects or sabayon overlays (i think). you might have to enable some USE flag for it to compile. i'm not sure though as I only use KDE4 live ebuilds at the moment...
oh yeah, just because the maintainers do not find your favourite toy so useful to be worth wasting their time by including it within the main package tree, you call the whole distro lame? - great flames for Friday morning ;-)
it is included as patch for kicker in xeffects overlay
add xeffects overlay and emerge kicker with kickoff use flag
I just looked at the study, as far as it is presented at http://en.opensuse.org/Kickoff/
The sampling used is quite questionable if one wants to find out wether kickoff is more efficient than the classik kde menu.
They used 30 participants and divided those into 3 groups of 10 each: One using kickoff, one using the classic kde menu and one using the windows vista beta2 menu. According to SuSE, the groups consisted of the follwing (they considered some more factors that I omit here, I also omit the windows testing group):
kde classic:
by OS/Desktop:
Windows: 7
Linux/Gnome: 0
Linux/KDE: 3
by skills:
beginner: 9
advanced: 1
expert: 0
kickoff:
by OS/Desktop:
Windows: 3
Linux/Gnome: 4
Linux/KDE: 3
by skills:
beginner: 1
advanced: 6
expert: 3
Then they compared those groups solving 13 tasks (of which I would not solve more than half by using the start menu at all) solely by using the menu. Basically they compared 9 beginners and 1 advanced user (kde classic) to 1 beginner, 6 advanced and 3 experts (kickoff group). Is anyone really surprised that kickoff seems more efficient when comparing those two that way? Not to mention that they used 7 linux users in the kickoff group, while they used only three linux users in the kde classic group. At least both groups included 3 kde users. :-)
Bottom line: while this study most likely had some usful results (eg. by watching the participants and their reactions), it is hardly suitable to compare kickoff to the classic kde menu, at least in terms of efficiency. Whatever the intention of this study was, it most certainly was something different.
Two misconceptions:
(i) Kickoff did get some changes between 10.2 and 10.3 (like tab location), a few extra non-gui options
(ii) the one in KDE 4 _is_ beta, everything has not been implemented into it yet.
..and yes, it's been in openSUSE for some time, it's proven itself, and openSUSE users have been _very_ positive about it.
"openSUSE users have been _very_ positive about it"
At least some of them. I am an openSUSE user and many other users do not like it after trying it, just see the comments on the dot.