i need to read email in the evening

i really need to start reading my email in the evening rather than the morning. it's sort of like eating a good breakfast (for those that do that sort of thing): it gets you off to a good start. a bad breakfast does the opposite. email seems to be more and more the latter.

and speaking of breakfast, i'm looking for other contributors to kde://radio. if you've listened to the recordings there already you may notice that it's pretty easy. the format is two or more people chatting about something kde related which is pretty natural for most of us kde peeps. unfortunately for me i don't live near many other kde people and these things go over much better in person. (aren't people funny like that?) seeing as i won't be travelling again until the end of september, it would be very cool if some of the other kde people out there were to record a relevant conversation or two that could then get posted.

new konqcast and kicker feature (!); otherwise the play was fine

posted another konqcast over at kde://radio today. this time it's with ade of the english breakfast network complaint automator. (listen to the podcast to get the joke).

someone came on irc today saying they had a friend who would switch to kde if only they could make the taskbar blink indefinitely when a window demanded attention, ala gnome. this seemed like a silly thing to let get in the way. but seeing as i was already 5 hours into boring administration tasks any sort of hacking sounded really great so i spent 10 minutes adding the feature. after lunch with the p-man i tested it out and then uploaded the patch to my site.

with that patch you can edit your ktaskbarrc to have an entry like this:

[Appearance]
AttentionBlinkIterations=1000

and it will blink forever. anything less that 1000 and it will blink that many times. hooray.

other than that i did a metric butt-load of paperwork and organizing, two loads of laundry and spent some time outside with my son who is taking off for a week's vacation in california with his mom on wednesday.

post-ols, how big is the market?, organic interfaces, working with others

back home and today i'm going through the usual post-conference confusion syndrome: keep forgetting it's actually saturday, lots of discouragement looking at my house which needs a bunch of tlc and my todo list which has only grown.

but the conference was definitely a plus. the kernel people are becoming more aware of what we need, x.org is moving forward and several projects that had low profiles before this got to peek out at the world such as the open content project. i also got to karaoke with a bunch of people including a couple new friends from the gnome project. meeting adriaan degroot for the first time in person and hanging out some more with maddog was also cool.

a highlight was seeing an unpublished study during breakfast one morning that measured the linux desktop as having ~6% market penetration in north america and higher in some parts in europe. given that events such as the waitress that served me lunch the day before noted to me that her brother uses linux on the desktop are increasingly common, i suppose i shouldn't've been surprised ... but i was. that's significantly more than "common wisdom" says we have and certainly jives with numbers i've been playing with and hearing about from other circles. there was a lot more in the study and i'm very happy to have received that knowledge, but i probably can't say much more publicly at this point since the study is still private. hopefully it won't be forever.

another highlight was hearing someone present on the future of graphical interfaces talk about "organic interfaces" which is something that i and a couple others from trolltech came up with about a year ago. the basic idea is that millions of years of humans evolution was directed by the traits of the world around us, molding us to be hands that fit the glove. (or if you believe in "intelligent design" one might say that we were created as perfect fits for the universe we live in.) so if we create software that behaves more like the world around us, these organic interfaces will resonate more deeply with users who will grok the software quicker and feel more at ease. this leads to changes in both visual design as well as metaphor such as the use of non-square corners and pervasive subtle gradients as well as avoiding iconification when avoidable leading to a more realistic one-to-one mapping on screen between thing and action (think drag and drop).

it was also useful to talk with kernel and desktop people together so we could get a much better handle on what the kernel needs to provide the desktop as well as what userspace apps need to avoid doing so as to make life for the kernel utter hell (such as our rampant stating and opening of files... it's really insane right now). there was enough emphasis from the kernel people on supporting the desktop (which includees laptops as well) that i'm optimistic we'll see real improvements here.

and finally, while i was away i got a package in the mail from the K3M meeting people. inside were two booklets that looked quite professional wrapping up the meeting. i wrote the foreword for the 28 page "final report" that is being sent to our corporate sponsors and meeting participants, which is why i suppose i received a copy of these items. i think it's a wonderful way to give those who were there something to remember the event by as well as a perfect means to communicate the benefits of the event to our financial supporters and to those who would like to write about the event. it's the first time i've seen such a thing for a kde event and the coordination team (martijn klingens, claire lotion, adriaan de groot and jos poortvliet) deserve a huge amount of respect for the truly professional results. i know they worked hard on it and it shows. i wish we had such a thing for trysil; next time we'll have to convince claire and martijn to join as and document the event.

kde fundraiser drive

right now, kde is running a fund-raising drive for some new hardware acquisitions. we really do need more hardware to support the needs brought on by the growth of the project over the last few years. we've come a long ways from the time when kde's cvs was hosted on a machine under someone's desk. we've received some donations of hardware from industry (announcements of that forthcoming) and that's a great start, but now we need to do some additional purchases for things like more disk space.

so donate today and help us keep our infrastructure fresh!

recycling kernel developers

we're ~80% of the way through ottawa linux symposium (ols) and the desktop developers conference (ddc) week. thus far it's been moderately useful. the birds of a feather meetings on the second and last day of ddc were particularly good as they allowed people from disparate projects to come together and short circuit months of email conversations with high bandwidth communication. technical achievements have been low, but networking and information transmission has been huge.

today i attended a presentation on perfmon2, a git revision control workshop as well as a presentation on the new dbus interface for bluetooth on linux. all useful for desktop work.

it also occurred to me that a lot of what we're (using the term "we" very broadly) doing in x.org could really use with some attention and participation by the sorts of people who attend ols: kernel developers. we need a preemptive scheduler for access to memory on 3d graphics cards; we need hotplug support for monitors (which ajax has been working on this week and has in some part working already, though i'm not sure poking /proc every few seconds is the nicest long term solution). these are the kinds of things that people working on kernels have been doing since the beginning of time.

i bet if we looked around the desktop we'd see more examples of such stuff. why aren't we reaching out to them more directly and inviting them to bring their expertise and talent to bear more directly on the desktop? historically the desktop has been all about a gui candy coating of software. today it's growing eyes and ears and becoming a lot more integrated with the physical world around it. that may not be the current desktop developers' primary area of expertise, but i'm surrounded by a couple of hundred people for whom it is.

i'll be here all week

i arrived in ottawa this morning and came directly to the desktop developers conference (ddc), bags in tow. many familiar faces are in the rather small crowd: chris schlager, waldo bastian, brad hughes, daniel stone, chris blizzard, keith packard, etc, etc, etc .. and i think it's the first time any of them has ever met me in my country of origin: canada.

i arrived just in time to catch the tail-end of waldo's talk on portland and as i write this i'm listening to the presentation on network manager.

after the two days of ddc is the ottawa linux symposium, which i'll also be attending. my goal is simple: to gather information to bring back to the kde project about what is happening in the rest of the free software world, to ensure that the kde perspective is represented and to find time in between to hack on code.

i spent pretty much all of last week dealing with administration duties. i think i'm going to try and avoid any more of that this week and do a pure tech week. i do have one or two minor administration issues to take care of (hi andreas!) but i'm looking at soaking in a nice vat of tech.

i'll be blogging as the week progresses and hope to record some podcasts for kde://radio as well ... though right now a shower and a proper meal would also be nice.

kudos to dental-on-line

aurelien gateau blogged about dental-on-line's new open source patches website. it presents their in-house modifications to the free software they use in a nicely sorted and annotated fashion. with the reality that not all patches will make it upstream, this is certainly one way to be a useful and upstanding free software citizen! kudos to this free software lovin' dental software company.

now i'm left dreaming of a day when there's an easy to set up website that anyone can self-host to present their patches to the world like this ... which are then linked together via some sort of grid-like web service so i could easily grep for kde patches ... a queryable patches planet ... which would give me a list of such patches from around world along with who they came from, what they do and a link to the patch itself ... all decentralized so there's no mother ship to have to coordinate with, no single point of failure ... just a low barrier mechanism to publish patches for up-, side- and down-stream adoption ...

i also want chocolate that doesn't rot my teeth. either that or the ability to grow new teeth like sharks do. that'd be nearly as cool as that distributed patch publication system idea. not quite, but nearly.

equal and opposite reaction

luis villa has been posting various thoughts he's had in the wake of attending guadec. some i really agree with, like the idea that it's almost certainly possible to bring a non-trivial amount of monetary support into free software desktop projects through web based services that we offer access to from our desktop apps.

some of us have been talking about that in kde for at least a year (probably more, but that's the first conversation i can remember having about it explicitly)... maybe once we're out of the kde4 woods some of us might actually start working on that.

but some of his thougts don't jive with my experience. such as this one about project portland, which is an effort to bring a simple set of desktop-neutral APIs to ISVs (those people who write third party software). in his missive, luis said:

Given two mediocre ISV platforms*, and a finite amount of time and money, why spend that time and money to fix one ISV platform to kick ass** when you can spend the time and money to add a third mediocre ISV platform instead? Yay!

first, this is an effort done in direct response to ISV requests. they want something simple to use now not n years from now. portland is a pragmatic project that most of us realize may well have a limited shelf life. and that's ok.

it's ok because it doesn't take a lot of effort, not compared to the work to "fix one ISV platform to kick ass". not that doing that would fix the issue at hand, which isn't "the platforms are mediocre" but rather "how do we do X" where "X" might be "install an icon on the desktop". now, i'm not trivializing the efforts people have put into portland, but it's pretty telling that the lsb is already looking at it after a very small number of people have put limited amounts of time into to produce something useful already.

did you know that google ships some of these portland tools with their google earth software?

now, creating a right-size solution at low cost that people are needing now may seem like the greatest idea since sliced bread. well, at least like an obvious idea that deserves to be done. but this obvious idea does not come at the expense of making kde's api kick ass (i'll leave comments on gnome's api to those involved with it). we're making kde's api kick ass right now even as portland continues. and note that the driving coding forces behind portland are mostly people who hail from the kde project, so one might expect that if portland was going to cost anyone major API kick-assery slow downs it would be kde. but then one would have a hard time explaining our moving to dbus, adoption of cmake, development of akonadi, solid, phonon, liveui, the k3m and trysil meetings, etc... these are all API level, ass-kicking-potential improvements and only represent a fraction of total activity in this area.

so while i respect the desire to keep an eye out for things that will detract from the long-term sustainability of the free software desktop, ISV and API issues among them, i don't feel portland represents such a risk while delivering real world value to our ISVs.

having problems with the konqcast?

if you're having problems hearing the konqcasts, try using this rss feed url instead http://radio.kde.org/pub/konqcast/konqcast.rss.

this should get you around the dns issues some are still having.

measuring the kde user base

the current size of and growth rate of an open source project's user base is notoriously hard to measure. we can't rely on sales because the software is freely available with many independent distribution channels. in the desktop space, many products allow one to pick the desktop installed post-purchase anyways. our software doesn't phone home(though that would make for a nice slashdot article;) and participation levels on user forums gives us only general guidance that we are growing without actually revealing any numbers ... web stats are not specific enough for us (the browser often belies the desktop), but even if they were they are often skewed by usage patterns that vary between different user segments. to complicate it further, many kde systems are simply not on the internet at all. some have used cd handouts to imply user base numbers with, but that's an amazingly shakey (some might say meaningless) bridge to errect.

but there are some tantalizing hints as to the extent of our user base growth rate out there. everywhere i go it seems i run into kde users: from the streets of paris to print shops i patronize to cocktail party goers to random business people i meet in assorted american cities to users in small towns in norway. it happens so often it no longer really surprises me. this certainly wasn't the case when i first got involved with kde. interestingly, though, i have yet to see a single linux laptop on an airplane besides my own or those belonging to people going to linux conferences. the implication of that is interesting. but my experience is simply anecdotal and hard data would be nice.

which brings me to the numbers in a recent article in brazil's "easy linux" magazine titled "linux para todos" which covered a brazillian national program to make inexpensive computers running free software available to the public. they all happen to ship with kde running on linux distributions (5 different ones if i recall correctly).

the article notes that 200,000 of these machines have been sold in the 12 months spanning from april '05 to april '06. of course, not all of those systems remain kde desktops. one survey i've learned of has found that approximately 80% of those who bought one of these computers ends up installing microsoft windows on it when they get it home. this is not surprising as most people only know windows. i'm happy the retention rate is as high as 20%, to be honest, since the majority of those people are almost certainly trying something new for the first time and that's always the harder path to take.

20% leaves us with 40,000 users. that's over a 100 new kde systems per day due to just one program in one country. how many of those represent new kde users is harder to pin down, but it would be interesting if we could get some data stating that the majority of people who use those systems are new kde users. it's also probably safe to say that many if not most of those systems are used by more than one person.

then there is red flag linux in china. according to this interview with the red flag people they ship over one million copies a year, use kde for their desktop (and do a fair amount of custom kde development to that end) and currently hold 80% of the chinese linux desktop market. given the growing importance of the chinese market, these are pretty important numbers.

of course, of those one million units shipped most are almost certainly for servers assuming the linux usage patterns are similar in china as they are in the rest of the world. it's still nice to know that if a gui is brought up on those servers that it is kde, but how many "traditional" desktop users might there be in that one million units? they didn't say in the interview and i haven't yet asked, so i'll just do some wag'ing (wild-assed guessing) and say that 95% of those shipments are for servers. it could be higher or lower, but let's go with 5% as a number. that gives us 50,000 desktop units or nearly 137 new users per day.

we end up with some 240 new installations each and every day due to two projects in two countries. i'll wager that many of those reading this had not heard of one or even both of these projects before. how many other similar projects are out there, large and small, pushing kde out into the market place all around the world? if those 240 installations a day represent 5% (which i think is probably generous) of our user base growth rate that would give us 1.75 million new desktop (not server) installations per year. i wouldn't fall off my chair if the number was even higher.

the best news is that there is simply no way that 5 years ago we were seeing this kind of growth. which means adoption is accelerating. this implies that if we keep on track even better numbers await.

p. the busker

p. decided to take his xylophone outside to the sidewalk along with a hat to engage in some busking. he has all the tricks down, including seeding his hat with some change he collected from my desk. clever lad.

he's already made a couple of bucks in his first half hour at it, and i must say that his musical efforts are actually pretty enjoyable if somewhat random in composition.

forget lemonade stands. public performance is where it's at.

konqcast: trysil

audio recordings of chats i had with people who attended the kde four core meeting are now up on radio.kde.org. they aren't slick with cutesy music intros and what not and there are the occasional audio glitches and i've only tested this with amarok and akregator. so apologies in advance for the probably mistakes from me, the podcast newbie.

i am looking for feedback such as "the rss feeds don't work for me" or "you know what would be a cool topic?" or "oh my god. please stop. now." ;) i hope to continue doing more of these if the response is positive.

konqcast (ogg format)
konqcast for those who need it in mp3

update: i've added a small crappy bit of html to radio.kde.org so that if your web browser supports automatically finding rss feeds and adding them to your aggregator you can go that route. a "real" website would be cool for this, but beyond my time budget at the moment.

update 2: if you're getting 404's, that's because the dns hasn't updated properly yet. it changed last night to point to the new location. we also now have an osusl mirror (huzzah!) and as soon as that is sync'd up (tomorrow) i'll be adding those links as well.

enjoy (i hope =)

more toolbars

in true "blogs are the new mailing list for odd conversations" fashion, clarence wrote a response to my response to his blog about the new toolbar default in kde4. (random thought: is there an agregator that uses trackbacks or other such metadata to thread blogs into conversations?)

i like clarence's new 6 element toolbar for kde4. it's much more task appropriate and easier to manage. he does bring up two issues that i agree are valid to talk about: system wide merging messing up toolbars and icons under text adding to visual clutter.

in kde3 we have a system wide file that defines standard toolbars. the problem is that toolbars simply aren't standardizable in the same way that menubar layouts are in my opinion. toolbars simply don't cope with the odd "accidental" entry as they are way too "in the face" of the user. perhaps in kde4 we shouldn't define a default toolbar at all and make it opt-in per-application so that we get task appropriate toolbars. the number of apps that don't use the system definition with noMerge is already pretty high anyways.

as for text adding visual clutter, if we use a nice (e.g. smaller) font for it it really doesn't get in the way. i've been using toolbars with text under the icons since 3.5 was released (to get some first hand experience with it in prep for kde4) and the problems are long names and big font sizes. they help disambiguate toolbar icons quite nicely so i don't have to learn every single app inside and out. and in clarence's screenshot it's obvious that sometimes it even shows contextual information, as can be seen in the "Undo: Line" button.

on toolbars

in a slightly more positive tummy-rumbling of a blog entry, clarence dang (of kolourpaint, a kde painting app) says:

From this screenie, it should be obvious that KDE's new toolbar default of "Text Under Icons" does not work unless one has a screen with 5,000 x 5,000 resolution.

which nearly misses the point. it is very true that many of our toolbars when used with text under the icons take up a lot of screen space. however the problem is not bigger toolbar buttons, which are good for usability in many ways (more on that in a second). the real problem is that we cram so many (often stupid) things into our toolbars.

so what are toolbars for? toolbars give rapid access to the key actions in a given context (which is usually a task-specific window).

toolbars are not a billboard to advertise "neat-o" but rarely used features. that's what menubars are for. so when "mail" appears in a painting program's toolbar (kolourpaint) one really ought to say themselves, "wait a minute! this isn't an email program! this is a painting program! emailing a doodle isn't in the top 5% of actions! it's not a core part of this task's workflow!" and for most apps, even the top 5% of actions is going to be more than you want to shove in a toolbar.

another tip: if your toolbar icon's text is more than 2 short words long, ask yourself why you're being so wordy. probably it's because the action is ill defined or you're trying to be overly exact. remember that users can mouse over for tooltips and that once they learn the toolbar (given that it's learnable, e.g. less than 10 or so items on it) they won't rely on the verbage anyways: it becomes a matter of muscle and visual memory (placement and iconography). that's right, i'm looking at you "next unread message" in kmail and "mark feed as read" in akregator! why not just "read next" and "mark as read"?

in other words, changing the toolbar defaults is meant to make application developers rethink their abuse of the user experience through really poorly thought out toolbar entries.

i mentioned earlier that bigger buttons in the toolbars are better for usability. why? bigger objects are easier to point at. bigger objects also mean bigger icons which means greater clarity (22px icons are pretty hard to make really clear, particularly given the specificity of some of our actions). bigger objects mean fewer objects which means less clutter and greater learnability.

so rather than bemoan text-under-icons, perhaps take it as an opportunity to make your toolbars suck less(tm). users of kde4 will thank you. chances are you'll thank yourself.

the kde effect in trysil

when kde comes to town, e.g. at the trysil meeting, we tend to have an effect on the information flow in the area. right now if one searches google for trysil, a kde blog entry comes up in 5th position. it even appears ahead of the wikipedia entry. wow.

we've also been in the regional newspapers twice: once while we were still there and meeting and again just today. the writer, inge scheve, brought by two copies of the paper for us just before we left on saturday and also did a second interview with myself and the other kde devs who were still present. our first article appeared on page 19 with a teaser on the front page; both snippets included photos. inge covered the social aspects of free software and kde meetings. today's article also included photos and covered the technical goals of kde4 and what we accomplished while we were there. according to our native norwegian 'man on the ground' olav, she got the details right. given that inge is actually a sports writer that's pretty terrific. i think it speaks to both her abilities as a journalist as well as the increasing clarity of our communication.

inge's also been keeping me up to date with the ongoings in trysil, such as the recent take over of the public transit system by the resident rumanent union (ovis local 232). or something like that. here, look for yourself:

© 2006 Ingeborg Scheve

in any case, it's probably fair to say that people in the trysil area are now much more aware of kde and the whole free software desktop thing than they were a couple weeks ago. this obviously wasn't our intention when planning the meeting in trysil, but it is a good example of how one can hook up effectively with local media when there is something newsworthy to promote.

i'm in much better spirits today thanks to a good night's sleep. even though it's "teleconference tuesday" (in which my morning is monopolized by my least favourite household appliance: the telephone) i'm having a good time of it.

in brief

t. picked me up at the aiport on saturday and promptly whisked me to a games night party at one of her friend's places ... great fun, but a bit much for having just walked off an airplane from europe ...

went out with t.'s family and her to celebrate her birthday on sunday. we ate at the coup, a local vegetarian restaurant. the food was good, the company was pleasant (if a little argumentative at times). i tried not to get involved too much as i was still recovering my energy from traveling the day before and it was, after all, t's birthday.

today was spent catching up with email, kde related duties and grocery shopping so i had some food to eat. peyton stayed home today and hung out around the house keeping himself busy (he's quite good at that) and shopping with me.

i still have a mountain of things to do ahead of me this week and right now feel an extremely low amount of energy in my spirit. a few good nights sleep will probably fix that.

i'll be at ddc and ols next week. find me if you're in the area. =)

trysil thursday

we climbed a mountain
in the middle of the night



and saw forever

tromping through norwegian forests to the highest peak in the area seemed a good way to start the last full day of kde four core four.

trysil wednesday

after dinner on the deck (which was mostly vegetarian and quite tasty, hooray!) we had a mid-week progress report meeting where each team of people (with some people being teams of one) got up and gave a quick update on what they were working on and the state it was in.

LiveUI (xmlgui replacement), graphical dbus explorer and dbus support enhancements, accessibility (hooray for at-spi over dbus working!), human interface guidelines (yes, we have someone actively working on that!), x.org composite integration, KJob (abstracting out the many implementations of jobs in kde; more significant than it sounds), KIO improvements (including proper seek support), kdelibs API reviews, akonadi, qgraphicsview use (e.g. in kpat ;) and khtml work were all covered. i probably missed something in there too. we've been busy.

it hasn't all been hacking though. there's been team building in the form of a daily walk and various meetings on important if not overly-technical topics. all of this (and the lack of a proper darkness at night) has left us fairly tired at night. how tired you ask?

george works till he drops

celeste is not to be disturbed
(more than she already is? ;)

i'm fond of the advanced yoga sleep position known as "the orkan"
(and no, i had not been drinking. honest.)

david does not know the meaning of sleep however
no wonder he gets so much coding done

... and then they got familiar

while working on accessibility in kde, gunnar schmidt was testing the text-to-speech support in various apps. he got around to testing with kde's konversation irc application and when people noticed this new toy in the room all hell broke loose. just imagine the following being read out loud in a mechanical voice with a slight british accent from gunnary's laptop:

[08:50] <icefox> I think it was copied to KColorEdit, need to double check
[08:51] <col> boemann: if koffice relies on it, it's easy to copy to kofficeui
[08:51] <boemann> well since it's a generic for a gradiented slider i think many would like to rely on it
[08:52] <[ervin]> hahahahahahahahahahah
[08:52] <col> boemann: this is not about people liking it, this is about people maintaining it
[08:52] <pmax> muahahaha
[08:52] <[ervin]> OH OH OH OH OH OH OH
[08:52] <boemann> well i've been maintaining it
[08:52] <[ervin]> OH OH OH OH OH OH OH
[08:52] <pmax> OH OH OH YES YES MORE
[08:52] <harryF> YYYYYYYAAAAAAAAAAAIIIIIIIIIIIIII
[08:52] <[ervin]> OH YEAH OH YEAH OH YEAH AGAIN!
[08:53] <col> is it really true?
[08:53] <[ervin]> tak tak tak tak tak tak
[08:53] <boemann> take a look at the commit
[08:53] <pmax> SLAP IT! HARDER!
[08:53] <will_> shoo bee doo bee doo wop
[08:53] <aseigo> do my hard baby
[08:53] *** will_ is now known as Bille.
[08:53] <[ervin]> OH OH OH OH OH OH OH
[08:53] <aseigo> in the midnight hour
[08:53] <aseigo> she screams more, more more
[08:53] <pmax> Now from behind
[08:54] <aseigo> you know how i like it baby .. rough
[08:54] <Bille> i think i'm turning japanese i think i'm turning japanese i think i'm turning japanese i really think so
[08:54] <[ervin]> nasty
[08:54] <[ervin]> oh? broken?
[08:54] --> till has joined this channel (n=till@kde/tilladam).
[08:54] <icefox> :(
[08:54] * [ervin] want's his toy back
[08:54] <Schmi-Dt> I have temporarily turned text-to-speech off now.

and so with that fun over we returned to hacking. i podcasted with the akonadi team today (still working on the hosting.. soon, i promise!) a journalist visited us today as well. she works with two regional norwegian papers. thanks to Olav of Trysil for hooking this up!

we've been going through kdeui and have found an astounding number of classes that are simply unneeded thanks to both kdelibs maturing and qt 4.2 being released. currently the number of deathrow classes stands at 20. we're about to have a meeting to discuss the final outcome of things, and after that will be reorganizing kdeui into sensible subdirs.

lots of other hacking is happening, but i'll blog about that this evening.. they are waiting for me to start the kdeui meet. cheers!

trysil tuesday

i received an email the other day from a trysilian (or whatever one calls a person who lives in trysil) who is a kde/linux user. apparently he switched over not too long ago and has discoverpued it does everything he needs quite nicely, even in his work as a translator. he and a local journalist is coming up tomorrow to chat with us a bit.

we had a couple of quite long meetings today about kdelibs: what should go into kdelibs; how should we handle the process of managing the public APIs; what are our goals and practices for binary compatibility ... not trivial topics even if the questions are easily stated. we hopefully will have some proposals to bring to kde-core-devel soon.

we had lunch out on the deck in the sun with music supplied by my laptop. it was a nice break from the more heady topics earlier in the day. people are now back to hacking on various qt, kdelibs, dbus and build system issues. later today i'll be presenting the plasma technical design.

i made a podcast the other day and uploaded it .. hit a snag with our hosting solution for it, but i'll be doing more over the week and uploading them as they come. when the hosting situation gets worked out i'll post the url and announce it on theDot

trysil sunday

coolo arrived this evening marking the end of the first wave of attendees. there will be a few more arriving mid-week, but we've got the bulk of the group here now.

one of the goals for kde4's interface is to be more organic, which is to say to behave more like the real world which we are well adapted to. in keeping with that we went for a walk out to a meadow for one of our two group meetings today.

towards a more natural interface

we shared our broad visions for kde, experimented with walking on bogs and climbed a tree or two. then it was back to the lodge for more hacking and in the late evening (though it's always hard to tell since it never gets particularly dark) we held a more structured event in-doors where we brainstormed first functional goals and then practical tasks. we put them on sticky notes and put them on walls in the main common room.

the wall of practical tasks

we arranged the practical tasks from least critical to most critical (they are all important, of course ;) from left to right. we then tagged them with some even smaller stickeys as to which were purely discussion meetings and which we were going to for sure tackle this week. then we each tagged those items with small yellow stickeys with our name on them which items we were committed to working on personally.

the idea was to emerge with some direction for the week ... and beyond. we witnessed the emergence of a new teamwork centered around kdelibs, something that has not always been so obviously there. as we closed the meeting (before going back to more hacking, at midnight no less) everyone shared what they hope to leave this week with, and resoundingly the consistent message was, "kdelibs in a state that has clear direction, is coherent and which can be used to start application work with."

looks like the days will be packed.

trysil

i arrived in trysil, norway last night for the kde four core library hackathon.

we are in a lodge with grass on the roof and pine wood covering most of the interior, except for the floors which are beautiful tile. the view is magnificent, looking down into a forested valley with a river running through it. the sun is shining, a light breeze is blowing. this is the perfect place to sit and hack without distractions.

there are already over 15 of us here with more set to arrive this evening. qt 4.2 is in qt-copy, kdelibs builds against it, there's a copy of kdebase in branches/work/bleedingedge that we will be working on against trunk/kdelibs, icrecream is humming and the meeting is coming to order nicely.

i've already seen some cool qt demos while here, including one that uses the java bindings ... they are amazing. the resulting apps are fast and good looking. other cool stuff are the new docker window animations in qmainwindow which are just gorgeous and smooth.

over the week i'll be blogging at least daily, though i'll try and post something in both the morning and the evening. we'll see how it goes. the entries will likely be a little technical heavy, but so be it =)

i get the feeling more than ever that kde4 really takes off here ..... =)