Task-centric system planning

Sat Dec 29 17:28:38 UTC 2012

Several people didn't make it. Some were late arriving.

Log attached.

Carl

On Sat, Dec 29, 2012 at 8:40 AM, Marco Martin <notmart at gmail.com> wrote:
> On Saturday 29 December 2012, Thomas Pfeiffer wrote:
>> > >
>> > > 14:00 - 16:00 CET in our IRC channel #active.
>> > >
>> > > I hope to see at least everyone who agreed to that date on Doodle, and
>> > > maybe  some more (Aaron? Sebas?)
>> >
>> > I think I've put it in the agenda already, sorry for not checking the
>> > doodle, but we should be fine. :)
>>
>> The meeting is on now. Marco? Carl? Sebas? We're waiting for you... ;)
>
> sorry i couldn't make it, poorly scheduled conflicting stuff
> completely gone out of my mind :(
> are there backlogs available?
>
> --
> Marco Martin
> _______________________________________________
> Active mailing list
> Active at kde.org
> https://mail.kde.org/mailman/listinfo/active
-------------- next part --------------
[Saturday, December 29, 2012] (times are US-Pacific; CET - 9hours)

[05:02] <bbalazs> Hi all - sorry for being late...
[05:02] <bbalazs> task-centric UI meeting will be here and now?
[05:03] --> bogdan has joined this channel (~bogdan at 37-8-178-215.coucou-networks.fr).
[05:04] <colomar> Hi bbalazs! Yes, it will be
[05:04] <colomar> Others are yet to arrive as well, so it's no problem that you're a little late ;)
[05:05] <colomar> And hi bogdan!
[05:05] <bogdan> hi, are we still waiting to begin the discussion ?
[05:05] <colomar> yes
[05:06] <colomar> kallecarl is still away, I'm not sure about sebas
[05:06] <colomar> notmart isn't here yet, either
[05:13] <colomar> Maybe we can give them a few more minutes. I'd like to have at least one of the core devs here for technical feedback
[05:13] <colomar> If nobody else shows up, maybe we can start discussing some ideas and then get technical feedback later
[05:14] <bogdan> what are the plans for switching to Qt5 ?
[05:15] --> niqt has joined this channel (~quassel at host139-13-dynamic.247-95-r.retail.telecomitalia.it).
[05:16] <colomar> I have no idea ;)
[05:16] *** trueg_away is now known as trueg.
[05:16] <bbalazs> colomar: I have a rather hard end at 1600, so I think it would be good to start
[05:17] <bbalazs> the log will be available anyhow :)
[05:17] <colomar> Okay
[05:17] <bbalazs> colomar: so what is the goal of this meeting?
[05:17] <colomar> (btw: Anyone else in this room is free to join the discussion, of course ;) )
[05:18] <colomar> The goal is to have some ideas that we think are worth working on
[05:19] <bbalazs> concerning the idea of a task centric approach in the UI?
[05:19] <colomar> Those ideas should then be developed further during this development cycle so that ideally, designs are ready for the next cycle
[05:19] <colomar> Yes, how to implement the idea in Plasma Active
[05:20] <colomar> Or, rather, steps toward it for now
[05:20] <bbalazs> yeah - I would not even know yet how a ui could look like ;)
[05:21] <colomar> Since you've probably thought about this most, maybe you can start with telling us which ideas you have so far?
[05:21] <bbalazs> ok...
[05:21] <bbalazs> very basic though :)
[05:22] <colomar> Well, we're just starting, so basic ideas are fine ;)
[05:23] <bbalazs> When you look at the way computers work, you find that they are divided by applications. I think the main reason for this is that it is the way of sharing the market in non-free software environments. Not the way users would like to do thinks - which is - as we know from psychology - rather taks centric than tool centric
[05:24] <bbalazs> Now in free software we do not have to care about market shares or such - so the idea is to forget about the traditional thining of applications and find a way for more flexible tools that help us to fullfill our tasks
[05:26] <bbalazs> let me give you a small example: When I want to writer a letter to a friends birthday, first my calender reminds me, that there is a birthday, then I have to open my adressboook, copy the adress (if possible at all), open any writing application, find my right template and handle a UI that I could write a novel with, just to write down a few lines for my friend...
[05:27] <bbalazs> How cool would it be to have this done in one step: I see a birthday reminder and magically a UI opens with the right template, the adress filled in and a UI I need for writing a short letter...
[05:28] <bogdan> so you are proposing to add another software layer on top of the existing tools so that the task of writing a short message is automatized
[05:29] <bbalazs> I am not proposing anything technical at the moment :) - I just see the flaws and think this is something only free software can do better (or put is the other way around - non free software will probably not be able to achieve this)
[05:30] <colomar> bbalazs: Agreed. I see this scenario as more of a long-term goal though, since it probably means rewriting many components
[05:30] <bbalazs> I also see that activities is a great step towardds this idea, as it already allows to be task specific to some extend
[05:30] <colomar> I agree that in thze ideal world, it's possible to combine any number of different components to any number of workflows
[05:31] <bogdan> actually we need to reuse as many existing components as possible. That might be one starting principle
[05:31] <bbalazs> bogdan: true - we have all the needed underlying technology
[05:31] <bbalazs> I have no clue about a good way for the transition though
[05:32] <colomar> bogdan: Yes. I think that during the transitional period, we'll probably have to string several "classical applications" together to create a task
[05:32] <bbalazs> We could start by a random task, just to gather some experience with it
[05:32] <colomar> KParts will surely be helpful for us as well
[05:32] <bogdan> maybe a list with most useful tasks shall be needed. A voting system could be put in place
[05:33] <colomar> bogdan: Do you mean voting for tasks to implement first?
[05:34] --> pinheiro has joined this channel (~pinheiro at 2.80.36.94).
[05:34] <bbalazs> bogdan: I do not even think we need to start with the most needed things - a good task would be one we could learn a lot from 
[05:34] <colomar> bbalazs: I think one crucial part we'll have to come up with soon is the UI which is used to start a task
[05:34] <bbalazs> colomar: true
[05:34] <bogdan> colomar: I mean first defining tasks to implement, then choosing the most useful features, how these features could be implemented with existing components
[05:35] <bogdan> a small set of most useful tasks could be a good start
[05:35] <bbalazs> colomar: as most people usually only do few different taks with their computer, it should be easy to create and share these - however we call the resulting UIs
[05:36] <bbalazs> this could be done via khownewstuff (or whatever it is called)
[05:37] <colomar> bbalazs: Sounds useful. Finding a good way to start a task will still not be easy, though. On a system with a keyboard, it might be a good idea to just have the user type in what she wants to do, but that's not very convenient on a touch device...
[05:37] <bbalazs> colomar: true 
[05:37] <colomar> bogdan: Finding the most important tasks is important, but maybe the next step after defining a general way to implement tasks
[05:38] <bbalazs> It will be a very long way untill this will work well - and I am only talking about the UI perspective here as I have no clue about the technical stuff :)
[05:38] <bogdan> colomar: The user input should not be limited to the touch screen. I personally prefer to have severel ways of doing it: touchscreen, keyboard is available, sensors
[05:39] --> luisgabriel has joined this channel (~lampih at 186.212.104.69).
[05:39] <colomar> The immediate challenge we're currently gfacing is the following: Calligra Active currently can only read files. Therefore, the current entry point to it is always selecting a file to open (from Files or the Activity Screen)
[05:39] <colomar> bogdan: Agreed
[05:40] <colomar> So currently there is no need to even have Calligra Active in the Launcher, which is a good thing
[05:40] <colomar> However, now the Calligra Active team plans to implement file creation
[05:41] <bogdan> colomar: you mean that there is a supperposition between caligra active and okular active ?
[05:41] <colomar> The usual way would be to have a Calligra Active "Start Screen" from which users wozuld choose which kind of file to create. This would be 100% appp-centric
[05:42] <bbalazs> for things like office suites I think we could start with defining UIs for the most common types of tasks, like I need a different UI for writing a letter than when I write a mathematical book
[05:42] <colomar> What we want now is for users to just tell the system "I want to write a letter", and then the3y're automatically presented with the UI for it, without having to know that it's Calligra Active they're using
[05:43] <bbalazs> colomar: perfect
[05:43] <bbalazs> that is a good start
[05:43] <bbalazs> if we have this, it is only a small step towards automatically adding the address and fetching the right template for writing letters
[05:43] <colomar> bogdan: No, Okular Active is still an ebook/PDF reader. Calligra Active is for reading - and later creating - office files
[05:44] <bogdan> colomar: multiple choices are a good thing. It looks to me that the task centric approach is similar to using a template where the user can configure what she wants to do
[05:44] <bbalazs> bogdan: it will sort of UI templates
[05:44] <colomar> yes
[05:45] <bbalazs> one thing we have to ease for that is working with stuff like qt creator 
[05:45] --> pvdm_ has joined this channel (~pvdm at pvdm.xs4all.nl).
[05:45] <colomar> Oh and btw: we'll probably not use GHNS for this, but the Add Ons App
[05:45] <bogdan> okular active can still read office files (*.odt), but they are presented as static images, no change is possible => useful only for reading
[05:46] <colomar> Ideally, users would go to the app and instead of choosing apps to install, they'd choose workflows and the system would automatically fetch everything needed for that task
[05:48] <bogdan> bbalazs: could you explain a little more about working with qtcreator ?
[05:48] <bbalazs> colomar: I dont understand - go to the app?
[05:48] <colomar> Okay, I admit that was confusing ;)
[05:49] <colomar> I mean "Use the Add-Ons App"
[05:49] <colomar> Or "Go to the store" ;)
[05:49] <bbalazs> I my dream world it would be so easy to creat UIs on top of existing technology that end users simply can create their own task-solution UI and share it with others
[05:50] <bbalazs> this would need some software like qt creator, just simpler
[05:50] <bogdan> bbalazs: like saving the UI template for later use or sharing it
[05:50] <bbalazs> bogdan: yes
[05:51] <bogdan> bbalazs: that should be the primary task of the task centric interface, no need to a programming tool. The user selects the tools she wants then saves the entire workflow as a task
[05:51] <-- pvdm_ has left this server (Ping timeout: 265 seconds).
[05:52] <bbalazs> colomar: so they would tell the store: I like to write personal letters on the occasion of the birthday of a friend - and then install this task rather than installing applications
[05:52] <colomar>  bbalazs: Yes, that's what I was thinking
[05:53] <colomar> It would install apps, templates, everything that's needed
[05:53] <-- icota has left this server (Ping timeout: 248 seconds).
[05:54] <bbalazs> bogdan: seeing the increadible ammount of possible tasks, we will have to crowd-source creation of these - so there would be the need of some sort of configuration tool that not-so-technically-gifted people can use for that
[05:55] <colomar> And since Aaron describes the store system as quite flexible (aka "You can offer pretty much everything in it", I guess this should be possible
[05:56] <bogdan> bbalazs: yes, that's why this tool should be the task-centric interface
[05:56] <colomar> bbalazs: Yes, but I think this is probably still a very long-term goal. Offering a way for users to connect existing UIs to create task-specific tools would already help users a lot
[05:57] <bbalazs> colomar: yeah very long term - as I said, we should just pick random tasks and start gathering experiences with them, before we can start to make it generic in any way
[05:57] <bogdan> colomar: not sure how you will cope efficently with several tools: you need to propose the user a list with choices each time she wants to use a task, or change the server should take care of changing the task with most popular tools
[05:58] <bbalazs> bogdan: that is an unsolved UI problem
[05:58] <colomar> I think the "Write a letter to XYZ" could be a nice example (although it's probably not really a common task for a tablet): It would fetch address data from a contact, choose a letter template, fill in both sender and recipient address and let the user write the actual content
[05:59] <bbalazs> I have been talking to inge (calligra) and volker (kmail) about it and they would probably be in such an experiment
[05:59] <bogdan> bbalazs: I would call this a project
[06:00] <-- mck182 has left this server (Ping timeout: 264 seconds).
[06:00] <bbalazs> bogdan: The UI is solvable, when you consider that not all people need to be offered all types of tasks in all situations - so smart limitation is the key to a successful UI there
[06:00] <bbalazs> bogdan: true :)
[06:01] <bbalazs> and again, activities come into play nicely there
[06:02] <bogdan> bbalazs: well, you actually try to remove the load on the user side when trying to complete a task. Anyway, the environment is changes, so you need to allow the server to update the tools the task is using
[06:03] <bogdan> bbalazs: for the beginning static task would be enough I guess
[06:03] <bbalazs> bogdan: agree
[06:04] <colomar> bogdan: Yes, I think so, too. If there are several tools available for a given task, I think we should offer users a choice when they install a task, not every time they start it
[06:04] <bbalazs> colomar: so what are the plans for the calligra writing mode?
[06:05] <bogdan> colomar: in a long run, the user will have the tendency to use different tools, so periodic task update migtht be needed
[06:06] <colomar> bbalazs: The only thing I know is that they plan to add file creation capability but they are not sure about the UI for it. And that's where we should come in
[06:06] <colomar> I don't want them to create a "Calligra Active Start Screen"
[06:06] <bbalazs> bogdan: I think this is one of the experience we have to gather with a project like this
[06:06] <bogdan> bbalazs: agree
[06:06] <bbalazs> colomar: please not!
[06:06] <bbalazs> :)
[06:07] <bbalazs> esp. on a tablet you will not write a master thesis I guess
[06:07] <colomar> And that's why we'll have to have somethin in place for them to use soon
[06:08] <bbalazs> understand
[06:08] <bogdan> colomar: tablets can be used with keyboards, but this is not the primary task for a tablet
[06:08] <colomar> So we can tell them "Just implement interface XYZ which we can call with parameters ABC, we'll do the rest"
[06:09] <bogdan> colomar: you mean tests :)
[06:09] <colomar> By "we", I mean the Plasma Active team as a whole, of course ;)
[06:10] <bbalazs> So if that is the current need, what can we offer there?
[06:10] <colomar> Nothing yet ;)
[06:11] <colomar> I think we'll need to have a general idea of where we're heading and then think about the UI to start a task first
[06:11] <colomar> Since we'll need that sooner or later anyway
[06:11] <bbalazs> ok, what tasks would make sense on a tablet that include calligra writer
[06:11] <bbalazs> ?
[06:12] <bogdan> the idea of having a task centric interface is interesting, but at a higher level. And generating the workflow for a task is something worth to start implementing (to be put on a features list)
[06:12] <colomar> bogdan: yes
[06:13] <colomar> Well... One example might be to start writing a rough draft of a text document on the tablet which would then be worked on further on the desktop/laptop
[06:13] <bogdan> among the most useul tasks: reading e-books, review a paper (meaning to be able to write comments), review code 
[06:14] --> icota has joined this channel (~quassel at 31.45.217.87).
[06:14] <bogdan> colomar: making a review would be something doable on a tablet
[06:14] <colomar> I guess mostly users won't create a document on a tablet but rather only edit them a bit while going over them once more before a meeting
[06:15] <bbalazs> yaeh commenting / bookmarking on something you read seems reasonable to me
[06:15] <colomar> That would be easy to do. Still, I'm pretty sure users would ask for a way to create new documents sooner or later
[06:15] <bbalazs> taking meeting notes / preparing a meeting is also a good case
[06:15] <colomar> yes
[06:16] <bogdan> bbalazs: I hardly beleave that someone is able to take notes at the presenter speed with a tablet
[06:16] <bbalazs> colomar: we do not have to disable the creation of something new right now
[06:16] <bbalazs> bogdan: with a good UI, perhaps  ;o)
[06:17] <colomar> And "taking notes" does not necessarily mean "writing lots of text" ;)
[06:17] <bbalazs> I like the meeintg notes thing - it could be started from the calender somehow
[06:18] <colomar> ...or from an Activity created specifically for a meeting
[06:18] <bbalazs> or an activity gets created from the meeting in the agenda
[06:18] <bbalazs> that includes the notes :)
[06:18] <bogdan> well, for taking notes you should be able to use a pen and use automatic handwriting recognition. Which is already done by comercial OSs
[06:18] <colomar> We've always advertised that: You create an Activitx for a meeting, which has all related documents, contacts and the event in it
[06:19] <bbalazs> which is nothing different that a task specific ui
[06:19] <bbalazs> or approach
[06:19] <colomar> definitely
[06:20] <colomar> That's why I thought that the Activity Screen may be a good starting point for tasks (though maybe not the only one)
[06:20] <colomar> Or, rather, sub-tasks
[06:20] <bbalazs> it definitely is closely related
[06:20] <bogdan> I would go for Caligra Active being able to do handwriting recognition for editing docs
[06:21] <bbalazs> bogdan: that is a feature noone will say no to, I guess
[06:21] <colomar> bogdan: I'd prefer a general component which can be used with every application
[06:22] <bogdan> colomar: indeed, maybe such component already exists
[06:22] <colomar> Perhaps
[06:23] <colomar> Last time I looked (a few years go), there were no really good open source handwriting recognition systems on ther market yet, but that may have changed by now
[06:23] <colomar> That's a different story, though ;)
[06:24] *** trueg is now known as trueg_away.
[06:24] <bogdan> win 8 with notes application allow to write notes, but no sure about handwriting recongnition
[06:24] <colomar> Let's recap. What do we need for the "big picture":
[06:25] <colomar> - Tool componjents, as many and as atomic as possible
[06:25] <colomar> - An UI to connect the components to form a task workflow
[06:26] <colomar> - A way to share / get them
[06:26] <bogdan> Calligra Active: edit document is ok, using either the keyboard or the touch pad (with handwriting recognition if possible)
[06:26] <bogdan> touch screen I mean
[06:26] <colomar> - A tool to easily create new UIs
[06:26] <colomar> - A way to start a task with given parameters
[06:27] <colomar> Anything else?
[06:27] <bbalazs> I think that is a good summary
[06:27] <bogdan> popular task definition
[06:27] <colomar> bogdan: I think this is a precondition for the first bullet
[06:27] <bogdan> colomar: pl
[06:28] <bogdan> colomar: ok
[06:28] <colomar> Since it's usede to prioritize component development
[06:28] <colomar> -e
[06:28] <bbalazs> and the starting point will be the meeting situation?
[06:29] <bogdan> do we put handwriting as a feature to integrate in AP (see CellWriter) ?
[06:30] <-- icota has left this server (Ping timeout: 248 seconds).
[06:30] [Away] You are no longer marked as being away.
[06:30] <colomar> For Calligra Active, Activities as starting point may be enough. I'm not sure if we can start every task from an Activity, though
[06:30] <bbalazs> colomar: no definitely not
[06:31] <bbalazs> just for this first try
[06:31] * kallecarl apologizes for missing most of the meeting
[06:32] <bbalazs> idea: out of Kalender you should be able to create a new activity that allows you to take notes and such
[06:32] <colomar> kallecarl: You haven't missed it completely yet ;)
[06:32] * kallecarl scanned the log
[06:32] <kallecarl> one comment
[06:32] <kallecarl>  the way of sharing the market in non-free software environments.
[06:32] <kallecarl> this is true but not the big issue
[06:32] <kallecarl> people are habitualized this way
[06:32] <kallecarl> and we have the ability to look at a big, big picture
[06:33] <kallecarl> I would like to see something audacious
[06:33] <kallecarl> cuz KDE can do this Apple and MSFT can't
[06:34] <kallecarl> colomar: 's big big ideas about task-centricity
[06:34] <kallecarl> pardon the interruption
[06:34] <colomar> kallecarl: Yes. So are we not audacious enough yet? ;)
[06:34] <kallecarl> colomar: some is
[06:34] <kallecarl> i love handwriting stuff
[06:35] <bbalazs> kallecarl: that is what at least i try to express ;)
[06:35] <kallecarl> and that sort of rethinking from the users' points of view
[06:35] <colomar> And they're actually bbalazs' big big ideas ;)
[06:35] <kallecarl> I work with ordinary users quite a bit and they _always_ are file centric
[06:35] <kallecarl> trained that way
[06:36] <bbalazs> colomar: it is the result of numerous discussions ;)
[06:36] <kallecarl> and then there's another big concept that I didn't see mentioned
[06:36] <kallecarl> nepomuk
[06:36] <kallecarl> it's supposed to be my brain annex
[06:36] <colomar> yes
[06:36] <bogdan> kallecarl: I would say that this depends on the used device: on tablets they are mostly task centric in my opinion
[06:36] <kallecarl> so why not meeting activity & nepomuk watches what I do
[06:36] <bbalazs> yeah nepomuk is part of the technology we need for doing this kind of stuff
[06:37] <kallecarl> bogdan: yes
[06:37] <kallecarl> had an experience last week ... woman wanted to show me a particular picutre of the space shuttle
[06:37] <colomar> Absolutely. This is something where Nepomuk can finally really shine
[06:37] --> mcamaret has joined this channel (18d6accc at gateway/web/freenode/ip.24.214.172.204).
[06:37] <kallecarl> opened photo app and then looked through thousands of pictures
[06:38] <bogdan> kallecarl: not a good ideea to move processing on the tablet side. It might need to much processing power and storage
[06:38] <kallecarl> the way she found it was by what other picture were nearby and when she took them
[06:38] <mcamaret> ergh! i'm late! slept in.
[06:38] <kallecarl> bogdan: yes
[06:39] <kallecarl> please don't stop the note-taking use case conversation to address my idea
[06:39] <kallecarl> this has to start with something doable
[06:40] <kallecarl> processing and other heavy lifting on server side makes sense 
[06:41] <bbalazs> I think the note-takeing discussion has as far as we can go for this meeting
[06:41] <bogdan> colomar: for the summary I would prefer something less genering in order to have developers to work on
[06:42] <bbalazs> bogdan: I think it is UI first in this particular case - as developers have already worked on all these great bits and pieces thata we need to put together now :)
[06:42] <colomar> bogdan: Development probably won't happen during this cycle anyway, so we can have more meetings until something happens
[06:42] <bbalazs> the cycle is 3 weeks?
[06:42] <colomar> nope
[06:42] <colomar> the whole PA4 cycle
[06:43] <bbalazs> ah, ok, that gives us some time indeed
[06:43] <colomar> I'll check how long it's supposed to last...
[06:44] <colomar> mcamaret: Hi! Erm, sorry but I don't recognize your nickname. Who are you? ;)
[06:44] <bbalazs> just one more thing: I do not claim this to be a substitue for applications - they need to be around too!
[06:44] <bogdan> colomar: maybe schedule some meetings for specific tasks (once we know how to save the task workflow and make it available for the user)
[06:44] <colomar> March 25th 2013: Final tagging of PA4
[06:44] <colomar> February 25 2013: Beta
[06:45] <mcamaret> colomar: some kde/tablet/task-centric ui obsessed kid in the -5 timezone
[06:45] <bbalazs> So in every task I need to be able to load the full application at any time as a fallback
[06:45] <mcamaret> following the PA mailing list
[06:46] <colomar> mcamaret: okay, so welcome and feel free to jump right in :)
[06:47] <colomar> bbalazs: Yes, but in my ideal world, even the full application UI would be started by a task. I would like to get rid of as many generic app launchers as possible
[06:47] <kallecarl> bbalazs: not sure about that...what full application should be loaded?
[06:47] <kallecarl> reading app is different from note-taking app
[06:48] <bbalazs> As I said - sort of a fallback mode - I cannot reach my goal with the current limited veiw - go to the full app
[06:48] <bogdan> bbalazs: I would consider this to be two different tasks
[06:48] <kallecarl> that sounds like "task centric" to start and then app-centric
[06:48] <kallecarl> bogdan: yes
[06:49] <bbalazs> kallecarl: practically I think (i might be wrong here) that at every step we will have a modified UI of an existing application - connected intelligently with each other.
[06:49] <kallecarl> I don't have a solution ! but it seems like the progression should be task -centric and then task-centric in another way
[06:50] <colomar> I agree with kallecarl and bogdan here
[06:50] <kallecarl> "practically" might get in the way
[06:50] <colomar> If I start a simple task with a simple UI and then realize that my task is actually more complex than that, I should start a new tas (with the same file)
[06:50] <kallecarl> not in the way of development but rather conceptualizing task centricity
[06:51] <bogdan> bbalazs: let's keep things simple for now. Try to find a way to create tasks and use them
[06:51] <bbalazs> When I do enter notes, I will be in calligra - even though I might not notice that. But when my UI fails (e.g. not enough formating options), I need to go to something larger - in this case I would have to be able to start a full blown calligra
[06:51] <bbalazs> I was just meant as a side note anyhow
[06:51] <bbalazs> :)
[06:51] <kallecarl> this is key bbalazs
[06:52] <kallecarl> these are the kinds of examples that make sense of the task-centric way
[06:52] <mcamaret> so, will calligra have a simple rich-text ui in that case? a word processor doesn't really do notes well.
[06:52] <bbalazs> mcamaret: this is what we will have to define :)
[06:52] <kallecarl> "when my UI fails, I need to go to something larger". true, but the user shouldn't have to switch to app-centric mode
[06:53] <bogdan> bbalazs: in my understanding task centric UI is more about how to transfer and process data till the final tool is started (e.g. write some message to someone)
[06:53] <bbalazs> kallecarl: true
[06:53] <bbalazs> bogdan: I would not say there is a final tool - there is a necessary toolchain
[06:53] <kallecarl> bogdan: and the device starts the tool, right
[06:53] <kallecarl> right?
[06:53] <colomar> Yes. Users may realize they need more functionality at some point, but we don't want them to go the the app launcher and start Calligra Active from there in that case
[06:54] <bogdan> kallecarl: yes, but is more like defining a workflow of actions
[06:54] <kallecarl> bogdan: I like it
[06:54] <bbalazs> I have to leave soon - colomar thanks kicking this discussion of
[06:55] <bbalazs> I will leave the chat open, so I can follow up on the discussion later. Wishing all of you a good start into 2013!
[06:55] <kallecarl> colomar: ++ I think that this is an opportunity for KDE to make a considerable contribution
[06:55] <colomar> Okay, so how do we proceed? Maybe schedule another meeting for defining and assigning more specific tasks?
[06:55] <kallecarl> bbalazs: thank you
[06:56] <bbalazs> colomar: wikipage + new meeting
[06:56] <kallecarl> colomar: summary of this meeting... new meeting
[06:56] <kallecarl> wiki++
[06:56] <bogdan> colomar: please write a short summary in AP site with these ideeas on task centric UI
[06:56] <colomar> Yes, I'll put a meeting summary in the wiki
[06:57] <bbalazs> bye!
[06:57] <bogdan> colomar: I would go definitively for more specific tasks
[06:57] <bogdan> bbalazs: thank you
[06:57] <colomar> bbalazs: bye, and thanks for joining us!
[06:57] <kallecarl> colomar: I would like to get more of your university research into the mix
[06:57] <mcamaret> bye all
[06:58] <colomar> bye mcamaret Hope you'll be able to participate in the next meeting from the start ;)
[06:58] <mcamaret> me too ;)
[06:58] <bogdan> I'll leave too, happy new year everyone
[06:58] <kallecarl> colomar: IOW...if we could start from nothing, what would task-centric look like?
[06:58] <kallecarl> bogdan: bye thank you
[07:00] <colomar> kallecarl: I'm working on that, but it will probably take a while for a university project to materialize. Plus, such a project may require to be partly business-financed. In that case we might have to find some company willing to invest into that
[07:00] <colomar> bogdan: Bye, thanks, happy new year
[07:00] <kallecarl> colomar: it's the kind of project that we can get funded most likely
[07:01] <kallecarl> we have an interesting intersection of possibilities
[07:01] <kallecarl> activities ... task centricity ... nepomuk ... Frameworks 5 modularity
[07:02] <kallecarl> building blocks that need a different kind of interstitial connections
[07:03] <kallecarl> colomar: aren't you already doing things related to HCI? must be some research on that which is not bound by the habit of app-centricity
[07:04] <kallecarl> colomar: great conversation you started here
[07:04] <kallecarl> thank you
[07:04] <kallecarl> I gotta go too
[07:04] <colomar> Yes, I am doing HCI research
[07:04] [Away] You are now marked as being away (reason: Away for now).
[07:04] <colomar> I'll try to get something going. Would be great
[07:05] <kallecarl> colomar: it is possible for KDE to do outlandish stuff
[07:05] <colomar> yes
[07:05] <kallecarl> maybe not PA4
[07:05] <colomar> definitely not. Topics for PA are already pretty much decided
[07:05] <colomar> PA = PA4
[07:06] <mcamaret> wow, community.kde.org hates me
[07:06] <kallecarl> hmmm...I haven't followed HCI closely lately
[07:06] <kallecarl> mcamaret: prolly not
[07:06] <mcamaret> i'm always getting invalid http headers
[07:06] <kallecarl> in the kde community...all you need is love
[07:06] <colomar> *ggg*
[07:08] <colomar> I'll post another Doodle soon for the next meeting (I hope some core PA devs can be there then, too=