Reply to topic

Rethinking Activities

AGuiFr
Registered Member
Posts
77
Karma
0
OS

Re: Rethinking Activities

Sun Aug 09, 2015 7:45 am
alake wrote:So permit me to ask thread participants some difficult questions for a moment.
  • If all we can come up with is another abstraction layer for users to organize their work space, then why shouldn't most users just use Virtual Desktops that they're already familiar with and that provides enough workspace organization utility for the needs of most users?
  • Is that really the only thing of value that Activities provide to the user?
  • If not, then what other cool stuff can the user get done with the underlying capabilities that Activities provide? Ignore the current UI, or any repackaging of the current UI. Consider the underlying capablities.
  • If we can't come up with good answers for the above questions, are we really "rethinking' activities, or are we just rearranging furniture?

We have been mainly discussing window management for activities because there is currently some overlap between the functionalities provided by virtual desktops and those provided by activities regarding this point. Better defining the goal of each tool regarding windows management is important because this overlap creates confusion for the user. But everybody here agrees that window management is only a small aspect of activities.

I think the what of activities and its added value compared to virtual desktops is pretty clear for all of us discussing here. That is why we were more discussing the how. But you are right, I realise that it was never explicitely mentionned in detail in this thread. We all started discussing activitie with our own prior knowledge, but someone jumping into the discussion without a deep knowledge of activities could have the impression that it is mainly about window management.

The goal of Activities is to enable the user to have different customisations of his workspace adapted to his usual activities. Some examples of this customisation could be :
  • specific sets of widgets on the desktop, so that the relevant folders are more easily available.
  • specific sets of windows opened (if you open an application in an activity, it will not be displayed in another)
  • specific history (file recently accessed, applications recently used, recent contacts, etc...)
  • specific shortcuts (application launchers in the task bar, favorites in the file browser)
  • specific power management profiles (screen fading time)
  • specific network management profiles (proxy, VPN connection)
  • specific notification behaviour (all notifications, only important notifications, quiet)
  • specific default instant messaging status
  • specific number of virtual desktops
Moreover, as Activities are for regular activities of users, the state of the applications should be saved (documents opened, lay out of windows) when switching from one Activity to another. When the Activity is started again, even if the computer was shut down in the meanwhile, the user should find his computer in the exact state where it was when he left it.

Most of these features are already available or easily implemented, and windows management is only a very small subset of them. Activities have already some things to offer to users right now. Some users like Psifidotos or me find a real benefit in using them, but we had to make a big effort to understand the concept by reading about it on the internet, experimenting a lot with it until we found the right use. Most people who tried didn't see the benefit of it because the effort to discover the features is too big. So before thinking about what other user needs could be fulfilled by activities, we need to make a better job at presenting the features which are already there. Part of that requires to remove the overlap with VD. That is what we have been focussing on lately. Another aspect is to centralise everything related to activities in a central place. Finding a more descriptive name than "Activity", which is a very abstract term which different people don't understand the same way, could also help. For me, we don't need to rethink the concept of activities itself, but how it is conveyed to the user. Yes, it may be rearranging what is already there, but it is needed first.

I think I covered all the points of your questions, but I am wondering if you were expecting us to start from the beginning by describing a user situation identifying a need, explaining how it is fulfilled without activities and what activities can bring that improves the situation. I would be very interested in reading your own answers to these questions.
User avatar psifidotos
Registered Member
Posts
86
Karma
0
OS

Re: Rethinking Activities

Sun Aug 09, 2015 1:31 pm
alake,

I think that the current conversation has created a lot of common ground between the members of the discussion. I will make an attempt to articulate some very important points even though AGuiFr's earlier position to summarize is a fantastic base "here is a little summary. For the sake of simplicity,.." . Feel free to correct me!!! :)
  • the terms Activities and VDs are not sufficient enough from the user's perspective based on the technology offered (for VD I think we all agree, for the term Activity there is a running discussion) ,(personal position: the conversation shows that the term Room is fantastic in order to explain Activities but not sufficient enough as a term, I will definitely use it to explain but I would prefer something a little more abstract and not used so often in the regular speaking -I fancy a lot Sights(Activities) and Canvas(VDs) ;D - )
  • the new terms must be simple but descriptive enough for the casual user. (personal position: I agree with ken300 in this, even my earlier proposed term WorkArea possibly does not fit in this one)
  • there should be a definition for each of the new terms, the definition for the old terms is not sufficient enough. (personal position: fantastic job in this one from arucard and AGuiFr )
  • there should be a learning process for the user. personal position: the stage's articulated from arucard are extremely helpful in order to create a learning curve process. I didnt had it my mind so clearly defined, and that helps a lot in the mockups and the designing process
  • both Activities and VDs are fantastic technologies and we need them both, in a way that they operate logical and intuitive between them. (personal position: for the current way that these technologies are introduced to the users I feel that some furniture must be moved and some other dropped or hidden a lot in the garage for the advanced users. I feel sometimes like there is a refrigerator in the living room and an oven in the bedroom)
alake wrote:I think innovation happens when we solve problems that users either didn't realize they had, or for which they couldn't find or imagine a solution. :-)

of course :)

alake wrote: So permit me to ask thread participants some difficult questions for a moment.
  • If all we can come up with is another abstraction layer for users to organize their work space, then why shouldn't most users just use Virtual Desktops that they're already familiar with and that provides enough workspace organization utility for the needs of most users?
  • Is that really the only thing of value that Activities provide to the user?
  • If not, then what other cool stuff can the user get done with the underlying capabilities that Activities provide? Ignore the current UI, or any repackaging of the current UI. Consider the underlying capablities.
  • If we can't come up with good answers for the above questions, are we really "rethinking' activities, or are we just rearranging furniture?

  • I follow ken300's and AGuiFr's position in this one
  • no
  • I follow AGuiFr in this one... "The goal of Activities is to enable the user to have different customizations of his workspace adapted to his usual activities. Some examples of this customization could be..."
  • before adding I find it very important to clean up a bit and move some things. There is a reason why Activities are not being used a lot from the casual users (make an exception for me :) ) . Before adding something in the living room, it is very important to see if all in it makes sense...

I would also interested a lot in your position alake in your previous questions...
AGuiFr
Registered Member
Posts
77
Karma
0
OS

Re: Rethinking Activities

Sun Aug 09, 2015 8:05 pm
I have been reading the blog post from Ivan Čukić on "private activities" : http://cukic.co/2015/08/09/activities-c ... -settings/

I noticed that he has a "main" activity. I also use a "main" activity, which I use when I don't do something particularly activity oriented. I guess most activity users have a default one. Do you, Psifidotos ? So why not make it like that in the design itself ?

Here is another naming idea :
  • Your workspace represents the space which is available for you to interact with your computer.
  • You can extend your workspace, with additional areas (or desktops), but only one area is visible at a time.
  • You can create some alternative workspaces. An alternative is a modification of the workspace that you do to suit a regular activity. Almost everything is the same, except some specific settings which can be altered.

The settings available in system settings would affect the main workspace, and all alternatives which use the default setting. The size of the main workspace (i.e. the number of areas/desktops) could be different of the one of each alternative you have created. When you create a new alternative, by default, it clones the main workspace. But you could also clone an existing alternative.

Apart from the differentiation between "main activity" and "other activities", I am just proposing a naming change. All features would remain the same. To sum up :
  • virtual desktop -> area (of the workspace)
  • activity -> alternative (of the main workspace)
What do you think ?
User avatar ivan
KDE Developer
Posts
916
Karma
14
OS

Re: Rethinking Activities

Sun Aug 09, 2015 10:57 pm
Hi all,

First of all, I want to thank everyone willing to help making activities more prominent.

I have been following the discussion, although I refrained from commenting in order not to influence it. So, don't think that all the ideas you are proposing here are ignored just because you see a blog post where something is implemented, which is not previously agreed on here. I just can not allow this to stagnate, it has already been dormant for a long time (especially UI wise).

A few of the more general impressions I'm getting from this thread are (I'm sorry if this sounds harsh, it is not my intention):

* having a discussion on a variety of topics in a single place is not really productive. It is very hard to pull out concrete ideas from a huge wall of text where some parts focus on one thing, and other on something completely different;
* the second thing that I'm personally having problems with is when people write a lot of text (yes, I'm doing the same right now) because, again, important parts get watered down - this is a problem even when a discussion is focussed on a single topic;

The main problem I see is that people have different ideas of what activities are / should be. It sometimes seems like everyone (not only in this thread) has something that he desires Plasma to have, and somehow tries to explain the activities idea in the terms which will cover that specific missing feature.

What I think this thread needs is for us to collect all the ideas as bullet points (maybe in a new thread :) ), with *short* pros and cons for each. To check which fit the personas, and which can fit well together. Then, which are possible to implement and how.

As a separate topic, after the above image has crystallized, we can talk about the potential renaming of VDs and Activities. Not before, as far as I'm concerned (although there were a few nice proposals).

Cheers

----

"I noticed that he has a "main" activity. I also use a "main" activity,"

Not really, I just hate writing 'Test' when creating a dummy activity - I like to have a more sane name. I'm not using the 'main' activity (maybe I should, but that is another story :) ).


Image
User avatar Heiko Tietze
Registered Member
Posts
593
Karma
0
OS

Re: Rethinking Activities

Mon Aug 10, 2015 7:28 am
I read your blog post and it's good to have Activities implemented again in KDE. To not go too much into detail I agree with AGui's comments. Putting facts on the table is probably what we need.

During the VDG meeting yesterday we talked about P5.5 and the idea to declare a lighthouse project. Activities would be an interesting topic but starting the discussion goes immediately into very basic questions (what is it good for, how do we call the baby etc.). Personally I have a pretty clear picture of how the workflow could be (maximizing flexibility in order to let users find their own use cases). However I feel not being the driver here since other topics are stalled as well.

To put that all into an actionable question: How do we proceed? Keep the Activities implemented as you did with probably a few minor updates on the UI, or discussing alternative options?
arucard
Registered Member
Posts
53
Karma
0

Re: Rethinking Activities

Mon Aug 10, 2015 8:15 am
ivan wrote:I have been following the discussion, although I refrained from commenting in order not to influence it. So, don't think that all the ideas you are proposing here are ignored just because you see a blog post where something is implemented, which is not previously agreed on here. I just can not allow this to stagnate, it has already been dormant for a long time (especially UI wise).

Good to know that we're being useful :P
ivan wrote:A few of the more general impressions I'm getting from this thread are (I'm sorry if this sounds harsh, it is not my intention):

* having a discussion on a variety of topics in a single place is not really productive. It is very hard to pull out concrete ideas from a huge wall of text where some parts focus on one thing, and other on something completely different;
* the second thing that I'm personally having problems with is when people write a lot of text (yes, I'm doing the same right now) because, again, important parts get watered down - this is a problem even when a discussion is focussed on a single topic;

The main problem I see is that people have different ideas of what activities are / should be. It sometimes seems like everyone (not only in this thread) has something that he desires Plasma to have, and somehow tries to explain the activities idea in the terms which will cover that specific missing feature.

What I think this thread needs is for us to collect all the ideas as bullet points (maybe in a new thread :) ), with *short* pros and cons for each. To check which fit the personas, and which can fit well together. Then, which are possible to implement and how.

As a separate topic, after the above image has crystallized, we can talk about the potential renaming of VDs and Activities. Not before, as far as I'm concerned (although there were a few nice proposals).

I understand your impressions and generally agree with them (and don't worry, they didn't sound harsh to me), but I don't really agree with your solution. I think these impressions are correct, but they are due to the nature of what is discussed here.
  • As for the first point, I don't think we really discussing a variety of topics, rather one topic (Activities) that simply has many different aspects to it (VDs, configuration, multiple devices, user perception, user workflows, etc.). I think this is also why from time to time you'll find a post suggesting we focus on what is important. I think this is similar to what you suggest for a new thread but as you can see it's not necessary to create a new thread for this.
  • I can't really argue with your second point (since you can see I'm doing the same right now). It's a bit difficult to follow, but I think the summary posts that have been made a few times should help others follow and join the discussion more easily. But I don't think it's necessarily a bad thing to write long texts. In a discussion such as this, you often need to explain your point of view which I believe can't always be done in a few well-formed sentences. But we could try to maybe highlight some of the main points (like in psifidotos' most recent post and I've tried to do it in this post as well) or provide a TL;DR version for people not interested in the details (which I also do sometimes, when I feel it is warranted).

As for the discussion itself. I think alake's questions were pretty useful even if just to point out that we do actually agree on what Activities are. But from those answers, I noticed that we mostly defined Activities as alternative computer configurations, but none of the names we've suggested actually reflect this. I think the names we've been suggesting mostly try to relate to the window management part of Activities (which is what we've mostly been discussing due to the overlap with VDs). So I think in trying to think of names for both VDs and Activities at once, we may have been skewing our name suggestions in a way that they relate to VDs. I would like to point out that in these answers we've given, I think we've all given pretty much the same relationship between Activities and VDs; VDs provide extra space, Activities provide an alternative configuration (which includes VDs). As you can see, the relationship between Activities and VDs is pretty indirect so maybe it's time we start treating it that way (I think this is also reflected by the fact that it's been mentioned multiple times that the user should be able to use Activities and VDs independently from one another).

I think that we should just keep the name Virtual Desktop for now. It is a well-established name, but if necessary it can be changed later. For now, we need a more descriptive name for Activities, where I think we need to focus on the fact that it's an entirely different and separate configuration of your device. So I think names like "Device Configuration", "Alternative Configuration", "Virtual Configuration" or even "Virtual Device Configuration" could be used, where you could also use something like "Setup" instead of "Configuration".
Heiko Tietze wrote:To put that all into an actionable question: How do we proceed? Keep the Activities implemented as you did with probably a few minor updates on the UI, or discussing alternative options?

I think that most of us have pretty much agreed that a name-change with some minor updates on the UI would be the best way to proceed. There is already some level of agreement on how the UI should be changed, but there has been no real agreement on a new name other than that we do actually need a new one.

As for a summary of the discussed UI changes (please correct me if I'm wrong, I also tried to apply some priorities using the MoSCoW method):
  • there must be a separate UI for the VD Switcher as well as the Activity Switcher/Manager(Still some disagreement, see below)
  • The Activity Switcher/Manager should allow you to view and change the settings of each Activity (with something like a settings button for each Activity)
  • The VD and Activity switcher UI's should be available as Desktop Effects, which should allow alternative switchers
  • The VD switcher could display the list of active Activities as well, though this list should clearly be separate from the VDs, e.g. through a dropdown menu or maybe the List part of a List-Detail pattern.
  • The VD switcher could contain a link or button to go to the Activity Switcher/Manager instead.
I think the main disagreement in this UI part of the discussion is whether to limit VDs so they can only be laid out horizontally or vertically, which would allow us to display Activities in the other 2-dimensional direction. This has the benefit of providing a clear overview of both Activities and VDs in 1 UI, also allowing you to easily move things from one Activity to another. It has the downside of making Activities seem too similar to VDs which could confuse users and it limits the way VDs can be configured.
User avatar ivan
KDE Developer
Posts
916
Karma
14
OS

Re: Rethinking Activities

Mon Aug 10, 2015 10:14 am
I'll try the bold for TL;DR format.

> As for the discussion itself. I think alake's questions were pretty useful even if just
> to point out that we do actually agree on what Activities are. But from those answers,
> I noticed that we mostly defined Activities as alternative computer configurations,
> but none of the names we've suggested actually reflect this.

I don't agree with this statement.

These are the things that were mentioned previously (that I recall) regarding activities that I do not agree with:

  • Activities are related to devices and form factors
    The idea of convergence that when you have a tablet UI that magically turns into normal Plasma when you plug in the keyboard has nothing to do with activities (apart from both ideas coming from Aaron). The connection exists for one reason - because the term 'activity' was used in the first few releases of Plasma Desktop and Plasma Netbook for desktop layout and set of widgets, and some people still think in that direction.
  • Activities are meant to support different configuration profiles
    This is not the purpose of activities (the term 'configuration profile' is kinda well established, so that name would be used if this was the case), that is just a side-effect.
  • We should change the name, and then ...
    I don't see the name as the biggest problem. If you need to change the name in order to specify the idea of activities, you are essentially creating a new idea. That is, bending activities to do something else that you'd like to have (as I previously wrote).

The idea of activities, as initially envisioned and designed, was to allow the user to focus on a specific task at a point in time.

From that idea, the following came out:

  • Different widgets on the desktop (done)
    When you work on one task, you want the related files, applications, some specialized widgets shown.
  • Power management settings (done)
    This one is a feature that is as far as possible from the idea of activities. The activities should not have any connection to hardware. The problem is that there are some (for me corner) use-cases like you want to be in the performance profile when gaming, or you do not want your system to go to sleep when you are doing a presentation.
  • Usage statistics tracking (done, mostly done plasma-wise)
    Since the aim is to make it easier to focus on specific tasks, recent documents, important documents and similar should be tied to the task. You should not see a recent document from 'Gaming' activity in the 'Some project' activity and similar.
  • Per-activity favourite files / applications / contacts (some done, some not)
    The same as above. The difference is here that the user explicitly marks something to be important for a specific activity.
  • Spacial separation of windows in different activities (done)
    This one collides with VDs a bit. Rationale: since you need to focus on a task, you do not want to see applications from other activities. This is mostly because applications do not internally support activities. The correct approach would be for some like kmail, telepathy etc, to be on all (most) activities, but to modify what is shown depending on the current activity.
  • Stopping / starting applications in activities (aka, the evil buggy X11 session stuff)
    This one is one of the most praised features that does not work correctly, and that is not likely to change unfortunately. We'll see...
  • Automatic activity switching based on environment (location, time of the day, etc.) (not done)
    This is one of the things I desired for a long time and which we started doing in Plasma Active, but that never got further than a proof of concept (I didn't really like the proposed workflow we had back then). When you take your laptop to work, it would be nice if the activity automatically switched to 'Work'. In Plasma Active, we had a scoring system for this, and nothing was automatic - it is only the activity switcher that did show a yellow star on the activity that it thinks will be the most likely chosen one.
  • Form factors do not enter into it
    If you are in the 'Work' activity and get an email that you should fill out your time-sheet, you realize that it is tedious to type on a touch screen. You plug in your keyboard (plasma adapts, etc.) - you haven't changed your current task. You are still working on the same thing, thus you are in the same activity.


I do not want to comment on the proposed names. I do like some for VDs, I even like one alternative one for Activities (though, I do not consider it a significant improvement at this point, just one that might have been acceptable when this baby first got its name),


Image
User avatar hook
Registered Member
Posts
197
Karma
0
OS

Re: Rethinking Activities

Mon Aug 10, 2015 11:32 am
As probably many in this thread I have a very big interest in Activities and also my own workflow and a strong idea how to use it.

The reason why I haven’t participated in this thread recently is that just as Ivan, I’m feeling that there is a lot of brilliant ideas, but scattered throughout a too wide spectrum of depth.

I think at this point we have to decide on what aspect of Activities we want to concentrate ourselves:
  • pure tech, thrown at users to do with whatever they want (i.e. relatively workflow-agnostic)
  • a specific workflow that would be part of Plasma experience, of which Activities are part of the tech that enables it

Currently we’re doing a bit of both and that’s confusing. For pure tech discussions it’s too specific, for a specific workflow, we’re not thinking outside of the box enough.
i think that we should follow the Plasma principle here: Simple by default, powerful when needed, with that specifically I would propose that we continue in the following order:
  • get a moderator for this thread to make sure discussions are on topic (or find a better communications channel)
  • decide upon some useful workflows that Plasma should provide by default – simple by default
  • figure out how to present those workflows in UI/UX
  • figure out how to code those workflows using Activities, VD, Baloo, Akondi, … whatever is needed to get it done.
  • make sure that people who have a different workflow can still use Activities as they wish, preferably with nicer tools and in a way that would enable them to share their workflows with others – powerful when needed

In short:
The way I see it the point of this thread is not rethinking Activities per se, but rethinking how Activities could (and maybe should) be used/expanded.


It's time to prod some serious buttock! ;)
AGuiFr
Registered Member
Posts
77
Karma
0
OS

Re: Rethinking Activities

Mon Aug 10, 2015 6:13 pm
I do agree with all the critics of our discussion here. But as arucard, I think sometimes it is needed to detail your point of view to make it understandable by others. I think using bold to highlight important points would be useful. Using a wiki as a complement could be a way to sum up the main points discussed. I have no idea if it is possible to create a page on one of KDE's wikis and to edit it by logging with a KDE user identity. I guess it is. I would allow us to carry on discussing here (but with shorter posts, got it ;) )

I also agree with the points made by hook : we discussed too specific points, mainly because I first made a mockup on another thread and came here to explain the reasoning behind it. And regarding activities, we are not thinking out of the box enough. I tried to in the last few posts.

As for remarks on some more specific points :
ivan wrote:The idea of activities, as initially envisioned and designed, was to allow the user to focus on a specific task at a point in time.

Thanks a lot for your input. As users, we can only explain what we understand about activities from the different features which are already implemented. You were part of the people who designed the concept, so without surprise, you are the one who give the best explanation of the concept. I think this goal is still valid and doesn't need to be changed. This should be the first line of the wiki page.

ivan wrote:Activities are meant to support different configuration profiles
This is not the purpose of activities (the term 'configuration profile' is kinda well established, so that name would be used if this was the case), that is just a side-effect.

But this is a direct consequence of the goal described above. I can't think of any other way to "allow the user to focus on a specific activity" than allowing him/her to customize his/her settings to this task (settings being understood in a large way, encompassing which windows are displayed and not). Or am I missing some other implications of the concept?
User avatar alake
Registered Member
Posts
591
Karma
3
OS

Re: Rethinking Activities

Mon Aug 10, 2015 6:49 pm
ivan wrote:The idea of activities, as initially envisioned and designed, was to allow the user to focus on a specific task at a point in time.


Thanks so so much Ivan for providing this clarity. This, to me, is the fundamental proposed benefit of activities to the user:

To allow the user to focus on a specific task.

This is what I think we should use to guide the discussion. I'll commit to starting a separate thread with this as primary overall user goal against which any proposed design should be judged.

From that idea, the following came out:

  • Different widgets on the desktop (done)
    When you work on one task, you want the related files, applications, some specialized widgets shown.
  • Power management settings (done)
    This one is a feature that is as far as possible from the idea of activities. The activities should not have any connection to hardware. The problem is that there are some (for me corner) use-cases like you want to be in the performance profile when gaming, or you do not want your system to go to sleep when you are doing a presentation.
  • Usage statistics tracking (done, mostly done plasma-wise)
    Since the aim is to make it easier to focus on specific tasks, recent documents, important documents and similar should be tied to the task. You should not see a recent document from 'Gaming' activity in the 'Some project' activity and similar.
  • Per-activity favourite files / applications / contacts (some done, some not)
    The same as above. The difference is here that the user explicitly marks something to be important for a specific activity.
  • Spacial separation of windows in different activities (done)
    This one collides with VDs a bit. Rationale: since you need to focus on a task, you do not want to see applications from other activities. This is mostly because applications do not internally support activities. The correct approach would be for some like kmail, telepathy etc, to be on all (most) activities, but to modify what is shown depending on the current activity.
  • Stopping / starting applications in activities (aka, the evil buggy X11 session stuff)
    This one is one of the most praised features that does not work correctly, and that is not likely to change unfortunately. We'll see...
  • Automatic activity switching based on environment (location, time of the day, etc.) (not done)
    This is one of the things I desired for a long time and which we started doing in Plasma Active, but that never got further than a proof of concept (I didn't really like the proposed workflow we had back then). When you take your laptop to work, it would be nice if the activity automatically switched to 'Work'. In Plasma Active, we had a scoring system for this, and nothing was automatic - it is only the activity switcher that did show a yellow star on the activity that it thinks will be the most likely chosen one.
  • Form factors do not enter into it
    If you are in the 'Work' activity and get an email that you should fill out your time-sheet, you realize that it is tedious to type on a touch screen. You plug in your keyboard (plasma adapts, etc.) - you haven't changed your current task. You are still working on the same thing, thus you are in the same activity.


Again, this helps so so much. This nice thing about this list is the rationale that relates each implementation piece back to the overall vision of allowing the user to focus on a specific task. Thanks!

For clarity, I'm not suggesting that Ivan or anyone else did not possess this clarity. I'm fairly confident they did. I'm just not sure that we were all on the same page. Now I think we are. :-)

So per Ivan's request, I start a clean thread. The new thread will focus on:
  • Ivan's clarification that foundational concept for activities is to "allow the user to focus on a specific task". Can I call this the "vision" for Activities?
  • Ivan's summary of implementations that are done/not done/limited in support of the foundational concept/vision.
  • Start a discussion on some "simple by default" user stories/workflows that directly support the foundational concept/vision.
  • A summary of the specific ideas from this thread so we can figure out which ideas best serve those user stories/workflows.
  • A link to the VDG comunity wiki that will be a collection point for settled design decisions from that thread.
User avatar ivan
KDE Developer
Posts
916
Karma
14
OS

Re: Rethinking Activities

Mon Aug 10, 2015 8:14 pm
alake wrote:So per Ivan's request, I start a clean thread. The new thread will focus on:


Not that I requested anything, (rather insinuated :) ) but I think that might be helpful.

alake wrote:Can I call this the "vision" for Activities?


+1 from my side. The only downside to the sentence is that people might include playing relaxing music to help them focus, which is not the point. :)


Image
arucard
Registered Member
Posts
53
Karma
0

Re: Rethinking Activities

Tue Aug 11, 2015 9:18 am
alake wrote:The new thread will focus on:
  • Ivan's clarification that foundational concept for activities is to "allow the user to focus on a specific task". Can I call this the "vision" for Activities?
  • Ivan's summary of implementations that are done/not done/limited in support of the foundational concept/vision.
  • Start a discussion on some "simple by default" user stories/workflows that directly support the foundational concept/vision.
  • A summary of the specific ideas from this thread so we can figure out which ideas best serve those user stories/workflows.
  • A link to the VDG comunity wiki that will be a collection point for settled design decisions from that thread.

I think what AGuiFr said about how customizable configuration is a direct consequence of the first point above should also be included. It's also what I was trying to describe when I called it a Device Configuration. Though ivan's points of features made me realize that this includes more than I initially realized (not just system configuration). I think the functionality of Activities should be seen as though the user has got a separate device for each Activity and optimized it entirely to suit the needs of one specific task. So switching between Activities would be like switching between these task-optimized devices. Other than "Device Configuration" (or maybe "Device Instance"), I couldn't think of any more-descriptive name for this.

I would also like to suggest that the vision be expanded to include multiple devices. So it would be something like "allow the user to focus on a specific task, possibly on multiple linked devices". These linked devices should be assumed to be the user's personal devices, though in a later stage it might be interesting to be able to link devices from different users, allowing them to collaborate (if that's even possible or useful). Note that this has nothing to do with convergence as it's just about synchronizing Activity switching across devices. So when I put my computer in the Work Activity, my phone and tablet go into the Work Activity as well, but my HTPC isn't linked in the Work Activty so that remains as it was. At a later stage, it might be nice to add some data transfer to this though that might also just be handled by something like KDE Connect.
ivan wrote:I don't see the name as the biggest problem. If you need to change the name in order to specify the idea of activities, you are essentially creating a new idea. That is, bending activities to do something else that you'd like to have (as I previously wrote).

I think the name Activity relates to the user's goal of "focusing on a specific task" whereas the feature's name should relate to the solution of how to do this. This is why I don't think that changing the name would actually entail changing the idea behind it.

The main reason for this, which is also why I think Activities have been seen as confusing, is that it's easy to understand the mental link to the user's goal from the name "Activity", but the user's goal doesn't really have a mental link to the name "Activity". I hope I can explain this clearly enough. When you look at what an Activity does, it's pretty easy to understand that it's about allowing a user to focus on a specific task by allowing several customizations and providing filters on your device. But users will usually need to understand it the other way around, so they will need to understand that when they want to focus on a specific task they can use "Activities" for this. The user would look for things that might help him/her focus on their specific task, but when they encounter something called "Activities" I don't think the user will recognize this as what they need. This is why I think the name should relate more to the solution than the goal, because the user will be looking for a solution (even though the functionality is implemented by trying to achieve the goal).

Another (possibly less important) reason for this is that users can come up with their own, different solutions to this goal of focusing on a specific task. At which point the name that relates to achieving that specific goal is no longer relevant for that user. Conversely, the user may want to use Activities to achieve different (currently unforeseeable) goals, where the current name is also not applicable.

I think that the new thread should include the suggestion for a name change as, for reasons mentioned above, I do think it's necessary. It would be a shame if Activities got massively improved and no one notices it because they still haven't been able to see that Activities might be useful to them while working in the Plasma desktop.
User avatar ivan
KDE Developer
Posts
916
Karma
14
OS

Re: Rethinking Activities

Tue Aug 11, 2015 3:55 pm
> I would also like to suggest that the vision be expanded to include
> multiple devices.

+1 This was one the planned things - but it is a behemoth to implement.

> I think that the new thread should include the suggestion for a name
> change as, for reasons mentioned above, I do think it's necessary

I'd advise to create a separate one for that.

It is not likely to happen from what I've seen so far. Most of the names I've seen proposed (here and elsewhere), are either as vague (generic) as Activities, or they are already used for something else.

> the feature's name should relate to the solution of how to do this

I'm not sure I consider this to be a fact.


Image
arucard
Registered Member
Posts
53
Karma
0

Re: Rethinking Activities

Tue Aug 11, 2015 5:48 pm
ivan wrote:> the feature's name should relate to the solution of how to do this

I'm not sure I consider this to be a fact.

I actually didn't mean for this to be a fact. It is my opinion and I give my reasons for it in the next part of the post (At the part starting with "The main reason for this"). I might not have been clear enough about that.
User avatar ivan
KDE Developer
Posts
916
Karma
14
OS

Re: Rethinking Activities

Tue Aug 11, 2015 5:53 pm
I sometimes have creative wording of stuff :) Meant to say that while I do understand your point, I don't really agree. Not opposed, but still I don't agree.

That is why I want the vision and ideas first, naming later.


Image

 
Reply to topic

Bookmarks



Who is online

Registered users: Baidu [Spider], Bing [Bot], Exabot [Bot], farid, Google [Bot], Google Adsense [Bot], gregorymartinez, Hito, Sogou [Bot]