Please remember that poeple uses destops, because they want to work with the applications and not starring at the desktop because its such a great UI. If someone is new to linux-desktop, the first impressions, which are based on optical goodies and functionality, decide if he like the new ui or not. In this case an awesome UI is really important.

Giving users a lot of choices makes them often a little bit "lost in choices". Especially brand new users had often other problems as
to confugure the ui. I think, it would be difficult, to explain new users such things like "Activities", "Widgets", "Dock", .... which would come up, if there is a preset window at first startup.

Distributions have not enough manpower to provide their own ui. But a distrobased, specialized, KDE-UI could be something like a trademark for this distro. Distros know which packkages are configured together and which package contains e.g. Lancelot or a special widget. So, in my opinion, you should support the distributions to create their own "thing", not the users.

In general a good idea.

PS: Sorry if my criticism sounds really bad, i did not ment it so. My english is just to bad to write such things in a more friendly manner...

franku wrote:Giving users a lot of choices makes them often a little bit "lost in choices". Especially brand new users had often other problems as
to confugure the ui. I think, it would be difficult, to explain new users such things like "Activities", "Widgets", "Dock", .... which would come up, if there is a preset window at first startup.

I don't think people need to understand any of those concepts in order to decide which layout they prefer. We can just show them a few screenshots (it should certainly not be more than a handful, and they should differ substantially) and they can select the one they feel most comfortable with. I'm pretty sure people can recognize an application launcher, a system tray or a taskbar visually without knowing much about the concepts, simply because those are familiar to them.

Maybe i miss the point. What is the issue here?

1. Predefined layouts (solely optical things)
2. Predefined UI (which includes a different workflow on the desktop)

A preset window for the first option (layout) seems to be ok for me. But the second option could imho not be satisfied with a "simple" preset window. Beause this means, the user make a choice in the preset window and after a while he recognized, that he could not work with it. So he has to test other presets...

franku wrote:Maybe i miss the point. What is the issue here?

1. Predefined layouts (solely optical things)
2. Predefined UI (which includes a different workflow on the desktop)

A preset window for the first option (layout) seems to be ok for me. But the second option could imho not be satisfied with a "simple" preset window. Beause this means, the user make a choice in the preset window and after a while he recognized, that he could not work with it. So he has to test other presets...

As far as I've understood, we're talking about the selection and placement of panels and widgets, and I think that can be seen in a visual preview. What aspects of workflow other than selection and placement of panels and widgets do you have in mind?

I read russh statement and thought, he would like to bring a new UI:

russh wrote:We can do 'Unity', we can do a Dock.

e.g. Unity is more than a panel on the left with autohide. It includes a HUD with a global menu.

Can we have some options to take this forward instead of a new (Konky, icon, colour scheme, wallpaper)? Rather than refreshing the design, can we evolve?

This sounds to me, that he want more than a new (graphical)Design.

If i am wrong, sorry for confusing.

franku wrote:I read russh statement and thought, he would like to bring a new UI:

russh wrote:We can do 'Unity', we can do a Dock.

e.g. Unity is more than a panel on the left with autohide. It includes a HUD with a global menu.

Can we have some options to take this forward instead of a new (Konky, icon, colour scheme, wallpaper)? Rather than refreshing the design, can we evolve?

This sounds to me, that he want more than a new (graphical)Design.

If i am wrong, sorry for confusing.

He wants more than just changing purely the visuals of elements, but he also said that these things are all already possible with today's technology. Things like a global menu, for example, can easily be shown in a picture. I wouldn't just show the desktop in that picture, but an application window as well so that things like menu placement, but also widget and windeco style can be seen.

I wouldn't just aim for e.g. an "as close to Unity as possible in everything" option anyway. If people want Unity, they should install Unity, not Plasma. However, if they're used to and autohide dock on the left plus global menus and want to have that in Plasma as well, we can create a preset which does that and visualize it in a screenshot.

On many forums is a heading like " My awesome desktop". Such a topic here on this forum could bring up some samples for the idea.

But i am sceptical about success because the problem of unmet package-dependencies per distro persists.

franku wrote:On many forums is a heading like " My awesome desktop". Such a topic here on this forum could bring up some samples for the idea.

But i am sceptical about success because the problem of unmet package-dependencies per distro persists.

The layouts offered should only consist of things which are installed by default, of course. If we think that for example a certain Plasmoid is useful enough to be part of one of the default layouts we offer, why not make it part of the base set of Plasmoids that all distributions shipping Plasma install anyway?

I'm a bit skeptical about the "show options first" dialog. From my observation, I would say that most "common" users don't care about the appearance of their desktop except for the wallpaper, they just want something that looks reasonably nice. For those users, the dialog will only act as an irritation ("woah, what's this now?"), and they'll likely just click "Cancel" because that's the default "I have no idea what this is, get me out of here!" button.

The users who do want to customize their workspace will usually find the options themselves. It would be nice if there was an easier way to change the theme and layout of the workspace though, similar to EraX' mockup.

As a general principle, I would not present anything that's not absolutely necessary at startup/first launch. Remember those "Tip of the day" dialogs back in the days? They sound like a nice idea - present some useful tips to the user when an application is launched - but in practice, how many users actually read those? Everyone I know would just close them without a second glance, because they just wanted to use the application.

To summarize, I would advocate for an awesome default that works well for most users (remember that distributions can change it), and a simpler process for changing the theme and workspace.

Hans wrote:I'm a bit skeptical about the "show options first" dialog. From my observation, I would say that most "common" users don't care about the appearance of their desktop except for the wallpaper, they just want something that looks reasonably nice. For those users, the dialog will only act as an irritation ("woah, what's this now?"), and they'll likely just click "Cancel" because that's the default "I have no idea what this is, get me out of here!" button.

The users who do want to customize their workspace will usually find the options themselves. It would be nice if there was an easier way to change the theme and layout of the workspace though, similar to EraX' mockup.

As a general principle, I would not present anything that's not absolutely necessary at startup/first launch. Remember those "Tip of the day" dialogs back in the days? They sound like a nice idea - present some useful tips to the user when an application is launched - but in practice, how many users actually read those? Everyone I know would just close them without a second glance, because they just wanted to use the application.

To summarize, I would advocate for an awesome default that works well for most users (remember that distributions can change it), and a simpler process for changing the theme and workspace.

^This. I aggree on all points: Don't bother the user with unnecessairy stuff, but let him start as soon as possible exploring the system - but with a nice default config!

Nuc!eoN, Hans: Your arguments are based on assumptions about what users want and don't want. We cannot know what they want or don't want until we've asked them (even though we might think we could). That's why I'd prefer to make a mockup of such a dialog with a few presets, show it to some people (not just geeks, but our friends and family, "normal people") and ask them
a) Which one they'd choose and why
b) Whether they'd find such a choice right after installation useful or not

If different people pick different layouts and give better reasons for their choice then "I don't know, it just looked nice..." plus say that they'd find such a choice useful, we have real-world data indicating that it would indeed be useful for them. If they cannot make an informed choice and/or do not find it useful, we should scrap the idea.

@colomar:

Sure, I can only base it on my own experience with a small sample (I teach college students Linux/programming and have helped people install Linux). But I also believe that it's possible to infer what the majority care about by looking at what's popular. E.g. Apple products are not exactly known for their customizability, but rather for how simple they are to use.

Something like this would also make helping new users a nightmare (hyperbole, yes). Imagine a user who doesn't have any idea what's going on and asks how to launch an application. Other than the distribution and version, we would also have to ask which desktop preset they use. And I bet that most people don't remember what they chose. So we would have to play the game of guessing the preset in order to do something simple as starting an application.

To use my own experience: if my students were able to choose their own preset the first time they log in, it would just add another problem for me. I would have to tell the students to "make sure you don't choose a different thing in the window that pops up, just click OK", in order to ensure that everyone will be able to follow what I'm doing. And of course there will always be one or two who still choose a different present and become confused.

Letting the user choose their preferred desktop preset during the installation makes it a bit different, but I can still see it making giving support and writing documentation harder. The advantages are heavily outweighed by the disadvantages. In that case it's better to make a default that rock, in my opinion.

With that say, it seems like there are others who are more positive about the idea: http://sessellift.wordpress.com/2014/04 ... s-for-real

Hans wrote:With that say, it seems like there are others who are more positive about the idea: http://sessellift.wordpress.com/2014/04 ... s-for-real

Actually, that was my trick to "test" the idea on more people before continuing the discussion by using both the April fool and the "real" blog post to introduce it to the public. You have your opinion based on your experience, I have my opinion based on mine. By publishing the idea in my blog, I was able to broaden the discussion and get input from more people.

Your concerns based on your experiences in your teaching scenario make perfect sense... in your scenario. If I were you, I wouldn't want my students to all have different setups, either. However, this is not about being able to choose an Experience at startup, it's about whether you allow customization or not. If you don't offer an Experience Picker but allow your students to customize Plasma manually, I think you have an even bigger "nightmare", because people trying to manually fiddling with Plasma settings without knowing what they do are prone to messing things up pretty bad (at least that's what many people have told me from their experience).
Are your students currently able to customize their desktops? If so, do they do it?

However, even if on the one hand I like the idea of different types of configuration for different use cases and to show the great flexibility of Plasma, on the other hand I think Plasma should have its own unique style, which just see immediately recognize, moreover, the different styles may well be imitated, but never work as good as the original, and would require too much attention, perhaps you could better devote to other things. I rather like the possibility to create Unique Bundle, ch econ one shot you can transform the entire look, but this only after installation and user's choice

I would add, ch ein really something, maybe you could do with the different "activities" by default, however, since even today, I think it's one of the things not used by users, and myself, sometimes I do not understand fully the 'utility