Hello!

I'm Lukas and my main job in KDE is to maintain KGet, the KDE download manager.

KGet has a long history in KDE, the first version being shipped with the KDE3 series. For KDE4 it was completely rewritten and redesigned.
I started porting it to KF5 a few weeks ago and I feel like the time is right to make it look more beautiful again!

It currently looks exactly like the version in KDE4, a screenshot of the main window can be found here.

I currently have no idea on how to improve it, I'm just so used to having it look the current way, but I'm very open to anything new! If there are any questions please don't hesitate to post here, mail me (lukas dot appelhans at gmx dot de) or chat with me on IRC (boom1992 in #kget).

Thanks a lot already!

Do you consider to improve the workflow, or do you just want to be complain with the VDG styleguide? The first one could mean to run a survey before the redesign. With VDG styleguide I think about a simple, flat layout (https://techbase.kde.org/Projects/Usabi ... ganization).

boom1992 wrote:Hello!

I'm Lukas and my main job in KDE is to maintain KGet, the KDE download manager.

KGet has a long history in KDE, the first version being shipped with the KDE3 series. For KDE4 it was completely rewritten and redesigned.
I started porting it to KF5 a few weeks ago and I feel like the time is right to make it look more beautiful again!

It currently looks exactly like the version in KDE4, a screenshot of the main window can be found here.

I currently have no idea on how to improve it, I'm just so used to having it look the current way, but I'm very open to anything new! If there are any questions please don't hesitate to post here, mail me (lukas dot appelhans at gmx dot de) or chat with me on IRC (boom1992 in #kget).

Thanks a lot already!

Hey, could you maybe include some more screen shot e.g. of some dialogues and explain what the use case of this application is?

First of all, thanks for the replies! I created a group of more pictures: http://imgur.com/a/OO7ZS

I would also consider improving the workflow, after all I just want to create an awesome product! Are there any examples of survey in KDE?

Thanks,

Lukas

boom1992 wrote:Are there any examples of survey in KDE?

You will find some studies here: http://user-prompt.com/de/portfolio/kde/

User centered design would start with the prerequisites of vision, persona, and scenario (an example: http://user-prompt.com/de/what-is-ktracks/)
Second step is to outline the requirements (http://user-prompt.com/de/what-do-you-r ... m-ktracks/)
And finally we can sketch a mockup (http://user-prompt.com/de/how-could-ktracks-look-like/)

What would a study be good for?
* How many people actually do use KGet? Why and why not?
* Do people have unsatisfied requirements with the current version? E.g. implementation of (basic) torrent features.
* Which aspects of KGet are annoying? E.g. cluttering dialog instead of plasmoid, notification stuff.
* When do people limit the bandwidth?
* ...
(I have to guess because I'm fine with the inbuilt FF downloader or wget. But I would take KGet into consideration if it works as easy as FF + flash downloader plugin and fits into the workflow. For example, to have the download progress and finish notification in the browser is the second best solution.)

If you want to keep the devil down in his hole, we can just talk about simple and flat design. That means, for instance, to hide the menubar, the toolbar, and probably as well the columns, and put all along with the configuration dialogs into a sidebar that slides in (?) on click at a menu button.
Why: because your (right now implicit) requirements do not include the interaction with items beyond pause/cancel. And you would say so because your vision is to be simple by default and powerful on demand. (I hope to get you right )

Which way do you wan to go?

I do agree that some user research would make sense here, however I do not consider the comment section or a poll embedded in a PlanetKDE blog post the right tool for this because
a) As a quantitative tool, it lacks representativeness
b) As a qualitative tool, it lacks depth

I'd rather go out and find a few people from the groups of
a) Existing KGet users
b) Potential KGet users
And then do more in-depth studies with them, ideally observing their use and doing some contextual inquiry interviews with them.

Potential KGet user here.
I'd be glad to take part in any studies that lead to a better UX.

Hello, I'm a long time user of KGet. Though, I think that too many of KGet's actions are hidden behind context menus and menu bars. Thus, my goal was to make it so the average user, who just wants slightly more control than just a list of past and current downloads that web browsers give.

png: https://imgur.com/Rh3WfjI
svg: https://www.dropbox.com/s/kztw5nq5njllp ... r.svg?dl=0

The first thing I did is see why Heiko recommended a Flat and Simple design pattern. It seems that it what is recommended for an application that only lists documents without depth (depth would be clicking on a chain of folders or actions).

The next thing I did was try to remove the menu bar. KGet includes a lot of functionality. Much of what's in "File", is what I would consider advanced or not frequenctly used tasks. This part led me to some questions that can be asked in the usability survey:

How many times does the average user important a list of transfers? Perhaps, importing/exporting is great for moving data to a new PC? That's a one time on average use case. I would think most just initiate the download from another application. Also, I wonder how many users know what type of file is generated with KGet's "Create metalink" functionality. These options can go to the "Menu bar" depicted as a [1] on the mockup. As for "History", I moved it to the toolbar.

[2] on the mockup is a filter bar. This lets' the user only show downloads from a specific group. The user can also use the bar to only show completed downloads as well.

[3] shows what became of the "Downloads" menu. This menu is where the main user experience changes comes from. In Dolphin, there is a button to help users that don't know how to select multiple files using the keyboard to do it completely by mouse. That is why there is now a checkbox column in the mockup. Also, instead of highlighting the entire download when using the keyboard to select multiple files, only the checkboxes will change states. This way, there is only one change when a user hits, "ctrl+A", all of the checkboxes will change state. So, the checkboxes don't change or remove KGet functionality, they just make selecting multiple files easier for users who live and swear by the mouse!

When multiple downloads are selected, the status bar on the bottom pops up (or I guess it can be made static if you prefer that). It consists of a dropmenu to select an action of what to do with the downloads. KGet needs to be smart here. It should show actions that are only available to multiple downloads if multiple downloads are selected. The user hits, the confirm button to make that action happen. Also, the user should be able to make it so a "confirm click" is unnesary. Making it behave just like the "Downloads" menu from the menu bar.

As before, the downloads are collaspable (and should defualt to being collapsed as well). Here, the first download is not collasped so as to show info. This way, the context menu is only needed for file managing options. However, those options can also be found by clicking the "open" button by the destination and using Dolphin itself.

I haven't gotten to the multiple files download dialog yet or transfer settings dialog, yet. I would prefer to have feedback on what I have at this point (in case this is to invavisve or not invasive enough). Plus, I don't really think those are considered features average users will need (again a survey is needed)

I wanted to make sure that KGet didn't change too much. It's already pretty simple to use (though that means I may only think it's already simple because I'm used to it).

If nothing else, I hope that I least brought up many/most of the points needed to make the survey.

Edit:
I noticed I had a lot of extra elements in the svg file that I didn't need. So, I cleaned it up, added tags, and moved the Filter bar to the left side of the app. Then, I also made a version with no status bar by placing the available actions for downloads in a new sidebar on the right. This seems more traditional.

Statusbar:
png: https://imgur.com/2FDDWB0,1iaqcqH#1
svg: https://www.dropbox.com/s/ekp97hmlwbja5 ... s.svg?dl=0

No Statusbar:
png: https://imgur.com/2FDDWB0,1iaqcqH#0
svg: https://www.dropbox.com/s/9ric4smq36crt ... r.svg?dl=0

CTown wrote:If nothing else, I hope that I least brought up many/most of the points needed to make the survey.

Um, I only saw one point for the survey in your post:

How many times does the average user important a list of transfers?

How do you mean your mockups and suggestions to be used for the survey? An initial user survey is meant to find out how people are using download managers and what they need from them. It would not make sense to start the survey from a mockup of a redesigned UI.
Although it would make most sense to wait for the results from the user survey/interviews/observations before proposing design changes, but of course if people already have ideas without research input, they can be presented in parallel to the survey, although that may mean they have to be adjusted later if they don't fit with the research results.

colomar wrote:
Um, I only saw one point for the survey in your post:

Sorry, I guess I wasn't as clear as I should have been. My points from my original post were:

1) How often do users import/export lists of files to download?
2) How many users even understand the "Create Metalink" option?
3) How many users use the "Transfer Options" settings?
4) Do users use the inbuilt File Manager options from KGet (versus opening the file in Dolphin for the same context menu options)?
5) Do most users consider "multiple file downloading" to be advanced?

I guess these could be added:
1) How many people use the Notification Center integration in KGet?
>a) Use the option to get individual downloads to show up in the notification center?
>b) Use the option to get a summary of what is going on in KGet in the notification center?
>c) Should this option stay in the "Advanced Setting"? I personally don't think so.
2) Do people still use the Drop target?
3) Which plugins do people use?
4) Any feature requests?

colomar wrote:How do you mean your mockups and suggestions to be used for the survey? An initial user survey is meant to find out how people are using download managers and what they need from them. It would not make sense to start the survey from a mockup of a redesigned UI.
Although it would make most sense to wait for the results from the user survey/interviews/observations before proposing design changes, but of course if people already have ideas without research input, they can be presented in parallel to the survey, although that may mean they have to be adjusted later if they don't fit with the research results.

In that case, thanks for the advice colomar. I guess a survey should be drafted. Probably should start with some use cases as well, right?

CTown wrote:Sorry, I guess I wasn't as clear as I should have been. My points from my original post were:

1) How often do users import/export lists of files to download?
2) How many users even understand the "Create Metalink" option?
3) How many users use the "Transfer Options" settings?
4) Do users use the inbuilt File Manager options from KGet (versus opening the file in Dolphin for the same context menu options)?
5) Do most users consider "multiple file downloading" to be advanced?

I guess these could be added:
1) How many people use the Notification Center integration in KGet?
>a) Use the option to get individual downloads to show up in the notification center?
>b) Use the option to get a summary of what is going on in KGet in the notification center?
>c) Should this option stay in the "Advanced Setting"? I personally don't think so.
2) Do people still use the Drop target?
3) Which plugins do people use?
4) Any feature requests?

Aha, much better, thank you for the clarification!
This is indeed valuable input for detail questions specifically targeting current KGet users.

colomar wrote:In that case, thanks for the advice colomar. I guess a survey should be drafted. Probably should start with some use cases as well, right?

Yes, it should. Apart from the detailed questions regarding specific KGet functions, the use cases (or rather user scenarios) are the most important part. And here, especially the potential (i.e. not yet actual) KGet users come into play. Existing KGet users have already adapted their behavior to how KGet works, whereas those who do not use it yet may show us the potential of more radical changes.

Here is my shot at creating personas based on new potential users.

Bill is a very hardworking, yet simple guy. He lives in a rural area. Unfortunately for Bill, his Internet connection isn't to reliable. Thus, Bill needs a download manager with the ability to resume a download once his connection becomes stable again.

Bill seems to be a user who can favor from KGet's ability to integrate into Plasma's Notification Center. Configure and forget!

Olivia is a proud resident of Canada. Though, she isn't very happy on how ISPs treat her fellow citizens. Like most Canadians, she has a relatively small monthly data cap limit. Thus, she uses KGet's "History" feature to watch how much she downloads.

Perhaps, KGet's History feature can take some cues from Android's ability to show the user how much data. Android even supports the ability to hard limit data cap usage?

Heiko Tietze wrote:
(I have to guess because I'm fine with the inbuilt FF downloader or wget. But I would take KGet into consideration if it works as easy as FF + flash downloader plugin and fits into the workflow. For example, to have the download progress and finish notification in the browser is the second best solution.)

If you want to keep the devil down in his hole, we can just talk about simple and flat design. That means, for instance, to hide the menubar, the toolbar, and probably as well the columns, and put all along with the configuration dialogs into a sidebar that slides in (?) on click at a menu button.
Why: because your (right now implicit) requirements do not include the interaction with items beyond pause/cancel. And you would say so because your vision is to be simple by default and powerful on demand. (I hope to get you right )

Taking cues from Firefox I tried to design a new view for KGet (I already tried getting rid of the menubar). This time, I tried to get rid of the columns. I call this new view "List View" and am referring to KGet's current view as "Detailed View". Firefox shows just the (1) filename, (2) size, (3) domain the file came from, and (4) the time finished for a completed download or stopped download. For a running download, Firefox shows just (1) the filename, (2) size, and (3) time remaining of the download. That's a lot less than what can be found in KGet.

Not only that, to make things "flatter" and simpler I removed collapsible downloads. Instead, the right side becomes a "Detail" view like in Dolphin. However, this means that details can only be found for one download at a time. I guess the question of "Do you look at at the details of more than one download at a time" needs to go on a survey to see how user reacts.

As for the toolbar, it only includes two actions: "New Download" and "Remove finished downloads". I feel that the latter is probably the most convenient button ever! The toolbar also includes three views. There's grid view (for an idea I just had), list view (which is shown in the following mockup), and detailed view (shown in the last mockup).

svg: https://www.dropbox.com/s/jize4lmywhys7 ... k.svg?dl=0
png: http://imgur.com/LNpOT0b

Edit:
After thinking about, that stausbar dropmenu might be a nightmare due to all different the syntax between languages are. Not only that, but those designs had three panes at once (Filters on the left, Actions on the bottom, and Details on the right). Actions are the most important of the three. I'm not sure how I didn't see that before... So, I removed the Filters from the next mockup (a user can just add it to a right or left sidebar if KGet ever gets a Filter panel). Then, I moved Actions to the right. It's pretty much how some of those "Gallery-Next" mockups were. Also, the ability to un-collapse a download to get info is back, to make things more similar to the current way of doing things. It's definitely still more busy than I would like. I need to add more room between elements in the "download widget". Also, I am looking for feedback of course! So, I present a hopefully simpler mockup:

svg: https://www.dropbox.com/s/am9mfqvofybuj ... e.svg?dl=0
png: https://imgur.com/szxylIa

Edit2:

Wow, I think something might be wrong with me. After all, I have to wake up in a few hours. I must be coming down with design fever...

Anyways, I added a Legend for consistency. I also removed the toolbar as Heiko suggested; that includes the menu bar. I think it all came out nice; though that could be the lack of sleep. Though, I guess a survey would leave clues to what the order of the Actions sidebar should be. I think this one mockup includes ever feature that I know KGet has. Though, this list view will be terrible for people who do many downloads. The only other issue that I can see is for people who want to reorder the list based on staus, time remaining, etc. just like how current KGet does. One solution to may be going back to placing all text in a horizontal fashion. This way downloads take less room and the user can double click on text to other by that text's category. At that point, the detail bar might have to come back. The only thing I can think have the option to go back to "Detailed View" if people prefer that (another thing that could go on the survey - what do people order their downloads by?). Any suggestions.
png: https://imgur.com/dpLD7OV

I think my computer is tired, too. As the multitouch just stopped working... Good night.

Nice work CTown. I also never can't wait until the user scenarios are done to make mockups

I prefere icons in the icon bar more than actions in a side panel. One thing I'll miss it the file mimetype icon. Additional informationes and special actions like transfer settings and source, tags would be nice to have in a sidepanel so when the user need this you can turn it on/off.

When you can't wait, what is no problem, start with a minimal interface setting in combination withthe VDG-Mockup toolkit. Maybe from the Firefox or Chrome downloader and then look what are the additional features from kget and put them to the place where the user would search, when he need this additional features. Of corse it is not necessary to have a 1:1 copy of a Firefox/Chrome download manager, because than people will use them, but simple as possible and powerfull when needed told us start with simple and show the user powerful when needed.

As I said, the mockups are realy nice and it is also nice to see the process behind as you show us with the different mockups. If you like you can use the VDG Owncloud storage for your mockups (viewtopic.php?f=285&t=123494).

Go ahead, when your computer is ready to work

The design is nice and goes in the right direction. Though I wouldn't use this KGet. What you do is to rearrange controls, redesign the layout, and get rid of features you are using rarely. Someone else might have different priorities. I wouldn't use KGet for the (nice) feature of resuming a download but for the global feedback, for instance. And if boom1992's vision is to have just a frontend for wget the approach has to be completely different anyway.

TL;DR: Third step done first.

I'm waiting for boom1992 who should decide what he wants from us.

I agree with Heiko on the procedure: While I like many ideas from your mockups, before we put a lot of effort into discussing and iterating them, I'd prefer that we do the user research.

Now the question is: Who will do it? I must say that I won't have the time to do the actual interviews at the moment, but I can of course provide guidance for whoever does it.
So who would be interested in gaining some experience in user research?