5 ms·
Free OS on your phone is great and all, but it's got that quirky linux UI problem where usability is an after thought to "me too" features. For example, I can'
by afro88 11y ago
Free OS on your phone is great and all, but it's got that quirky linux UI problem where usability is an after thought to "me too" features.
For example, I can't read any of the text on the phone image on the front page, and physically the image on my screen is bigger than my iPhone. So the text is unreadable in the OS by default?
- tho9Ohx1eo 11y agoJust watch the video. Everything is similar to a current Android phone.
- jbk 11y agoThe front page is just a marketing thingie. Look at the video. It looks great.
- veeti 11y agoYou must have some very low standards for "looking great". Like they say, this is a prototype and it shows. The design is all over the place. The one thing that sticks out the most is that there is no sense of scale at all. Everything is either ridiculously huge or tiny. For example, look at the compact status bar: really small text and icons. Then look at the navigation icons on the bottom of the screen, the dialer tab icons, the contact pictures: all huge. But then there's the text next to each contact picture: tiny again.
- wz1000 11y agoEverything you mention is relatively easy to fix. Submit a patch if you don't like it. That's the beauty of open source :)
- veeti 11y agoAh, the good old "submit a patch if you don't like it". This is a common problem all throughout the user interface shown, not an isolated issue. And as mentioned, it's a _prototype_: I'm going to give them the benefit of the doubt and say this is far, far from the final user interface anyway.
- wz1000 11y agoIn the end, it is a community effort. It needs people from the community to contribute. If you think you can do something better, I really encourage you to do so.
- keithpeter 11y agoUI design - how would someone with strong views (and experience!) on gestures and UI features but with little coding experience get involved? Just put some mocks somewhere and post them to where the UI team meet? Edit: In fact, where do the people interested in UI design actually meet and make their sausages in KDE? I can't seem to find a central place similar to http://design.canonical.com/ http://design.canonical.com/ There is this gap between having the programming knowledge to 'submit a patch' and other kinds of large scale organisation/design skill. Disclaimer: not me I'm a civilian.
- zanny 11y agohttps://vdesign.kde.org/ https://vdesign.kde.org/ And the UIs for all this stuff are going to be done in QML, and if you can do HTML and CSS QML is much easier and more intuitive. It puts all the animations, gestures, etc in the QML files and they have a visual designer in QtQuick so you don't actually need to write code at all if you don't want to.
- keithpeter 11y agoThanks for pointing out that page. I'd not found much but then I was searching for 'user interface testing' &c. "The goal of the KDE VDG is to create a central point for all visual design work – a one-stop for all KDE's visual design needs, if its logo's, icons, visuals, illustrations, launch pages, plasma themes or presentation and promo-material." Above from the 'what' page on that site. Might be just my expectations but I read that Web site as being about branding and visual design rather than UI and workflows. The 'who' page mini-bios do not appear to have anyone from a UI / workflow background - or is it that am I just out of date with the job titles that cover those skills these days? https://mail.kde.org/mailman/listinfo/kde-usability https://mail.kde.org/mailman/listinfo/kde-usability There appears to be a kde-usability group, but their mailing list archive is not public, you have to subscribe. PS: Fedora 22 KDE spin does look really nice in my opinion.
- DanitaBaires 11y agoDesign cannot really be done by comitee.
- jbk 11y ago> You must have some very low standards for "looking great". For an early project, no, it's quite nice. They did not change a thing from QML apps, and it works fine already, without any design changes. > The design is all over the place. The design is clearly not all over the place, and is way better than any early Android or Ubuntu version. It's an early prototype.
- legulere 11y agoUser experience should be the driving factor when prototyping. This looks more like a me-too copying of features from competitors.
- sho_hn 11y agoFWIW, this is very early and not representative of where we want the UI to be. A lot of our effort so far has gone into e.g. getting our Wayland story sorted out (we knew we wanted to be Wayland from day 1 on the phone, and do it properly to where every line of code we write also gets Plasma Desktop closer to being on Wayland) and getting all our service infrastructure ducks in a row on a phone device. Our community designers have in the meantime poured a lot of effort into generating user personas, design guidelines and mockups, but we're only just beginning to reallocate developer resources to realization. This is open source -- you get to see how the sausage is made.
- afro88 11y agoFair enough. I know my post was quite negative, but the first thing people see is that first page. It's definitely worth making that initial image look like something you can and would want to use. Text so small as to be unreadable, on a huge screen, was just a big red flag for me.
- sho_hn 11y agoI understand, and I don't mind it at all. We should all get upset at things that need improvement to drive that improvement.
- Qwertious 11y agoI'd actually disagree - this is a problem that KDE has in general. KDE has a focus on features and customisation but chronically lacks in the UI/UX department, whereas GNOME has a focus on UI/UX but is obsessed with one-size-fits-all and is actually going backwards with options, and ends up somewhat braindead as a result if you don't like exactly what the GNOME team likes. Seriously, KDE isn't new. Desktop has really shitty defaults, which is a major problem considering how many people don't know how to change the settings. I'm talking about "not a computer person" people, here.
- sho_hn 11y ago> I'd actually disagree - this is a problem that KDE has in general. Are you actually disagreeing, though? We both seem to agree it needs work. And you're correct the challenge is partly an institutional one -- recruiting and retaining design talent and manpower, and fashioning new ways of working together. It's something we're aware of and working on, and if you'd like to work on it with us we'll welcome you with open arms. The KDE community isn't unaware of its weak spots and isn't satsified with its weak spots; addressing weak spots is Real Work[tm] though, and doesn't always happen over night. We'd like to think we've made some progress (for example we've identified some things that don't work from past attempts, and that's useful institutional knowledge as well), though. And that it's very much worth doing. tl;dr If open source is bad at product design, we'd rather not accept that as law of nature and work out how to get better. If that's a challenge you have an appetite for ...