Today at 17:30 there is a panel presentation here at Berlin Desktop Summit that is unfortunately titled "KDE Platform 4 Roadmap" and the schedule says I'm presenting it. This was submitted prior to the Platform 11 meeting in Randa so it could make the speaking schedule here at the Desktop Summit. At the time I didn't know what precisely we'd decide on at Platform 11 .. and the title reflects that.
What I did know was that we would want to communicate the results (whatever they would be) from Platform 11. That is in fact what we will be doing. Better yet, I will be joined by David Faure, Kevin Ottens and Stephen Kelly in doing so.
Interestingly, however, the presentation will not be about KDE Platform 4. It will be about KDE Frameworks 5.0.
Yes, you read that right. Coming out of Platform 11, we have a roadmap for the next major releast of KDE's libraries and runtime requirements. The emphasis is on modularity, dependency clarity / simplification and increasing quality to the next level. Our goal is to give us better tools for desktop app development, give our KDE mobile projects a leg up and make KDE's libraries something that Qt developers can and will use.
There are many steps to get there: reexamining what is in KDE's libraries that ought to be in Qt proper; dividing up the libraries along the lines of the new organizational charts we've drawn up at Platform 11 and subsequently presented on kde-core-devel, etc.
We do not wish to introduce anything highly disruptive, however. As with Qt5, we want this to be a mostly-under-the-hood set of work. We will be taking this opportunity to adopt some new technologies behind the scenes to increase interoperability, such as introducing a Secret Service implementation that can phase out KWallet. (Yes, we have automated migration code ...)
Application development will not be pausing as we do this: releases every six months of application improvements will continue based on the 4.x codebase. When Frameworks gets to the point where it is ready for serious banging on, then we will start repurposing our highlight applications to the new codebase. We don't want application development to be held up by the library development, and we don't want the library development to create much, if any, need for "porting" application code. We want "just recompile and test" to be the common case, with whatever changes do become necessary to be of the simple and even automatable sort.
If this sounds rather different from how we approached 4.0, that's because it is. The requirements, needs and context for this release are utterly different. We're after evolutionary improvement and broadening our developer ecosystem, and our plans therefore need to, and in our opinion do, reflect that.
We will be communicating these developments over the next months in more official and comprehensive means than this personal blog entry written while I'm sitting in presentations at the Desktop Summit. ;) However, I wanted to make sure people knew what was coming in our presentation and hopefully motivate people to therefore show up and participate!
So one more time: Today at 17:30 in our panel discussion in Kinosaal room at the Desktop Summit we will be discussing the plans for Frameworks 5.0 in detail, taking questions and entertaining the thoughts shared by those who come. Be one of those people! :)
Subscribe to:
Post Comments (Atom)
13 comments:
"reexamining what is in KDE's libraries that ought to be in Qt proper"
Great - it's not uncommon to see developers agonising on whether to go for a pure Qt (fairly easily deployable) vs KDE (much richer than Qt; harder to deploy) solution, and it would be nice to see the pain of this reduced. The GNOME/GTK guys seem to have made some very effective strides in this direction.
Any chance of a "sneak preview" as to the functionality/ frameworks that are likely to be migrated? I've heard that QAction vs KAction listed as one of the "pain points", and presumably KIO is a major factor in whether to go for pure Qt vs KDE; do any projects (I'm thinking Amarok and Calligra in particular) have a list of things that they would love to see moved upstream into Qt?
Cheers :)
Those are great news Aaron. You guys do a great job working with the underlying technologies of KDE. My suggestion would be to also discuss how to simplify KDE's option (not take away from them) but how to make powerful UI changes with simpler methods and less clicks. I will probably be posting things like these on my blog anyway. Keep it up, KDE is awesome.
I think that today, it is possible to say that we need to rethink plasma as a whole. While full of promisses, this technology does not lived up to it, and is not doing anything much better than KDE3 did.
Please note that this is not me saying we should trash it or the code in way way, no trolling here!
I am saying we need to rethink the plasma goals (what we do want to provide?), and HOW plasma is used.
I am an android user, and I find incredible that android UI is simpler/better and faster in my Galaxy 3 (666Mhz CPU, horrible GPU) than plasma on my desktop (core 2 duo 2.3 Ghz). How can we use the KDE5 transition to improve/fix this problem? How can we advance plasma to the next level?
User/UI wise I know some answers, the about the technology, no one better than Aaron to tell us.
Thanks! And sorry if this sounds wrong, I want not to critiquize, but to incentive us all to work together in order to find what is wrong or can be done better :)
Can't wait for Wayland. Thanks for your hard work. :)
truthfully, it took me a long time to move from kde3.x to kde4.x because so much functionality was simply completely missing from the newer version. overtime, you have addressed that issue... though I still find myself missing kde3.x rather often. it was simply a better experience imho. I do not care about all of the flashy effects kde4 can offer, I really do not care about widgets, and I hate that all of the icons on my desktop are treated like widgets even when in Folder View. KDE5 sounds like a continuation of KDE4... so no radical changes? Please god, no more. GNOME was ruined, XFCE was bloated, Unity is a terror. LXDE has never worked properly... please, God, leave us one DE.
For me KDE5 needs to take major lessons from the mistakes of Gnome 3 and the previous KDE.
Firstly offer full UI customization right off the bat as soon as KDE5 hits the market, no silly third party packages, no stupid moronic uncustomizable UI's.
We need a DE that is flexible, user friendly and not something that is a chore to use.
In short dont make it like Gnome shell.
Off topic but..... :o)
Activities: can each activity be made to launch separate instances of apps like kontact/kmail with their own unique configurations? if not, are there any plans to do so?
I also like simpler with lesser clicks. With the later KDE's,my favorite Karamak theme is gone.I'd like it back with round min/max close buttons..And the 3d buttons...Gone as well.These are examples of the kinds of eye candy I like.
@SSJ: "it's not uncommon to see developers agonising on whether to go for a pure Qt"
we've noticed that too, and are aiming to fix the situation.
"Any chance of a "sneak preview" as to the functionality/ frameworks that are likely to be migrated?"
it's all been covered on kde-core-devel and if you do a search for "modularize" on community.kde.org you can find various things, such as links to spreadsheets and pdf diagrams that are online.
we're still going through the definition processes and will update as we go.
KAction and KIO are indeed two of the (many) issues we've addressed and are building plans for.
@Andy: "how to make powerful UI changes with simpler methods and less clicks."
we managed to do this from the 3.x to 4.x releases. just compare dolphin and gwenview, fo rinstance.
"I will probably be posting things like these on my blog anyway. Keep it up, KDE is awesome."
we'll try :) thanks for your support.
I am not a coder I am a user. In the last few years I think the priorities of the Linux world have been going on a strange path. I like the glassy look of KDE4 but I wish I still had KDE3.5 extra functions. I can't stand Amarok 2.0 compared to 1.4. Thankfully Clementine is resurrecting the superior Amarok 1.4. The latest thing with Gnome 3 and unity is just horrible. It makes no sense as a GUI. Its not intuitive. I use both KDE and Gnome regularly. I will be shifting to KDE more due to mess of Unity and Gnome 3. But if KDE5 is anything like those 2, then I don't know what I would do. Can you all please just resurrect 3.5, make it look like KDE4 and call it KDE5. This whole craze of making things simple is being done in the wrong way. Sure simple is good, but functional is better.
Thanks
a normal user...
Please keep your desktop UI as it is. GNOME 3 AND GNOME SHELL are atrocious and not many people like them. Please don't succumb to silly fads. we need a professional UI.
Great news, as a long time user:
And my hope -& advices, if I may- goes along with those of Andy, Iuri Fiedoruk, Diego Viola, Ford, dancingmadrb3, nyn & RemixedCat.
Make it simpler by making it modular.
No dependency hell please. Make it better by making it functional and flexible. Configuration may be general to the entire DE; but also need to be by app to provide full independency.
If apps are not so depending on the entire DE we can make of KDE whatever we wanted, fit it to whatever hardware and whatever needs, and that is what will make KDE the DE of choice.
Post a Comment