Linux

For me, personally, it's better because it's more configurable and beautiful than ever. KDE and Xfce are winning the popularity polls these days, I read.

BUT some KDE developer(s) thought users would be better off if we weren't in charge of our own data, emails, calendar etc. They integrated a non-optional Networked Environment for Personal, Ontology-based Management of Unified Knowledge.(Nepomuk) into KDE ... "a social semantic desktop that enriches and interconnects data from different desktop applications using semantic metadata stored as RDF." Google "Nepomuk" and you'll see that the hits are mostly about how to disable it.

I still use KDE4 and love it but had to give up using the PIM and anything that Nepomuk wanted to get its evil mits on (the last time I tried to open KMail, Nepomuk had the system at a standstill for over 24 hours as it went about its mad indexing).

The point of this post is to spread the good news to KDE users. Just read today:

"Michael Larabel of Phoronix.com said the other day that "Nepomuk Doesn't Seem To Have A Future." After putting users through years of tumult, KDE developers have just about decided to phase out Nepomuk ".

This isn't really true. First of all, it isn't non-optional if you can turn it off. Secondly, Phoronix is a horrible media source. Nepomuk isn't for PIM anyway, that is Akonadi. And it isn't because they don't want you in charge of your own data, it is because data indexing is part of every modern desktop. It is not like they were shipping it out to Google or Amazon :p . It is also highly configurable. By default, at least on my install, you have to check the box in system settings to enable PIM Data Indexing under Nepomuk. And yes, performance was terrible, but they are well on their way to fixing that.

A recent post by Phoronix predicted that Nepomuk would stop being supported and be obsolete by this year. The article claimed, “It appears there isn’t much of a future left to KDE‘s Nepomuk framework that was developed at a cost of 17 million Euros… It’s going to be replaced going forward in the KDE land.”

That’s not true. First of all those 17 million Euros were not spent on KDE; those were invested in the Nepomuk project and Nepomuk KDE was just a small part of the entire project.

That work was continued by France based company Mandriva, once EU’s project was finished. So the claim that 17 millions were spent on KDE’s Nepomuk is incorrect. The second point is that this investment is not wasted as it enabled research and development of work which builts the foundation of the next version of Nepomuk.

Vishesh Handa, a KDE developer told us, “Additionally, the Nepomuk research project produced tons of academical papers, ontologies which are still being used in Tracker and Zeitgeist, and spawned many other independent projects such as Refinder.”

Another KDE developer Thomas Pfeiffer agrees and says,”….Nepomuk is still being used in academic contexts (I just heard a talk at a conference last year where people used it), for projects where it fits the bill much better than for KDE. Plus – and the article at least mentions that – Baloo still reuses some concepts from Nepomuk.”

Linky.
 
Last edited:
And I find it darkly hilarious to introduce MBA-speak like that into Linux. The product is fill-in-the-blank, but let's sell it hard. It's all about the sell and not what it's good for. Many Linux users don't find this amusing at all, and have voted with their feet. Myself, I consider GNOME3 unusable.

Apparently we've both been around Linux for decades. I'm sure that you've seen projects die just as I have. Do you think that GNOME can't die? If the GNOME3 devs think that they can act like Windows or OS X and force their will on the masses, they will learn the hard way that Linux is not monolithic in a way that allows that kind of assumption.

GNOME3 has pissed a lot of people off, and there seems to be a huge disconnect with their vision of the future. I think that this won't end well unless a change in attitude takes place.

I miss xmmp every time I install a new distro, so. I actually agree with the sentiment that if you don't like the brand vision you should beat feet. Keeps the community from becoming overly stagnated.
 
This isn't really true. First of all, it isn't non-optional if you can turn it off. Secondly, Phoronix is a horrible media source. Nepomuk isn't for PIM anyway, that is Akonadi. And it isn't because they don't want you in charge of your own data, it is because data indexing is part of every modern desktop. It is not like they were shipping it out to Google or Amazon :p . It is also highly configurable. By default, at least on my install, you have to check the box in system settings to enable PIM Data Indexing under Nepomuk. And yes, performance was terrible, but they are well on their way to fixing that.



Linky.

Thankyou.

I'm sorry I got it wrong though about Nepomuk though because I was assuming it was part of what I had come to think of as an Evil Triad (with Akonadi and Strigi). I had never succeeded in uninstalling the indexing or stopping it from installing itself. I did turn all three off but if I wanted to refer to any of 10 years worth of saved PIM data they sprang back into action (if it could be called action).

Good that the performance is well on the way to being fixed. Though it depresses me that "data indexing is (an inevitable?) part of the modern desktop." All I'd like is the choice to index my own (it was my profession, after all).
 
I don't think what the GNOME3 devs are doing is the same kind of thing. Their goal is not to perform a rebuild of the framework, which gets messy and temporarily drops features and functionality. They seem to want to reinvent Linux as something quite foreign to what Linux users currently value. Who gives a crap about their brand?

In many respects, though, the 2->3 transition is far less radical than the changes between 1 and 2. Gnome1 was a wild free-for-all with about a bajillion options for everything, and almost no "blessed apps". There were nearly a dozen choices for mail-reader, chat program, etc. And the degree of customizability and themeability was almost frightening. Then Gnome2 came out, and it was all locked down and inflexible, and there was initially no sign that they were planning to loosen things up in the long run. When I asked about focus-follows-mouse, I was told that there were no plans to support that feature. No, "we might add that in the future." Just a simple "no". And that paled in comparison to the wars about the modal file browser, where pretty much everyone hated it, but the Gnome devs refused to even consider changing it. Right up until the moment they changed it. :)

Seriously, Gnome1 fans used to mock KDE fans because KDE was so rigid and inflexible, and had no options, while KDE fans were mocking Gnome1 fans for having no standards, no consistency, mass confusion, little-or-no discoverability, and a desktop that could only appeal to the geekiest of hackers. With Gnome2, that pretty much reversed completely! Talk about alienating your core audience! :)

Of course it's true that users are fickle, but I think the GNOME3 devs have gone too far in alienating their user base, and many of these users will never go back.

I thought that with Gnome1->Gnome2. I thought I'd never go back. I was wrong then, so I'm less than convinced that it's going to be true this time, especially since the changes seem far less drastic.

I'll freely admit, I don't know the future. Gnome3 might be the final nail in Gnome's coffin. I just know that I haven't seen anything to make me think that's even vaguely plausible.
 
Thankyou.

I'm sorry I got it wrong though about Nepomuk though because I was assuming it was part of what I had come to think of as an Evil Triad (with Akonadi and Strigi). I had never succeeded in uninstalling the indexing or stopping it from installing itself. I did turn all three off but if I wanted to refer to any of 10 years worth of saved PIM data they sprang back into action (if it could be called action).

Good that the performance is well on the way to being fixed. Though it depresses me that "data indexing is (an inevitable?) part of the modern desktop." All I'd like is the choice to index my own (it was my profession, after all).

Hah, yeah Strigi was pretty bad. They got moved away from it in 4.10. While I don't know about potential dependency hell in KDE4 of uninstalling it, you can disable or enable them as a whole, or whichever parts you like in the settings.
 
Been playing with Elementary OS and, as a Mac user, it's easy to pick up since it's similar in so many ways. And it runs pretty quick as well.
 
KDE 5 is dead, long live KDE 5!

Word has gone around that KDE 5 will not be, but it usually spread without an explanation of what that meant. KDE 5 separates the desktop into Frameworks, Workspace, and Applications. Frameworks are themselves modularised and less interdependent, so for example installing one Kapp won't pull 400MB of dependences. Workspace (Plast Next) and their first party Applications are developed and released separately. Further, releases, specifically in Plasma, will be incremental in nature to give community feedback along the way.

We all know how these things are supposed to go: some teasing and then at a given date "The Big Reveal". Not so with the next version of Plasma. One of the things that I am so completely over-the-moon about with the way the Plasma Devs are handling the release for Plasma is that it will be done incrementally. What that means is that beyond the basics - the cool backend things, some slight visual changes - the first release of Plasma wont be anything drastic. The next release of Plasma will contain a little bit more in terms of change. The one after that a little bit more.
Why this slow pace? Because it's one thing knowing that your ideas and changes are sound and correct - another to see how we all react to them IRL. By changing bit by bit there is more room for community participation - for everyone to try them, see what they think, give feedback and suggestions and change or add to them if we have to - when there isn't ten thousand different things that are changed, hinging on each others existence to work.
This may not be great for marketing - everyone likes to write about a big splash - but it will be an insane improvement for the community and when the choice is between those two... well you know which side of the toast our bread is buttered on. :-)

http://wheeldesign.blogspot.com/2014/02/monday-report-3.html

Also, abstractions. I used to hate them and wonder why KDE proliferated them. But now that the Linux base is becoming more diverse (Wayland), they suddenly make perfect sense. They are a reasonable answer for Linux fragmentation.
 
In many respects, though, the 2->3 transition is far less radical than the changes between 1 and 2.

It's possible that my recollections aren't as densely populated with fact as I thought they were. I did consider the early GNOME and KDE to be horribly ugly, and I jumped on Enlightenment with both feet as soon as I saw it. When I did finally see some virtue in GNOME2, it was probably refined quite a bit. For a while I happily used Dropline GNOME. I didn't actually like or use KDE until KDE4, but that didn't last very long. The compositing was cute, but those big DEs eventually seem to just get in my way.

When I asked about focus-follows-mouse, I was told that there were no plans to support that feature. No, "we might add that in the future." Just a simple "no". And that paled in comparison to the wars about the modal file browser, where pretty much everyone hated it, but the Gnome devs refused to even consider changing it. Right up until the moment they changed it. :)

I understand your rancor perfectly. At one point I ditched GNOME because it wouldn't center the windows I opened. Little things like that can easily be a dealbreaker for me--why should I live with constant annoyances? And I also insist on focus-follows-mouse. It's the only civilized way to do it.

Anyway, I don't know what's going to happen with GNOME3. I'm just mouthing off on the internet :)
 
For me, personally, it's better because it's more configurable and beautiful than ever. KDE and Xfce are winning the popularity polls these days, I read.

Agreed. KDE looks and feels better than ever. I was one of the users who left KDE when KDE4 came out, but I came back when I'd had enough of Gnome. By that time (KDE 4.5?) most of the bugs had been ironed out.

BUT some KDE developer(s) thought users would be better off if we weren't in charge of our own data, emails, calendar etc. They integrated a non-optional Networked Environment for Personal, Ontology-based Management of Unified Knowledge.(Nepomuk) into KDE ... "a social semantic desktop that enriches and interconnects data from different desktop applications using semantic metadata stored as RDF." Google "Nepomuk" and you'll see that the hits are mostly about how to disable it.

I still use KDE4 and love it but had to give up using the PIM and anything that Nepomuk wanted to get its evil mits on (the last time I tried to open KMail, Nepomuk had the system at a standstill for over 24 hours as it went about its mad indexing).

This is curious. Nepomuk's been something I've barely noticed in all of the years I've been using KDE. In fact, I didn't even know what it did until just now. It's just one of those KDE system apps that sits in the background and does...something. The only time I noticed it at all was when it needed an update.

But, then, I do all of my mail and calendar stuff with Thunderbird or Google, anyway.
 
Personally, looks is my biggest problem with KDE. Visuals are only a concern in that they are properly rendered :p .

The potentially good news is that the design group has been working, and they will release things slowly and get community feedback. Or so they say. Not sure how things will pan out.

I am actually a bit concerned about what the design committee will have developed in house. So far they have released colours, a moodboard, and marketing speak. A decent theming framework and default theme are desperately needed, so this is an indication that they might be going in the right direction. I have reservations that committee->public input->final product may go wrong at a number of points. But we shall see.
 
Last edited:
So I just tried openSUSE. It is very slick and GUIful. The instsal process was very thorough.

However, I wanted to try it to test out Tumbleweed. The way they have repos split up can cause problems. YaST is actually quite good at handling dependency hell, but it comes up far too often. At one point there was a failed dependency for YaST itself that I had to fix manually by looking at the logs.

Compared to Arch, that is not as clean. Knock on wood, because I've heard the tales, but it has been working beautifully. Knock on wood again.
 
Isn't OpenSUSE supposed to be a little more experimental than Arch? I guess it's hard to compare a feeds-into-main-distro system like Fedora or OpenSUSE to a rolling-update system like Arch.

Since I was a Debian Developer for over a decade, I haven't really played around with a lot of other systems. (Debian at home, RHEL at work, mostly.) But I must say that as a Debian Developer, I've had a few occasions to compare notes/collaborate with people from Arch, and I've always been very favorably impressed with their competency, cooperation, and friendliness, which, IMO, is an ideal set of characteristics for people running a distro. My interactions with the team members have left Arch very high on the list of distros I'd be willing to try, so I'm pleased to hear that you liked the system.
 
My main problem with the changes in KDE was that it seemed to be "infested" with google and I couldn't be bothered figuring out how. I don't know whether it is, I just don't want to hear that something is all set up so that google widgets will work. I don't want anything to do with google, or its widgets.
 
Isn't OpenSUSE supposed to be a little more experimental than Arch? I guess it's hard to compare a feeds-into-main-distro system like Fedora or OpenSUSE to a rolling-update system like Arch.

Since I was a Debian Developer for over a decade, I haven't really played around with a lot of other systems. (Debian at home, RHEL at work, mostly.) But I must say that as a Debian Developer, I've had a few occasions to compare notes/collaborate with people from Arch, and I've always been very favorably impressed with their competency, cooperation, and friendliness, which, IMO, is an ideal set of characteristics for people running a distro. My interactions with the team members have left Arch very high on the list of distros I'd be willing to try, so I'm pleased to hear that you liked the system.

Well, I tried out openSUSE because they have a rolling release called Tumbleweed. Basically, you install your distro like normal, then you switch repos from the 13.1 or whatever to Tumbleweed and Current.

There are problems right off the bat. They have by default Apper as a system tray updater, but if you let it run while you are on Tumbleweed it breaks things. So you have to uninstall it. But the user isn't given any indication of this. Then a YaST update from Tumbleweed required a lib that only exists in the 13.1 repo. First of all it didn't warn me in the package manager. I had to deal with the program breaking and then looking through the logs. Secondly, you are supposed to disable those repos when you go Tumbleweed, so it was more difficult to find then it should have been.

It felt like a good distro for a normal distro, but their "also rolling" implementation is just too messy for me to run. Also, their OBS feels more akin to PPA than AUR. AUR is still the best.
 
My main problem with the changes in KDE was that it seemed to be "infested" with google and I couldn't be bothered figuring out how. I don't know whether it is, I just don't want to hear that something is all set up so that google widgets will work. I don't want anything to do with google, or its widgets.

Could you clarify? I'm not finding what you are referring to. I see a few things like syncing contacts/mail/calendar from Google, and then some discussion about Drive.
 
Ugh, just found out one of the KDE design heads is a fan of full screen start menus. Someone tell me the reason for these, other than mobile platform carryover where everything if full screen. Vast grids of empty space and big icons just don't make sense in and of themselves. This isn't a "modern" tend that needs to be followed.

KDE ships with three launcher choices right now anyway, so is not the end of the world. But defaults are hugely important.
 

Back
Top Bottom