I've been using KDE since the 1.0 beta was released with the SuSE 5.3 distro in September of 1998.
I've been using KMail since I left Thunderbird about 10 years ago.
I am currently running the latest version on Kubuntu 14.04 fully updated, with filters, encryption, etc...
It has been and is running fine.

Remember when KDE4 was released?

The developers opened it up to any and all suggestions and because of the power and rapid ease of development using the Qt API they went through a whole series of experiments interfaces and appliations. One volunteer, who was in grad school at the time, offered a web page to explain the new apps and features. He was crucified by those who abhorred change. Their attacks got personal. Some of the attacks were drive-by shootings by people masquerading as KDE users. He quit in disgust and devoted that time he used to his wife and graduate studies.

I suspect that the same thing will happen with this venture. My recommendation is to continue to polish the KDE UI and remove conflicting dialogs, fix the things that don't work properly, or don't work. Like this problems mentioned in this YouTube video: http://youtu.be/N7-fZJaJUv8

Above all DO NOT hide the current power and flexibility of KDE, i.e., "dumb it down", under a plethora of "useful" or "helpful" buttons, menus or dialogs. Windows does that. So does Unity. If I wanted that kind of interface I can use one of them. We saw what happened to GNOME2 when it was dumbed down to make it "easier to use". Is it possible to make a GUI "idiot proof"? Idiots are extremely ingenious, but simple interfaces are, well, simple. As in not powerful.

KDE dev team: IF you insist on shooting yourself in the foot with this scheme would you make it so that the user, during the installation process, could select the type of interface the users wants, say a mutually exclusive check box offering either the "Experienced User GUI" or "Novice Use Interfacer"? Either that, or make it easy for distro developers to select the kind of user GUI that want to default to and make the alternate option a Muon choice.

Baloo, at least, needs to be sufficiently visible that you can turn it off. It eats up an incredible amount of CPU time to, for me, no benefit. KWallet has it's points, but it's not THAT great. In my use case sticky-notes would be quite reasonable for passwords. Just don't make them accessible over the net (i.e., to other programs running on the same machine). I'm not worried about shoulder surfers. As for Akonadi...I had no idea what it was until I just now looked at the web page, and I still don't know if it is of *ANY* value to me. But if it doesn't take up CPU time when I'm not using it, I guess it doesn't do much harm.

OTOH, I find Gnome3 unusable. Gnome2 was decent...I preferred it to KDE4, but then I preferred KDE3 to Gnome2. xfce would be a good system, but when I tried it, it got confused about which window was on top of which (more specifically, windows tended to get stuck under the menubar at the top of the screen). It's usable, but with several misfeatures, so currently I'm using KDE4. I'm also wondering about razorQT, but I don't want my window manager to be flakey, and the last I heard razorQT was in very late alpha. I've also heard about LXQT recently. Don't know what it's status is, but it isn't in the system repository, and this makes me dubious.

There were (and are) people who like MSWind. Agreed.

MSWind became dominant because the people who made the purchasing decisions trusted IBM. Not because people who used the computers liked it. Most of them didn't. Now most of them do, because they've become habituated, and the thought of putting in that much effort again terrifies them.

If you want to pick a company that became dominant because people liked it, pick Apple. I, personally, don't use or want to use Apple, but those who do use it like it. (When I used it, I liked it...but they made a change in the EULA that I found unacceptable. Now I no longer know it, though I don't actively dislike it the way I do MSWind.)

IIUC, this is not a survey of the *level* of sexual assault, but of the rate. And if the sample questions quoted above are typical, then I'm surprised that it isn't higher.

OTOH, the questions that were listed above (in the discussion about poorly worded questions) don't distinguish between a bit of uncomfortable humor and forcible rape. One presumes that actual criminal activity is rare, but this isn't evidence of that.

That said, in groups that are predominantly male and relatively isolated from external contact, one might expect that undesireably agressive sexual behavior would be relatively common. The real question to me is how moderate is the degree of undesireably agressive sexual behavior. (The rate would be interested *IF* coupled with the degree.)

I believe that there are theoretical designs for computers (using reversible computation) that can compute without using any energy in computation. What I'm not sure about is that there's anyway to retrieve the results of the computation. (I've also got no idea of the speed of the computation. It might depend on random motions for all I can remember.)

Whatever, that's merely a theoretical quibble about your point. But then your point itself was a theoretical quibble.

The real weakness of 256 bit keys is poor implementation (of something). And you can't know that everything is properly implemented.

Weeel....sort of. Depends on what you mean. I looked at a lot of (well, several) computers before the Apple ][ was released. They were all interesting, but not quite interesting enough. Then the Apple ][+ was released with a Pascal card, and I bought it. A lot of other people made about the same choice at about the same time. That was when the PC bacame notable. A few years later IBM released the IBM PC with no significant advantage over the current Apple product...but that was when it boomed.

This is sort of like Apple releasing the Apple ][. Not the ][+. OTOH, Dremel is a much bigger name now than Apple was then. Perhaps that will be a big enough kick...but my expectation is that there will turn out to be the need for much fine-tuning of the design. Then Dremel will release a greatly improved model. And then someone who's the darling of a business segment will release a different, probably incompatible, model with some useful differences, and many user drawbacks...but it will sell into businesses, and Dremel will be edged out of the market...though not completely, and they may continue to dominate among home users and certain niche segments.
But THAT will be the boom.

Makerbot, etc. is just like the S-100 computers that predated Apple.

N.B.: This is all reasoning from analogy, and therefore not to be trusted. But it's still a good guess.

You're assuming that the universe is deterministic, or at least that the past is. I don't believe that to be correct. I believe the past to be as probabilistic as the future. Granted, it's probably that every backwards simulation will end up at the big bang, but in between now and then it's an uncollapsed wave function.

OTOH, I also believe in the Everett-Graham-Wheeler interpretation of quantum mechanics...but not in quite the same way that they did, as I'm considering branching to be essentially symmetric between the past and the future, so that not only does the present lead to multiple futures, but there are multiple presents connected to multiple pasts in a probabilistically branching net in both directions. Each present has multiple pasts, and each past has multiple presents (futures?). In a connected lattice that (perhaps) teminates in one single instant in the past where all the lattice links join (called the big bang) and less probably terminates in on single instant in the future where all lattice links join (called the big crunch). The big crunch, however, doesn't seem to be extremely plausible at the moment, given current knowledge and theories. And neither join is required by the theory.

FWIW, as far as I can tell this model is consistent with everything known about physics, but I'm neither a cosmologist nor a quantum mechanic.

Now that's just not true. Brainfuck is worse than PHP. So, for that matter, is whitespace.

Yeah, but are they ever going to fix the filters they broke with one of those updates? Doesn't appear so. If there were a decent and maintained email program I'd switch to it immediately. Unfortunately the others all seem worse...so far...until they break more stuff and decline to fix it.

The question is, what language would they write good code in?

FWIW, C++ has many features that are strong improvements over C. Class encapsulation, e.g. OTOH, it's also full of things that are only worthwhile if you are really interested in run-time optimization. Or maybe they serve some other function that I don't understand. Like the STL. Most of the code in the STL would be far better implemented as libraries, even if it might not be quite as fast. I also strongly dislike their implementation of iterators. Python, Ruby, D, hell, even Java, have much better designs for their iterators. I'd include Vala and C but I'm not sure that just iterating through a loop counts. (I know that in C++ you can iterate through an array just like in C, but Strings are a different case...and so it anything else that C++ calls an iterator.) Even Objective C is a better language than C++, but it has the major problem that nearly all the documentation and development is tied to the Apple version, and I'm not interested in accepting their EULA.

If Vala weren't so tied to GTK, and if it would ever get out of beta, then I'd consider it one of the best languages around. Pity about those two problems.

N.B.: despite the way I may have phrased things a few times, I'm well aware that my opinions are not universal, and also that different use cases result in different choices. So this is just my point of view. But I seriously consider Ada more often than I seriously consider C++.

The summary indicates that the source code is still there. They've just stopped updating.

Others have indicated that some of the projects moved off to their own web pages, so if you want up-to-date software, you should search for it rather than just relying on MozillaLabs.

They broke the filters quite awhile ago, and have shown no intrest in fixing them. If I could find a decent replacement I'd use it. Unfortunately, KMail isn't any better. (It's worse, but with different problems...don't use it these days so I don't remember quite what they were.) Seamonkey doesn't seem to work well on a 64-bit system. And every test of a new version of email package means that a bunch of emails aren't searchable. It would be worth doing if I could make one switch to a good program, but I've made several switches to programs that turned out to be worse, so I switched back.

I'm not pleased...but I don't really see what decent alternatives are...pine? Sometimes I actually DO want to enable html in a particular email.

Interesting. Is the geometry of that so-called "development vehicle" representative of the production vehicle's?

Back in 1981-1983 when I was local support team leader for Space Studies Institute in Miami, FL promoting the idea of space colonies among the locals, one of the slides we showed was of this artist's conception of a Single Stage to Orbit Vertical Takeoff Vertical Landing system proposed by Boeing to loft solar power satellites into LEO. This vehicle also appeared in Gerard O'Neill's original edition of "The High Frontier" that Jeff Bezos probably read while he was becoming the valedictorian of his high school class.

Looking at Bezos's New Shepherd Vertical Takeoff Vertical Landing vehicle you might think that somewhere along the line Jeff caught a glimpse of Boeing's old design.