We have something like that at VISL, but with zero statistical or machine learning or AI aspects.

We instead write a few thousand rules by hand (largest language has 10000 rules) that look at the context - where context is the entire sentence, and possibly previous or next sentences - to figure out what meaning of a word is being used and what it attaches to.

E.g.
Input: "They're looking at writing an AI which can in some sense understand what is being said."
Output: http://dl.dropbox.com/u/62647212/visl-eng.txt , http://dl.dropbox.com/u/62647212/visl-eng.png

This kind of system takes longer to develop and refine, but it also doesn't have any of the statistical problems. 95-99% "understanding" of text? Sure, we can do that. Statistics top out long before, and then have to add in rules to get the last 5-10%. And where statistics require giga- or terabytes of text, rule based systems only require a single example of a valid grammatical construct or word usage.

Why yes, yes it is: http://google.com/search?q=cry+it+out+brain+damage

jEdit's only flaw is that it is Java. Asides from that, it is a solid Unicode capable cross-platform editor that can work with files over SSH. Synchronizing your sessions, configuration, and plugins is as simple as copying over your .jedit folder, even between Windows and Linux.

I so far haven't found a single other editor with all those features. Do tell me if one exists...

Sure it does: View -> Compact View

True, that sucks.

It does? I've can't see any ads in Skype anywhere, but then I don't use the Home screen...you can turn that thing off.
I did enable it just now to see if there are ads, and there are not. It looks like a rather plain contact's status updates listing.

It does? Again, you can turn that off, or just ignore it.

But really, Compact View fixes most ills. It makes Skype look like it always has.

Then maybe what you want is DeepDict. E.g., magic is used like http://gramtrans.com/deepdict/lookup.php?word=magic&class=N&lang=eng&top=200 - it is not free, though all words starting with 's' are currently open to viewing for anyone.

It yields info such as: black magic, Orlando magic, ceremonial magic ... magic kingdom, magic roundabout, magic flute ... practice magic, radiate magic ... magic of animation ... etc

(disclaimer: I work on the DeepDict project)

At my job (plug: http://gramtrans.com/ ), we're doing human level machine translation now, if by human you're talking about high school level of quality.

But in difference to Google or Microsoft or all the previous attempts, we're NOT doing Statistical Machine Translation; we make Rule-Based MT.

Google's way is to scan huge amounts of parallel texts for good translation candidates, which works to some degree, but it has an upper limit that no amount of parallel text will help them surpass.

Our way involves writing a huge amount of rules so that our software actually understands the input text, and then rules for how things should be translated. This takes a lot longer to develop as you might imagine, but the results are great and there is no limit to how good it gets...if something isn't quite right, add a rule for it.

However, because it takes so long to develop and that we're only 2 people working on it, we only have a few really good translation chains.

Agreed. I also have Reply All as default, for the same reasons.

It's just easier, plus it makes you more aware in the rare cases when you don't want to Reply All.

Or, you could use POST requests like you're supposed to. Any number of intermediary caches and load balancing tricks can play foul with GET requests, but POST is not allowed to be cached.

With jQuery this is as simple as using $.post() instead of $.get().

Seriously? You're paranoid about letting go of your card for the 3 seconds it takes to enter the PIN? The card remains right in front of you, no more than 4cm away from your hands...

Where do you live where stealing cards at the payment terminal in full public view is so frequent that you feel a need to be paranoid about it? I've never even heard of such a case of theft/assault.

No, the real problem with the chip system is that when you put the card in the holder, the security code is facing away from you, visible to the store clerk...

Denmark already has a similar thing. We can perform most actions dealing with the government online, and we even get a gratis certificate for digital signing and encryption of emails. I haven't had to go to a government or city office in years.

Tabs on top is horrible...one of the key reasons I dislike Google Chrome.

I have looked around. I've tried them all, pretty much, on both Windows and Linux. Only one I haven't tried is XCode on OS X, but I've heard good things about that. I've dearly wanted to find a single powerful IDE that I could use cross-platform, but none exists.

Of all the IDEs I've tried on various platforms, none come even close to VC++ Express. A huge amount of that rests in the debugger alone...gdb and IDEs that use gdb are hilariously underpowered compared to VC++'s integrated debugger.

So for all development and debugging, I use VC++. For profiling and sanity checking, I use gprof/gcov/valgrind. Best tools for their job.

I'd even say Visual Studio is the best, by a very long mile. There just isn't anything as powerful in a single cohesive package. It is one of the main reasons I develop primarily on Windows (the other being gaming). And don't let anyone fool you into thinking that VC++ is evil...it doesn't lock you into making Windows only programs; you have to lock yourself in, if that's your desire.

You could get similar results with cobbling together various tools, but why bother when VC++ Express already Just Works...

That said, the final pin to make VC++ Express perfect would be some Valgrind functionality.