I do not like XML, or C#, or GConf too much, but I think they are doing their job good, and I do not care too much. Gconf is actually pretty simple to use, and very clear. I do not like XML, but for most configuration files it is perfectly good solution: small, good strcture, self descripting, familiar syntax to everyone, easy to handle in all languages (as for configuration). As for datastoring I do not see it as best solution. C# is AFAIK pretty good technology, and Mono is now very good implementation. I not use it mainly becasue I just do not want to waste next 100MB of memory to just single application. But people with better machines will not care.

Only think that i think was big mistake is a CORBA, and as we know GNOME is getting out of this rubish technology. I guess they used Corba, becasue it was only reasonable open-standard technology available for multiple languages at the time GNOME was create. I think they should just invent something simple, but I guess it will still took some work to do, and time. Thanks it now changes.

Yes there is not desktop in "gnome SHELL" (yet, and will probably not be). And anyway you have this "show desktop / minimize all" buttons, geastures in compiz, etc.

You do not need to use gnome shell, just use normal gnome desktop (gnome-panel + nautilus + metacity) with real desktop. And If you want you can easly readd minimize/maximize buttons if you wish. They are only chaning default for gnome shell.

And how is this releated to the story? I do not see, as gnome is disconected from its window manager, namely metacity, and (for gnome shell), mutter. And as many people already stated you can still maximize, minimize stuff even without buttons (in about 6 different ways). And If you want you can actually easly add them back. And actually they are not even removed if you do not use gnome shell...

GNOME is not losing anything.

This is only if you have Gnome Shell enabled, and it is actually only epxeriment. It will be probably more reverted back, as currently minimize button isn't really minimising, as there is no task bar in gnome shell. It is more a hide button. Also Maximize button is redundant. As you can easly use titlebar double-click. Also with minimize/maximize buttons missing, you can still access it's functionality in about 5 ways. Shortcut, Right-Click on title bar, click application icon/menu, Alt-RightClick. And you can also enable minimize/maximize buttons easly in gconf editor.

I disabled minimize and maximize button about 2 weeks ago, and noticed this fact only once during my normal work. They are useless to me. :) And i work in normal gnome desktop, not gnome shell. So I think experiment is good to be done on more people. :)

Uncertainities in constants (G, h, c, etc), are included in calculations. They do not explain why experimental values are different. Heim Theory is beautifull, but just wrong. Or there is some error. For example the measurments was actually done based on some incorect theories (for example you do not measure, for example mass directly, you measure energy it given to calorimeter, or how it behaves in magnetic field, and froom this infering mass BASED on other (well established) theory).

This is big problem in interpretation and desigining experiments.

Well it is quite straightforward calculations to optimize path, and could be easly done even 50 years ago. There is probably some reasons nobody done and used this results, it already.

I have read 20 more Tab-xxx.pdf files. Claiming that they are copy&paste is again nonsense.

They mostly consists of straight forward code, #defines or function definitions which looks similar.

Of course they are similar becuase this are POSIX functions and defines!

Their claims are just ridiculus.

What a nonsense.

ELF which is a file format, and is pretty stright forward to implement, and also have some functions and names directly embeded in the standard specification. It is not any proof.

FUD again.

Every sufficiently advanced cryptography technique is indistingushible from magic.

rfc1321
"The algorithm takes as input a message of arbitrary length and produces as output a 128-bit "fingerprint" or "message digest" of the input."

So it basically means that there is no limit.

(newer hashes, for safety or becuase they use ctr-like modes, define maximal lengths, for example SHA-1 and SHA-256 have limit of 2^64-1 bits. SHA-384 and SHA-512 have limit of 2^128-1. Still 2^64-1 bits is bilions of gigabytes.)

How long we have been waiting for this? 10, 20 years? It looks that designing, simulating and building was harder than it was initially projected.

And this is still prototype.

There is many script which scans webpage for tinyurl type links and performs reverse operation and make them show as original url. One of such scripts is URL Elongator Plus for Opera http://extendopera.org/userjs/content/url-elongator-plus (it is UserJS script). Works fast, is configurable and supports lots of pages.

There is also other similar scripts.

Eventually it would take maybe decade or two, but it will be discovered. And Einstein have big help of other physicists, and mathematics, notably Riman, Minkowski, etc.

It is stupid. Some things are better then othr in different things.

Currently people are good in generlizations, abstract thinking, spatial coordination, pattern recognition and finding. Computer are better at borring stuff which repeats the same many times, like searching for something on big list, or calculating trylions of formulas by repeativly applying rules.

Things are slowly changing and more and more stuff which was reserved for people can be done by computers, but this doesn't underestimate us in any way. We created this machines, we programmed them, now we have more time for other usefull things, we move humanity to the new future. I really don't see a problem.

I personally always very happy when my own programs can outsmart my or make me supprise :)

Chees or music isn't very hard thing. Actually Deep Blue or this robot achived they results slightly in different manner than we, and I would still say that they failed. Deep Blue was using brute force, not asbtract and general strategy planing, he was not very good at assosciating common patterns and predicting what will happen without simulation. This is what is called inteligence, to predict what will happen without actually simulate or carry experiment. But Deep Blue was performing full scale brute force simulation, so it isn't inteligent. It is soft-AI, Deep Blue, doesn't know about abstract properties of chees.

The same is with music. Maybe this program have rules, maybe it can compose greate music. But this music is created for people, and it isn't human, so he is actually creating music which he can't hear. This is ironic. But he doesn't really feel or know by itself how to create this music. Is it very innovative? Could he create new style in music? Or create new tonic system, could it have good imagination to create new music instrument? Good in the sense that it will be nice to hear such instrument be humans? I don't think so.

It is still soft-AI. All this is fake.

I don't see any problem. Whay is special about Mozart or Bethoven? it is simple, they was first to compose this masterpice of music.

What is a problem, is who will have copyright right to this AI works?