It can be done, the system needs an overhaul, as companies route data in an insecure or unencrypted manner then those companies are at least in a position to snoop. If these companies become excessively profit driven or are obliged to (Plc) then that data is a profit center and no longer private.
Encryption is no where near enough though! If anyone knows where our data is they can corrupt or steal it, or force you to give up passwords to it. It needs much more than pgp email or similar as these are layers on top of a currently brocken system. SMTP etc. will require servers and these can be snooped on and until we move away from servers and allowing others access to our data then it's going to continue as is with loss of privacy and ultimately liberty.
Huge disclaimer I work for this project novinet it's open source (dual license) and aims to provide people with a network that ensures privacy and security in a manner that's invisible to people and this is key. It's very new and like all new ideas will have detractors, but when people dig into the detail it becmoes clear that this or something very like it is required if we want privacy ever again.
If hackers focus on this issue with the above project or other ideas to achieve the same end goals then it will be achieved. My contention, however, is that we need another way of putting our data on this Internet of ours and we need to do so in a manner that allows more options than today with much better user experiences. From experience though this is not a simple job and does require a lot of new thinking and more importantly it requires to be available to everyone, not a % of the code and ideas but 100% available, however that's achieved.
This may not seem to bad until you consider that perhaps you wanted to actually build a business for the good of everyone, say to give everyone free Internet or something (anythings possible) and large ISP's etc. simply prevented it by altering their model slightly to keep them in jam and keep mad fekers like you from destroying their stranglehold on the customers.
So I like your idea it's closer to workable than now but perhaps this is something that requires consideration, it was one of the reasons for patents in the first place and it's abused to death now (I admit).
Perhaps there could be rules as to the size of the entity doing patents, i.e. anything above SME cannot prevent straight use and pay (small feasible) royalty but anything SME or smaller can. Just another idea.
Again though consider the guy who creates a pill to cure cancer and buggers about trying to create a company to exploit it, delaying the intro of the new tech. Not easy!
Theres many more explanations to sub atomic particle mass than the Higgs Boson, some have been though of and theorized, of that I am sure and likely the correct theory has likely not been discovered yet, but it may well be soon, who knows? Apart from that the discovery of new particles will be a further amazement to quantum physicists, even more conformation of quarks and the color charge would make a lot of people very happy, me for one or two or both!
I am always amazed by folks who profess to say a callback can appear multi threaded, I am not sure if this is what you mean but it would appear to me to be not possible, although I am genuinely interested.
I imagine Google could easily Open Source most of it's apps though as there only access to the Google back end therefor not giving away any bigtable, map reduce, pagerank secrets etc.
I am interested in how it can be made malware and virus free unless it's a read only OS and therefor either storing all your cache, data either in RAM and loosing it, or on Google servers or similar ? That could be a privacy nighmare for them but I would assume Google are way smarter than that and can employ some cool technologies in this privacy area to. Meanwhile has anyone an idea how they can be malware, virus free ? (you can get memory resident viruses per session even on a read only OS and
This in my mind is true cloud computing and not going back to central servers or even DNS which itself is reliant on centralisations and human authoritative controls. True cloud and true p2p must be brought together. We have released a library which allows for a NAT traversing (UDP hole punching) and extensible kademlia networking (DHT). This was due to the fact that although many libraries state they can achieve this in c, c++, java or python etc. we have never actually found a good base to create our main application from.
So it seems that open source has been a little slow in providing solid networking libraries to allow open source cloud computing which can easily challenge the giants head on and they should. Our library is at google code for anyone who is interested (BSD license).
I think that with a solid base then single developers, small groups a well as any other type of innovators can easily create an application to make proper use of p2p and therefor cloud computing, because simply using the Internet as a big cable to some servers which may or may not be centralised is not cloud computing, that to me is merely borrowing somebody else's computer when required and paying them for it. If it were a true cloud then this would all be happening in real time and not known to the users of the system. In my opinion real cloud computing is a mindshift along with a shift in how computers are used at a more fundemental level than connected VM's over long pieces fo virtual cable.
I am sure there will be differing opinions though
panic: kernel trap (ignored)