I was disturbed to see my office there because I don’t think I will be able to visit often enough to drink all that wine before it begins to leak, since it should be horizontal for long-term storage.
At any event, I thought I would try to tell you a little bit about the context of blockchain security.
All the work we are doing in this area grows from all the internet, and the internet has 3 properties, all which played their role.
The first one, the one that gets the most attention from the most people is the internet handles information by dividing it into packets.
The purpose of that first property is to support the second property.
The internet does not have to spend resources on doing anything with data that aren't travailing at that moment.
In the old circuit switch telephone system, when you picked up the phone in Shanghai and called Beijing, it created a wire from one city to the other like closing switches and that is very expensive, even if you have a very short call.
The third property of internet is probably the most fundamental, the most important.
The internet recognized that the correct basis for all communication is unreliable communication.
The internet protocol sends packets and pays no attention whether they actually get there or not but the transmission control protocol which operates above the internet
protocol, re-transmits packets unless it gets acknowledgment that they have been received.
Internet is short for “internetwork” , a network of networks but it is fast becoming the one and only network.
The jewel of the internet is the World Wide Web and the important thing is that world wide web could not have been built on top of the circuit switched system because when you connect to Baidu and search for something and pick some of the results and click on them, each thing hands you off to another thing and if something had to make a telephone call for each of those hand-offs, it would broken down as that is incredibly expensive.
So we’re about to expand the internet dramatically to become the internet of things.
So the numbers will increase for where we used to find, we would have at most 1 node per person, we now expect to have thousands of nodes per person.
And each connection, each node will have to drop substantially in size and in cost.
So the internet has 3 major themes.
One is openness that everyone can connect to the internet.
The second is decentralization, nothing that we know how to do in a streaming fashion should be done centrally.
And the third is the support of social processes whatever they may be.
So in these context, we can ask, what does “ secure internet” mean?
Because the internet is open to everyone, it is in every country, hundreds of thousands of companies, billions of people but vast diversity of objectives.
So I think what this means is we have a network rich enough that we can build secure networks, for example, for banking within the internet.
A major, new direction in internet security, in internet activity, in all configurable activity is blockchain.
I think we’re going to get more about that subject today.
Blockchain have the origins in 2 things.One of them 25 years old , another 40 years old
The first is merkle trees which are in effect are trees of signatures.
And the second is application of merkle trees that is practiced beginning in the 1990s
by two mathematicians named Lenstra and Haeber.
So the Lenstra-Haeber times tamping system created tree signatures for the time stamps of documents on the web and every week, they would publish the root of this merkle tree in the New York Times and what blockchain is doing is replacing this publication on paper that the constant negotiation among the members of the blockchain, which periodically achieves consensus as what the root should be.
And in the inside, to get people to participate in this energetically, it’s important to reward contribution to achieving consensus.
So to operate these, we need many internet security technologies and we depend on the two natural forms, one of which is static measure analogous to doors and walls.
And the other is dynamic measure analogous to guards and polices.
So there are a number of measures of internet measuring at techniques used in internet security and the first two are these algorithmic guard to walls and doors secured
completely and cryptography.
And the latter three are the dynamic technique of recognizing attacks, analyzing and
collecting evidence of attacks, and perhaps retaliating against the attackers to deter them from doing, can control behavior of much more freely occurring of components by large applications.
Unfortunately, it isn’t working very well.
It cannot make the operating systems as trustworthy as we need them to be and we cannot make the operating systems capable of controlling the applications in all cases.
So, in brief because I like the subject, the cryptography consists the measures for protecting confidentiality, integrity and authenticity.
And this is the best cooked part of internet security.
Transport layer security which your browser engages in with websites.
This is the most widely deployed cartographic security system ever built.
On the military, while they are the main customers for cryptography.
It was a lot of things.
They bought millions of something that was a lot of things.
Now, they’re opposed to a billion browsers in China alone and all of them have high grade cryptography built into them.
There are national standard, by in large public, speaks of symmetric algorithms called Kuznyechi of a recent Russia Federations.
These are companies by what I call a symmetric or public key cryptographic
These are used in signatures and managing keys.
Now, at present, the physicists have been promising us quantum computing and that may be absolutely wonderful,they may help solve problems of protein folding in biochemistry, but it is a threat to the current systems of a symmetric cryptography.
This has been rather slow in arriving but it is a recipient of a tremendous amount of investment, much of them are Chinese but also in the US, large corporations such as IBM and Google and Microsoft.
So there are projects underway to develop new, public key algorithms that will not be vulnerable to quantum computing.
They use parts of mathematics called lattice theory and coding theory and they are also being slow to arrive.So we’re in, by looking at three big problems in security and
in the security of the environment in which we need to operate blockchain.
The first we are about to discuss, we need cartographic algorithms that are not vulnerable, in the available to particularly quantum computing.
The second I said with the state of secure computing by in large is logic, we need to revisit that problem with new tools, perhaps with artificial intelligence will be smart enough to help us develop really secure operating systems.
Finally, more broadly than blockchain, more broadly than cryptography, more broadly than security, we need to make a qualitative improvement in programs.
We, the rate and number of progress that we accept in programs is totally unacceptable.
I believe in that to fully solve this problem, we will take new discoveries but I think good start could be made by just a change in the standards and the education of programs.
So those are some of my thoughts on where are headed.
I look forward to hearing the thoughts of many other people.