This is actually a pretty shit way of doing it; vehicles rarely hold a constant speed. If you build a database of locations and times for a vehicle doing the same route repeatedly then a more appropriate calculation is the median travel time from the given location.
Ok, so you seem to be more informed than I thought from your previous posts. In a market where the marketplace owner is a single central authority what you've said would hold about the privacy of bids / offers and instant settlement. But electronic traders are not working in those markets. For example, Nasdaq has distributed market making. The firms involved in trading are also making the market - this is a much closer model to a pre-electronic trading pit where the bids and offers were broadcast by shouting.
The actors in trading are the actors involved in making the market, so the signals and intentions are effectively public: or at least they are public to all of the major players in the market. One of the key behaviours that has been discussed is bots that try to identify arbitrage opportunities by leaping in to close bids before other actors, then hold the asset for microseconds before selling it in different sized parcels at different prices. They provide liquidity at the expense of widening spreads for traders. In the terms that you are described it these are middlemen, but in the case of markets like nasdaq they are also participants.
If this description does not convince you that the signals are public then I have a question for you: why do you think that firms pay so much to reduce latency to the exchange? Latency is only an issue in reaction times for deciding whether to trade or not, and the model that you've described so far with synchronous actions is non-interactive...
You seem to be content to argue from a position of ignorance. Maybe if you started with a very basic description you would learn the difference between an offer and the execution of an order. I'm sure that you would be capable of understanding if you tried.
No, they don't. They don't have ESP. They can't see what you do on a market before you actually do something on the market.
That's a bit of a silly thing to say, unless you actually believe that ESP would be necessary to see what is about to happen in a market. I don't know if you've given any real thought to this at all: but before a buyer and a seller can meet to make a trade they both need to signal that they have the intention to do so.
There was a link on the AltaVista front page to add a site that was not in the index, and its robot spidered all the links from that starting URL. Their big boast at the time was they had a bigger index than any other search engine (something like 8 billion pages). That was one of the reasons that Google used to crow about how many pages they could spider in their first few years.
Google had two big improvements over AltaVista (which at that point was the market leader by far). Relevance and speed. By that stage SEO was alive and well (although probably not called that). The big trick for fooling search engines of the time was spaff filler at the bottom of a page that contained search terms to make the page seem more relevant. When Google started page rank was mostly immune to this technique so it returned much more relevant results, and that is why it spread so quickly and overtook AltaVista. The fast loading simple front page really helped as well. Of course now SEO is mainly aimed at Google's algorithm of the week and the relevance and quality of the results has gone downhill since those early days. Of yeah, and their various AJAX / caching technologies have broken the web completely and are the reason that every machine I've seen recently has a bunch of semi-permanent connections to Google edge points.
Whoops, was working from memory and it has been a looong time. DH key exchange is indeed completely different to what I said above - as you've said it is an alternative way to set up a secret channel. Certifying trust on a public key is indeed the important issue, and solved in a different way.
I'm guessing the alternative issues on IBE are the need to trust the CA, which puts it roughly in the same level of messiness as using digital signatures on public keys anyway. Is that Matt Palmer at the National Archive?
What you have described is true when both parties hold the relevant keys, and they believe that the keys have not been compromised - this is when I can trust that I really have your public key and not one substituted by an adversary. To solve this problem of key distribution the DH key exchange algorithm is normally used, and this relies on the hardness of discrete logs. If the DH problem is weak (which now appears to be the case) then RSA would be borken in the sense that you could not exchange keys to use it.
If you do not have the knowledge to express that you are right and they are wrong, then how do you know that you are?
What you have written comes across as "but can't we all be inclusive and respect each others feelings so that I can play too". It really isn't about that - either you know your stuff so that you can explain why you are right on an issue, or you do not. Ultimately most engineers do not want to get sucked into "did everybody get included and take part" as it almost always destroys "was this the best result we could make".
Your phrasing is kind of hard to parse - I actually can't tell if you are agreeing with what I wrote, or arguing in a passive-aggressive way. This implies that I have had too many arguments with passive aggressive people recently and I need to learn to read things more neutrally again. But yes, that is what I was pointing out: tweaking the frequency in the fast sequential part is still covered by Amdahl's law, contrary to their wild hyperbole.
How dare you criticise the author - he is a physicist and he has stooped to coming and telling us computer science types how to do it properly!
There is a deeply appropriate xkcd but I cannot be bothered to find it. Decoding the garbage in the pcworld story tell us that he is going to break Amdahl's Law by dynamically partitioning the workload between a fast single threaded processor and many slower parallel processors. I would guess that my failing to make a fair comparison they can claim that the portion running under the boosted clock somehow beats the bounds predicted by Amdahl's law. Sadly it does not as the law is worded in the proportion of the code that can be executed on the parallel architecture.
It is quite possible that much of the hyperbole was added as sales pitch, which is a little unfortunate as the dynamic partitioning and the toolchain support are far more interesting anyway.
How? I've got a Mac mini plugged into a 40" TV and changing font sizes doesn't fix the size of non textual buttons, default image sizes, hit-zone size around window borders, scroll bars or any of the other UI elements that are not tied to the font size.
Are you assuming that all programmer are working on simple 1:1 transformations of data? It is impossible to encode anything with a summation term without using a gather operation. If there is a projective transformation in the algorithm (i.e. a change of representation onto different axes / number of dimensions) then it is impossible to encode efficiently without a scatter. Perhaps there are more algorithms out there that are suitable for vector architectures that you are familiar with?
The real key here is that there is no advantage to the device at all.
In the cryptographic protocol that the authors (all physicists) believe to be novel, but which every cryptographer is aware of: 1. The authors have a perfectly secure channel (separate from the one established in the protocol). 2. They exchange as much information over that channel as the device stores. 3. The later established channel can only use that number of bits.
For real excitement they xor together their OTPs. Sorry guys but this is called a pre-shared key and the crypto world is quite aware of it. Good luck with the window dressing getting you past the PC of a physics venue.
Organising your paragraphs around trolls - it is a brave strategy. I guess that you no some grammar nazi will bite on the first, and who could resist correcting the abortion of a description of a fusion bomb on a tech forum. Don't mind me, I'll just sit here to watch
Do you know anything at all about the Blue Brain project?
Serious question: if you do not then there is a video floating around from ICC'11 with Henry Markram explaining an overview of the project. Given that they are building artificial simulations of biology specifically so that they can explore how they work, build hypotheses and then experimentally validate them it is somewhat hard to see how this approach can be described as cargo-cult AI.
Slashdot Mirror
Are you referring to the recent news of an avian variety?
This is actually a pretty shit way of doing it; vehicles rarely hold a constant speed. If you build a database of locations and times for a vehicle doing the same route repeatedly then a more appropriate calculation is the median travel time from the given location.
Ok, so you seem to be more informed than I thought from your previous posts. In a market where the marketplace owner is a single central authority what you've said would hold about the privacy of bids / offers and instant settlement. But electronic traders are not working in those markets. For example, Nasdaq has distributed market making. The firms involved in trading are also making the market - this is a much closer model to a pre-electronic trading pit where the bids and offers were broadcast by shouting.
The actors in trading are the actors involved in making the market, so the signals and intentions are effectively public: or at least they are public to all of the major players in the market. One of the key behaviours that has been discussed is bots that try to identify arbitrage opportunities by leaping in to close bids before other actors, then hold the asset for microseconds before selling it in different sized parcels at different prices. They provide liquidity at the expense of widening spreads for traders. In the terms that you are described it these are middlemen, but in the case of markets like nasdaq they are also participants.
If this description does not convince you that the signals are public then I have a question for you: why do you think that firms pay so much to reduce latency to the exchange? Latency is only an issue in reaction times for deciding whether to trade or not, and the model that you've described so far with synchronous actions is non-interactive...
You seem to be content to argue from a position of ignorance. Maybe if you started with a very basic description you would learn the difference between an offer and the execution of an order. I'm sure that you would be capable of understanding if you tried.
That's a bit of a silly thing to say, unless you actually believe that ESP would be necessary to see what is about to happen in a market. I don't know if you've given any real thought to this at all: but before a buyer and a seller can meet to make a trade they both need to signal that they have the intention to do so.
Are you confusing Yahoo with AltaVista?
There was a link on the AltaVista front page to add a site that was not in the index, and its robot spidered all the links from that starting URL. Their big boast at the time was they had a bigger index than any other search engine (something like 8 billion pages). That was one of the reasons that Google used to crow about how many pages they could spider in their first few years.
Google had two big improvements over AltaVista (which at that point was the market leader by far). Relevance and speed. By that stage SEO was alive and well (although probably not called that). The big trick for fooling search engines of the time was spaff filler at the bottom of a page that contained search terms to make the page seem more relevant. When Google started page rank was mostly immune to this technique so it returned much more relevant results, and that is why it spread so quickly and overtook AltaVista. The fast loading simple front page really helped as well. Of course now SEO is mainly aimed at Google's algorithm of the week and the relevance and quality of the results has gone downhill since those early days. Of yeah, and their various AJAX / caching technologies have broken the web completely and are the reason that every machine I've seen recently has a bunch of semi-permanent connections to Google edge points.
Whoops, was working from memory and it has been a looong time. DH key exchange is indeed completely different to what I said above - as you've said it is an alternative way to set up a secret channel. Certifying trust on a public key is indeed the important issue, and solved in a different way.
I'm guessing the alternative issues on IBE are the need to trust the CA, which puts it roughly in the same level of messiness as using digital signatures on public keys anyway. Is that Matt Palmer at the National Archive?
What you have described is true when both parties hold the relevant keys, and they believe that the keys have not been compromised - this is when I can trust that I really have your public key and not one substituted by an adversary. To solve this problem of key distribution the DH key exchange algorithm is normally used, and this relies on the hardness of discrete logs. If the DH problem is weak (which now appears to be the case) then RSA would be borken in the sense that you could not exchange keys to use it.
Is he a considerate lover?
That is a movie that must be made. Shut up and take my money!
If you do not have the knowledge to express that you are right and they are wrong, then how do you know that you are?
What you have written comes across as "but can't we all be inclusive and respect each others feelings so that I can play too". It really isn't about that - either you know your stuff so that you can explain why you are right on an issue, or you do not. Ultimately most engineers do not want to get sucked into "did everybody get included and take part" as it almost always destroys "was this the best result we could make".
Your phrasing is kind of hard to parse - I actually can't tell if you are agreeing with what I wrote, or arguing in a passive-aggressive way. This implies that I have had too many arguments with passive aggressive people recently and I need to learn to read things more neutrally again. But yes, that is what I was pointing out: tweaking the frequency in the fast sequential part is still covered by Amdahl's law, contrary to their wild hyperbole.
That's a bingo.
How dare you criticise the author - he is a physicist and he has stooped to coming and telling us computer science types how to do it properly!
There is a deeply appropriate xkcd but I cannot be bothered to find it. Decoding the garbage in the pcworld story tell us that he is going to break Amdahl's Law by dynamically partitioning the workload between a fast single threaded processor and many slower parallel processors. I would guess that my failing to make a fair comparison they can claim that the portion running under the boosted clock somehow beats the bounds predicted by Amdahl's law. Sadly it does not as the law is worded in the proportion of the code that can be executed on the parallel architecture.
It is quite possible that much of the hyperbole was added as sales pitch, which is a little unfortunate as the dynamic partitioning and the toolchain support are far more interesting anyway.
How? I've got a Mac mini plugged into a 40" TV and changing font sizes doesn't fix the size of non textual buttons, default image sizes, hit-zone size around window borders, scroll bars or any of the other UI elements that are not tied to the font size.
Are you assuming that all programmer are working on simple 1:1 transformations of data? It is impossible to encode anything with a summation term without using a gather operation. If there is a projective transformation in the algorithm (i.e. a change of representation onto different axes / number of dimensions) then it is impossible to encode efficiently without a scatter. Perhaps there are more algorithms out there that are suitable for vector architectures that you are familiar with?
Dude I think he covered that: Moore Coors.
This seems a little bit more appropriate.
The real key here is that there is no advantage to the device at all.
In the cryptographic protocol that the authors (all physicists) believe to be novel, but which every cryptographer is aware of:
1. The authors have a perfectly secure channel (separate from the one established in the protocol).
2. They exchange as much information over that channel as the device stores.
3. The later established channel can only use that number of bits.
For real excitement they xor together their OTPs. Sorry guys but this is called a pre-shared key and the crypto world is quite aware of it. Good luck with the window dressing getting you past the PC of a physics venue.
Best slashvertisement. Ever.
Best editing of a summary. Ever.
Lowest point? We should be handing out awards for this shit.
Now that it has actually gone live and we can see what photos are being selected for relevance there is a longer and more complex answer:
Hell no.
Organising your paragraphs around trolls - it is a brave strategy. I guess that you no some grammar nazi will bite on the first, and who could resist correcting the abortion of a description of a fusion bomb on a tech forum. Don't mind me, I'll just sit here to watch
Do you know anything at all about the Blue Brain project?
Serious question: if you do not then there is a video floating around from ICC'11 with Henry Markram explaining an overview of the project. Given that they are building artificial simulations of biology specifically so that they can explore how they work, build hypotheses and then experimentally validate them it is somewhat hard to see how this approach can be described as cargo-cult AI.
Hopefully these guys can solve that problem.