HN user
Fast international money transfer is something that is already handled very well for those that truly need it. I know for a fact that the european central bank handles large volume transactions in a matter of seconds if not less. There is really no reason why traditional banks could not offer significantly shorter processing times for the majority of their services, it is just not in their interest to do so.
They actually have to at least in some countries, you can't just only accept a made up currency and not legal tender.
In the case of fiat money, it does have value because you have to pay taxes with it and anyone has to accept it for settling debts.
I remember when I first switched to kde 4.0 and everything slowed to a crawl because some background process (nepomuk) thought it was a good idea to index the whole filesystem.
In my eyes it is unfortunate that the terminal standard we are stuck with is dictated by a device designed in the 70s (vt100). There have been more advanced terminals, that had support for graphics and so on.
Just like Private Browsing Modes in Chrome/Firefox tell you that they won't protect from agents.
"Not even wrong" is something W. Pauli said first, although Feynman might have quoted him.
I think people solve universal login without any service fairly successfully. They use the same email as account name and the same password on every service. There is really no need to have some external provider that manages "universal login" however convenient this might be for advertising and marketing purposes.
Does stock based compensation of 640 million mean that they are buying 640 million dollars worth of their own stock each year?
In principle you can add full colors, image capabilities and direct manipulation of objects to a shell, see for example IPython (the Qt-shell) or Mathematica. Ultimately any large enough GUI application will add scripting support, take autocad for example. There are arguably tons of simple applications that are simpler to use if implemented as command line tools, for example video/audio conversion, archival tools
Events in the sense of: Here are all the queer reggae parties in Birmingham tonight/the next three months displayed on a map. If you decide to attend, get a semianonymous link you can send to all your friends by whatever method you chose, they can decide to attend simply by adding themselves to the list. Basically like doodle.com but with public location sensitive event listing. No signup or authentification required. There is no good reason to tighly couple all this to a online friendlist, you can just as easily manage your friends with an offline contact list that is synced to all your devices and use email/im to communicate.
People not using Windows, wont see this font to begin with, unless they chose to install Comic Sans for some reason.
There are efforts by Microsoft Research in this direction: http://research.microsoft.com/en-us/projects/slam/ They have developed several domain specific languages and specification tools for developing drivers. I think similar methods are applicable to protocols.
Of course anyone who has not installed Microsoft Fonts won't see this at all.
I think a startup focussing on only events and doing them well could have some potential.
England does also not have the right weather. It is much nicer to sit outside in the evening and enjoy a whine, if you are not freezing.
Lots of european cities have pedestrian zones, usually in the historical center of the city, there also aren't as many malls. The traditional medieval european city layout has market, town and church squares at their core and those are still places of public gathering today.
"Deutschland, Deutschland über alles, über alles in der Welt..." is the beginning of the first verse of the national anthem that has now been banned, although mostly because it refers to borders that Germany now does not have anymore.
One symmetric cryptography method are one-time pads. Essentially you generate a truly random key (for example by observing some decay process) that has the same length as the maximum length of a message you might want to send and distribute the key to the receiver in advance. You encrypt the message by modular vector addition and decrypt it by subtracting the key. If you use the key only once and it is random, there can't be any attack on this method.
What I don't understand why most crypto libraries are then implemented in an inherently unsafe language. I feel something like the secure subset of Ada would have been much more apropriate.
You only share the data model between the server and the client.
That is not true at all, the first five minutes will decide most of the early/mid game strategy. But if the build orders are executed flawlessly the match is far from decided at this point.
That number was an upper bound estimate obtained by visually integrating the commit history, it is entirely possible that it is significantly lower.
Rewriting from scratch can be beneficial, take the V8 engine in Chrome for example.
Obama went back to just killing suspected terrorists, instead of torturing them.
I think you have misconceptions about what American policy has been in the past. Remember all the US backed coups in South and Central America, accompied by mass slaughter of political opposition (e.g. Contras in Nicaragua) often by groups trained in US military schools, the extensive bombing campaigns in Vietnam, Laos and Cambodia with 100k-1m lives lost, all the US backed dictators (Suharto, Irans Shah,...), to name just a few.
Anyone with a shred of conscience would have resigned from the Bush administration sooner than later. The war against Iraq was against international law, since it was not sanctioned by an UN resolution, regardless of the war crimes commited. Unfortunately US officials are above international law, because the US has not ratified the ICC (international criminal court) treaty.
Remember there are lots of NP hard problems, that soak up any additional computing power you throw at them. A lot of them are very important (placement in EDA for example). Supercomputers today are already 1000000 faster than current desktop computers and still they are no where near where we would want them to be (to simulate the earth athmosphere at sufficient resolution for example, or all the neurons in a human brain)
I don't understand this attitude, if something is incredibly complex and very hard to get right, this can indicate that there is a mismatch between the tools you use to implement it and what you should use. With the right language it might be much more pleasant to implement: Consider for example https://github.com/GaloisInc/halfs a filesystem implemented in Haskell, I think it has ~10000loc
It can also mean that existing implementations are poorly documented and research papers on the subject place no importance on implementation details. Of course you should not use your first implementation in production, but that is as obvious as not letting someone do brain surgery who has no prior training in it.
What I don't understand is why there aren't companies around that have an interest in improving this software and pay ~1-5 people ~1 million to rewrite it in a year.