Game currencies have their own issues. If the money supply is managed poorly, then the game will fail. Too tight, and players can't afford to go on adventures, they spend 20 hours per day grinding away for a Sword of Boredom +.3141, with its special "tedium attack." Too loose and every noob who can kill three orcs is swinging a Sword of Godly Smiting +5000.
That's only true because the prices of items in the game are fixed. In the real world people adjust their prices to match supply and demand on both sides, currency as well as goods or services. There is no need to manipulate the supply of currency to maintain that balance. The rate of inflation or deflation should correspond, through the normal action of the market, to the balance between saving and consumption, as part of a self-regulating feedback system. Currency manipulation subverts that feedback and results in an excess of one or the other, to society's detriment.
I'm not quite sure where to start. They're completely different concepts. Bitcoin contracts, like the assurance contract or an escrow contract, are ways to use the Bitcoin protocol to create transactions which are only valid under certain conditions—when enough money has been collected from a variety of different inputs, or when two of three stakeholders (payer, payee, arbiter) sign off on a transfer, etc. All the data about the transfer is public, integrated into the block chain, and traceable to particular pseudonyms (public keys). Keeping track of how many bitcoins are associated with each key requires a complete record of the history dating back to the first block (gigabytes and growing).
Contracts in Open Transactions can be anything; they're basically human-readable text with semantic tags for computer parsing, identified by a cryptographic hash. Generally for currency contracts they would take the form of an agreement to pay a certain amount of BTC, USD, or some other commodity on demand. You could also have bond contracts, shares in a company, etc. A triple-entry accounting system ensures that all you need to prove who owns what is the receipt for the last transaction, signed by payer (authorizing payment and approving new balance), payee (accepting payment and approving new balance), and issuer (confirming sufficient payer funds and approving changes in balances). Payer and payee get different receipts from the issuer, of course, since they don't need to see each other's balances.
The system supports direct transfers (payer communicates with server), cheques (payer signs transfer and designates payee, payee submits transfer to the server), invoices ("negative cheques", requests for payment), and vouchers (like cashier's checks). All these forms of transfer leave a record of some sort.[1] OT also supports "cash", where you create a random token, "blind" it so that the server can't see the value, and have the server sign it (for a cost). You can then give that token to someone else, and they can deposit it with the server for credit in their own account. Once deposited, the server can see the value of the token to prevent double-spending, but can't connect the deposit with the previous withdrawal since didn't see the actual value until it was deposited. The token could have come from any previous withdrawal in that denomination (within a set time; the tokens do expire, and need to be periodically renewed).
OT also supports "smart contracts", which are programs which govern transfers. Assets can be transferred into the contract, and the program decides what happens to them after that. They can be used for implementing escrow and assurance contracts, enforcing company by-laws, or any number of other arrangements.
Because you only need to keep the most recent receipt, Open Transactions does not require anyone to store the complete history of every account, which implies much lower disk, RAM, and network requirements, and more effective pseudoanonymity compared to the Bitcoin block chain, even without dealing in cash tokens. In cash-only mode, given a reasonable amount of background noise to hide in, even the issuing server would have a hard time connecting payers with payees. Transfers are also instant, without the wait for confirmations required for Bitcoin. The downside, of course, is the requirement to trust the issuer.
[1] Vouchers need not record the payee if the voucher is open-ended (no designated payee) and the server supports converting the voucher directly into cash tokens without an account. It may also be possible to mask the payer by converting a cash deposit directly into a voucher. Either way, at least one side has to use cash to avoid associating both payer and payee with the voucher. The voucher could also be traded directly, but that exposes the recipient to considerable risk of double-spending.
Right, Bitcoin is pseudonymous, not anonymous, and even then you have to put some thought and effort into keeping your addresses independent, or the connections will be uncovered by some fairly basic traffic analysis. There are "mixing services" to deal with the latter issue, but use of one is somewhat suspicious in itself.
If you want fully anonymous transfers you need something more like an Open Transactions server running in cash-only mode. This is a federated contract-based derivative system, as opposed to a peer-to-peer base currency like Bitcoin, so you do have to trust the issuer to adhere to the contract. However, that does open up a number of interesting possibilities, including contract tokens in fixed denominations which can be transferred untraceably between users. Due to the use of blind signatures, the server cannot connect the tokens being deposited with the account they were withdrawn from, or the client making the withdrawal, and the fixed denominations make traffic analysis much more difficult.
NOTE: Open Transactions is still in early development, and is considered experimental software—more so than Bitcoin—though most of the principles involved have been well-known for some time and employed successfully in other digital cash systems. Do not reply on it for anything mission-critical at this time.
So, for somewhere between 5% and 0.8% of the cost of war that we shouldn't have started, the US Government can help to move us toward alternative energy sources, and off of foreign oil...
This may become a reasonable argument when you find a way to ensure that spending on alternative energy sources (and other projects) comes instead of, rather than in addition to, the money wasted on war. The "balanced budget amendment" proposal would be a reasonable place to start. Until then, massive overspending in one area cannot possibly justify spending other areas, no matter how small the latter might be by comparison. Quite the opposite, really; when you're that far over budget to begin with, any additional spending requires more justification, not less.
As long as DARPA's research comes to the public eventually (we got the internet, after all) it's still beneficial. Quite possibly delayed and almost certainly more expensive than it should be, but slow and expensive progress is still progress.
Unless, of course, it comes at the expense of the same result being achieved by someone else at a lower cost and/or in less time, or even other, completely unrelated, results which may be valued more highly. Whether it is beneficial for DARPA to do the research depends on the opportunity cost, even if the research ultimately produces usable results.
Naturally, when your source of funding is taxes, the (externalized) opportunity costs are never given any serious consideration.
Escrow involves trusting a third party. That adds risk.
Regular escrow with physical goods requires a fully-trusted third party, because someone has to hold the goods. Escrow with Bitcoin (or whatever you prefer to call it; it serves the same purpose) can use multisignature transactions for that purpose, preventing unilateral action on the part of the arbiter, and you can stick to pseudonyms if you wish. If you want third-party arbitration you'll obviously need to trust that the arbiter is impartial, and provide evidence to support your case, which may well reduce your anonymity.
Multisignature transactions are being implemented, but I don't see that N of M transactions are.
My understanding is the N-of-M transactions with independent private keys and signatures are supported by the protocol (via the CHECKMULTISIGVERIFY
operation), but lack a convenient interface in the GUI of the official client. It is possible to generate them with other programs and upload the transactions manually. The system I described is detailed on the Contracts pages on the Bitcoin Wiki.
If all else fails, standard cryptographic techniques exist for splitting a private key into several parts, a subset of which are required to reconstitute the original key, and these techniques can be applied to Bitcoin private keys.
The system you described, which is commonly known as "escrow", can be trivially implemented on top of Bitcoin. There are, in fact, several Bitcoin escrow providers already. Most people don't need escrow for most transactions, but it's available for those who do.
It is even possible to incorporate escrow and arbitration directly into the existing transaction format using N-of-M multisignature transactions. The simplest example would be a payment to a temporary holding address requiring two of three signatures to spend: the sender's, the recipient's, and/or an arbiter's. The sender and recipient can then agree to the payment (or refund), or the sender or recipient can appeal to the arbiter to refund the payment or complete the transfer. Since two signatures are required to spend the funds, no one can act unilaterally--not even the arbiter.
I've never really understood why it is that something you were going to do anyway becoming mandatory means that you should automatically resist it.
Some people choose to stand up for others' rights, as well as their own. The fact that I would choose to be vaccinated, and thus am not personally affected by the mandate, has zero impact on the fact that I believe it is wrong for anyone to be forced to accept any unwanted, invasive medical procedure, including vaccination.
You've lost nothing except the choice you weren't going to make, and society has benefited.
This is a contradiction. If you weren't going to make the choice to reject vaccination anyway, then society has not benefited from removing that choice. The goal of maximizing vaccination is only served by removing that choice from someone who would have taken it.
If anything, the choice itself (liberty) has positive value to society, even if no one ever chooses to exercise it, which means removing the choice from someone who would have chosen vaccination anyway is strictly detrimental. Is the heath benefit of maximizing vaccination of higher value than the cost in liberty to everyone, and in aggression against those who continue to refuse? There is no objective answer, of course, but from my point of view mandatory vaccination appears to be a very poor trade.
Which is one of many reasons the last story smelled bad, who'd wait 10 minutes for their grocery payment to clear.
Ignoring for the moment the fact that from the merchant's point of view the last story was about a traditional debit card, not actual Bitcoin transfers, ten minutes is still quite an improvement over the current system, which generally takes months to "clear" the payment beyond the risk of a chargeback. With Bitcoin, after about 10 minutes (the time it takes, on average, to incorporate the transaction into the next block in the chain) the payment is effectively final, and within an hour or so (six confirmations) there is no credible risk of reversal whatsoever.
Fundamentally, either you can say anything you want without adverse consequences, or you can't. That is the very essence of freedom of speech. If there can be adverse consequences purely because you said something, however hurtful or unpleasant, then you don't really have freedom of speech at all.
This is perfectly true, within the context of this discussion, which is the law. If there can be adverse legal consequences purely because you said something, then the law fails to recognize your freedom of speech. Adverse social consequences and the use of speech as evidence in relation to other actions have no bearing on one's legal freedom of speech.
And yet, you have also described in great detail numerous ways in which that freedom is effectively constrained, where while the speech itself is "protected" the person speaking may be penalised despite taking no other action themselves as a consequence of what they said.
The only example I gave where there was any kind of legal "penalty" was the one where theft was committed by means of fraud, and the penalty was for the theft. Similarly, self-defense is not a "penalty" for making threats. Your threats are merely evidence supporting the affirmative defense that the action against you was defense, and not aggression. You are not the one on trial here. The fact that your threats led someone to believe they needed to act in self-defense is a social consequence, not a legal one.
Were it not for the exception, anything compiled with the [gcc compiler?] would either be GPL (because of libgcc and libsupc++) or produce a linker error (because the libraries are called and not present).
I think you mean "linked with libgcc/libsupc++". One can compile code with gcc/g++ without linking against the bundled libgcc. For example, the BSD-licensed libcompiler-rt library produced for the LLVM project is said to be a drop-in replacement for libgcc, and as a bonus, it's even a bit more efficient. If the same is not already true for libsupc++, I'm sure it's only a matter of time.
And yet, for example, the Copyright Clause explicitly allows for reserving certain rights to authors. So even the Constitution itself doesn't support your position.
I shouldn't have to point this out, but the First Amendment amends the Constitution, including the Copyright Clause. Where there is a conflict, the First Amendment takes precedence.
No, you cannot legally be penalized for saying certain things. You can legally be penalized for doing certain things: for acting violently against others, as you expressed your credible intent to do by making threats
Ah, now I see. You have absolute freedom of speech, and you are penalised only for your actions, except in cases of thoughtcrime, in which case you can be punished for crimes you have not committed based only on something you said.
First, it's not punishment, it's defense. The standard "imminent risk of irreversible harm" rule for preemptive self-defense still applies, so if the threat is not imminent or the harm would be reversible then self-defense would not be justified. However, if someone believes your threat, and responds accordingly, you have only yourself to blame. If you don't want people to treat you as a threat, don't make threats.
By your argument, is hiring a hitman against the law? After all, you're not expressing any intent to commit any illegal act yourself even in the future.
That's a separate issue. The crime in this case would not be the speech, but rather inducing / conspiring with someone else to commit murder. You would be punished as an accomplice to the murder, which is a non-speech action causing actual harm.
If speech is protected absolutely, there's not much room left to prohibit fraud, so can just anyone declare themselves qualified to offer such advice (whether or not they actually are) and then charge for doing so?
I already addressed this. You entered into a contract to pay them in exchange for advise, under the explicit understanding that they were qualified to offer it. The contract is void for lack of "meeting of the minds" as a direct result of their deception. They knew otherwise, and took your money anyway: that is theft, accomplished through fraud. You would be justified in seeking both restitution and retribution for that theft.
For that matter, what if they make no claim to being qualified and simply charge for giving advice when they have no idea what they're talking about?
I have no problem with that. If they made no claim to expertise, then there is no basis for voiding the contract. Caveat emptor.
Obviously the Constitution is not self-enforcing, and if enough people decide the law doesn't matter then it won't offer any significant protection. That said, given the choice between a government whose formal case for legitimacy rests on adherence to a Constitution with protection for the rights of minorities, and whose officers swear to uphold said Constitution, and a government whose case for legitimacy rests only on maintaining the support of the majority, there is no question of which system I would prefer.
The government depends mainly on the support (or at least acquiescence) of the majority either way. Given that support, a government can get away with violating any legal limits on its power. Lacking it, even a tyrannical dictator will have difficulty staying in power for long. That being the case, I prefer a system where the majority, and especially the majority's political representatives, commit to holding universal rights (including the rights of the minority) in greater esteem than any political "mandate".
The fact that the First Amendment is often ignored by the legislature and the courts does not change the fact that it guarantees the legal right to free speech without specifying any exceptions whatsoever. The government routinely exceeds it constitutional authority; that fact does not change the substance of the law.
All the legitimate damages attributed to speech are actually caused by other actions which, unlike speech, are capable of violating another person's rights of self-ownership. Such damages can be countered without abridging the freedom of speech.
So you have freedom of speech except that you can be penalised if you say certain things and there are negative consequences? I think you and I have different understandings of the word "freedom".
No, you cannot legally be penalized for saying certain things. You can legally be penalized for doing certain things: for acting violently against others, as you expressed your credible intent to do by making threats, or for knowingly taking property which does not belong to you through fraud. The speech itself is irrelevant, except as evidence of intent or the means by which you trick someone into turning over their property. There are other means of gauging intent to cause harm, and other ways to take someone's property, and these would be treated exactly the same in the absence of any speech.
There are of course drawbacks in this system, but many people overlook the benefits - legislation cannot be challenged in court for constitutionality, resulting in a legislative program which can push changes through as long as the government has the mandate politically.
Frankly, I'm not sure any possible benefits can outweigh as major a drawback as this one. The stronger the political mandate becomes, the more need there is for constitutional protection for the rights of the minority.
And more important, you need to define what can be "property". In various times and places, such definitions included people and ideas.
I think those issues have already been addressed. You can't stake a claim to another person because there is already an owner—that person. And as implied by the GP's brief definition, ownership is not fundamentally the right to exclude others from using or benefiting from something, though that often comes with it to some degree as a side-effect of scarcity, but rather the right to freedom from interference in your own use of the property. "Ownership" of an idea is thus meaningless—as an abstract concept, there is no way for others to interfere with your use, at least not without directly interfering with you, so there is no point in staking a claim.
As far as I'm aware, nowhere in the world actually has an absolute right to free speech codified in law.
I'm amazed that you somehow managed to skip the U.S.A. in your research:
Congress shall make no law... abridging the freedom of speech, or of the press;....
There you have it, an absolute right to free speech in perfectly plain English. Any law which imposes a fine or punishment, or any other penalty, as a response to any speech on any subject whatsoever is unconstitutional.
That does not eliminate all potential fallout, of course. The First Amendment does not force any private individual to like you, or to associate with you; you can expect social ostracism and other extra-legal consequences should you choose to voice certain opinions publicly. If you declare an intent to take violent action against someone, they have the right to respond, preemptively if necessary, to that action—not the speech. Any contract founded on fraud is void, not as punishment for false speech but because it lacks "meeting of the minds". Property and/or services obtained through a fraudulent contract can be considered deliberately and maliciously stolen since, as the instigator of the fraud, you knew that the contract was void from the start and took possession of the property and/or accepted the services anyway.
All the legitimate damages attributed to speech are actually caused by other actions which, unlike speech, are capable of violating another person's rights of self-ownership. Such damages can be countered without abridging the freedom of speech. Your target is misplaced.
I never said it was reasonable to expect the "up to" speed on every connection. What I said was that the speed should only be less than advertised due to factors outside the ISP's reasonable control. Distance and upstream congestion are obviously outside of their control; throttling and underprovisioning are not.
The AC is correct; Wayland instances can be layered with minimal overhead. The idea is probably to run one system-wide instance for the login manager, which would also deal with compositing separate screens for things like fast user switching, and a separate per-session Wayland instance for each user. That not only allows customization, but also separation of permissions—the per-session instance can run under the user's account.
can Awesome be ported to Wayland itself, so that it manages X clients and native Wayland clients?
Yes, but with a twist. Wayland doesn't have window managers as a separate process. Instead of porting your preferred window manager to work with Wayland, one would implement the Wayland protocol support in the window manager, with help from libwayland for the common parts. Supposedly the Wayland support only requires about as much code as the boilerplate for an X window manager. Of course, X core rendering and XRender will be unavailable. If the WM already uses a portable library like Cairo, GTK+ or Qt for rendering that shouldn't be a problem; otherwise all the drawing code would need to be ported as well.
customers should be able to get that [advertised Internet] speed from any Internet site (network conditions permitting), not just sites which have payed the ISP for priority access.
Then you ARE a fan of so-called "net neutrality". Do you hate the terminology "net neutrality" ? You sure are for the concept behind it.
"Net neutrality" is too blunt; it would prohibit ISP from charging sites for priority access entirely. I'm fine with ISPs doing that; I just feel they shouldn't count that as Internet access for the purposes of advertising.
If an ISP wants to offer 10 Mbps access to normal sites, but charge Netflix extra for 20 Mbps, that's fine; it just can't throttle Netflix down to 5 Mbps for not paying (since that would mean the ISP was only providing 5 Mbps Internet rather than 10 Mbps), or advertise 20 Mbps Internet access if they agree. Advertising 10 Mbps from the Internet and 20 Mbps from Netflix would be fine. I'm only interested in truth in advertising, not complete ISP neutrality.
However, for security reasons, we block the Microsoft file and print sharing ports (which nobody should use directly over the Internet anyway) and outgoing port 25 (SMTP) traffic. The latter makes a huge difference in blocking spam from infected customer computers. If you ask for port 25 to be unblocked on your connection, we will unblock it.
Personally, I think this is exactly how ISPs should behave. Anything I should do differently? Is this an "Internet connection", or does the port blocking disqualify it?
Would you also unblock the file and print sharing ports on request? I don't have a problem with ports being blocked by default for security reasons, so long as they can be unblocked at no additional charge. Some ISPs block incoming HTTP/HTTPS, IRC, (non-ISP) DNS, and other traditional "server" ports, or use DPI to detect and filter specific protocols, unless you pay extra for a "business class" connection—assuming they even offer that at your address. What I really object to is essentially being sold a fancy cable TV package as if it were Internet access. The Internet is a network of peers, not just a way to consume content or upload it to someone else's servers. Far too much effort has been expended on workarounds for "peers" which can only initiate connections, not accept them.
It sounds to me like you're doing everything right, and I wish you the best of luck. I wish more ISPs were so enlightened. My own isn't so bad either (CenturyLink DSL, formerly Qwest, though it can vary by location). It's depressing how many manage to get this wrong, though—even Google's new 1Gbps broadband project has a "no servers" policy.
On the contrary, it is much easier to brute-force the password matching a known salt and hash on your own workstation, cluster, or botnet than it would be to brute-force it through repeated logins to a remote server, particularly if basic security precautions are implemented, such as rate-limiting login attempts and locking the account after several failures.
Salted and (repeatedly, as with bcrypt) hashed passwords are much better than merely hashed passwords, which are in turn somewhat better than plain-text passwords, but you really don't want any of the three out in the open. Actually reversing the hash is unlikely, but if a user with a valuable enough account picks an insecure password, not even salting will prevent it from being brute-forced from the password side.
If you really need all your accounts to be secure in the face of server data leaks, you're looking for public-key cryptography and challenge-response authentication. Server-side password checking against a hash isn't sufficient. However, if you must use passwords, at least generate them randomly on the server rather than letting users pick their own. Humans are really bad at randomness and pick passwords subject to trivial dictionary attacks far too often.
Technically, 0 bps is "up to" any speed. That doesn't mean they can get away with a connection that never works. To be meaningful, that "up to" number has to represent their best reasonable effort—in other words, you can actually get the advertised "up to" speed in the absence of factors outside the ISPs control, like upstream network issues or server-side congestion. Active throttling is something the ISP controls, and about as far from "best effort" as it's possible to get.
If you advertise a service "up to" a given amount, you may not be able to guarantee the customer will actually be able to use that amount, but you certainly shouldn't get in the way until they reach the advertised limit.
While we're on the subject, if an ISP restricts incoming connections or servers, or blocks certain services, protocols, or ports, then they aren't selling Internet service and shouldn't advertise as such. Similarly, co-located servers and caches within the ISP's network, and other services granted priority access based on a specific arrangement with the ISP, should not be counted toward the advertised Internet speed. I'm not a fan of so-called "net neutrality", but if you advertise a 10 Mbps Internet connection, customers should be able to get that speed from any Internet site (network conditions permitting), not just sites which have payed the ISP for priority access.
The strength of the concept is to have all owners join one, or perhaps a few, of these sites. 100 sites with 1.000 members is a lot less effective than 1 site with 100.000 members.
There is something to be said for a more federated approach. Sure, you don't want 100 completely isolated sites with their own distinct libraries, so that you have to join all of them to get a decent selection, but those 100 sites could get together and develop an inter-site loan system to pool their resources. Different sites could have different terms (though you would probably want the inter-site loan terms to be standardized), and a variety of sites in different jurisdictions and hosting environments would be much harder to shut down.
Slashdot Mirror
Game currencies have their own issues. If the money supply is managed poorly, then the game will fail. Too tight, and players can't afford to go on adventures, they spend 20 hours per day grinding away for a Sword of Boredom +.3141, with its special "tedium attack." Too loose and every noob who can kill three orcs is swinging a Sword of Godly Smiting +5000.
That's only true because the prices of items in the game are fixed. In the real world people adjust their prices to match supply and demand on both sides, currency as well as goods or services. There is no need to manipulate the supply of currency to maintain that balance. The rate of inflation or deflation should correspond, through the normal action of the market, to the balance between saving and consumption, as part of a self-regulating feedback system. Currency manipulation subverts that feedback and results in an excess of one or the other, to society's detriment.
How is this [Open Transactions] different from bitcoin contracts? https://en.bitcoin.it/wiki/Contracts
I'm not quite sure where to start. They're completely different concepts. Bitcoin contracts, like the assurance contract or an escrow contract, are ways to use the Bitcoin protocol to create transactions which are only valid under certain conditions—when enough money has been collected from a variety of different inputs, or when two of three stakeholders (payer, payee, arbiter) sign off on a transfer, etc. All the data about the transfer is public, integrated into the block chain, and traceable to particular pseudonyms (public keys). Keeping track of how many bitcoins are associated with each key requires a complete record of the history dating back to the first block (gigabytes and growing).
Contracts in Open Transactions can be anything; they're basically human-readable text with semantic tags for computer parsing, identified by a cryptographic hash. Generally for currency contracts they would take the form of an agreement to pay a certain amount of BTC, USD, or some other commodity on demand. You could also have bond contracts, shares in a company, etc. A triple-entry accounting system ensures that all you need to prove who owns what is the receipt for the last transaction, signed by payer (authorizing payment and approving new balance), payee (accepting payment and approving new balance), and issuer (confirming sufficient payer funds and approving changes in balances). Payer and payee get different receipts from the issuer, of course, since they don't need to see each other's balances.
The system supports direct transfers (payer communicates with server), cheques (payer signs transfer and designates payee, payee submits transfer to the server), invoices ("negative cheques", requests for payment), and vouchers (like cashier's checks). All these forms of transfer leave a record of some sort.[1] OT also supports "cash", where you create a random token, "blind" it so that the server can't see the value, and have the server sign it (for a cost). You can then give that token to someone else, and they can deposit it with the server for credit in their own account. Once deposited, the server can see the value of the token to prevent double-spending, but can't connect the deposit with the previous withdrawal since didn't see the actual value until it was deposited. The token could have come from any previous withdrawal in that denomination (within a set time; the tokens do expire, and need to be periodically renewed).
OT also supports "smart contracts", which are programs which govern transfers. Assets can be transferred into the contract, and the program decides what happens to them after that. They can be used for implementing escrow and assurance contracts, enforcing company by-laws, or any number of other arrangements.
Because you only need to keep the most recent receipt, Open Transactions does not require anyone to store the complete history of every account, which implies much lower disk, RAM, and network requirements, and more effective pseudoanonymity compared to the Bitcoin block chain, even without dealing in cash tokens. In cash-only mode, given a reasonable amount of background noise to hide in, even the issuing server would have a hard time connecting payers with payees. Transfers are also instant, without the wait for confirmations required for Bitcoin. The downside, of course, is the requirement to trust the issuer.
[1] Vouchers need not record the payee if the voucher is open-ended (no designated payee) and the server supports converting the voucher directly into cash tokens without an account. It may also be possible to mask the payer by converting a cash deposit directly into a voucher. Either way, at least one side has to use cash to avoid associating both payer and payee with the voucher. The voucher could also be traded directly, but that exposes the recipient to considerable risk of double-spending.
Right, Bitcoin is pseudonymous, not anonymous, and even then you have to put some thought and effort into keeping your addresses independent, or the connections will be uncovered by some fairly basic traffic analysis. There are "mixing services" to deal with the latter issue, but use of one is somewhat suspicious in itself.
If you want fully anonymous transfers you need something more like an Open Transactions server running in cash-only mode. This is a federated contract-based derivative system, as opposed to a peer-to-peer base currency like Bitcoin, so you do have to trust the issuer to adhere to the contract. However, that does open up a number of interesting possibilities, including contract tokens in fixed denominations which can be transferred untraceably between users. Due to the use of blind signatures, the server cannot connect the tokens being deposited with the account they were withdrawn from, or the client making the withdrawal, and the fixed denominations make traffic analysis much more difficult.
NOTE: Open Transactions is still in early development, and is considered experimental software—more so than Bitcoin—though most of the principles involved have been well-known for some time and employed successfully in other digital cash systems. Do not reply on it for anything mission-critical at this time.
So, for somewhere between 5% and 0.8% of the cost of war that we shouldn't have started, the US Government can help to move us toward alternative energy sources, and off of foreign oil...
This may become a reasonable argument when you find a way to ensure that spending on alternative energy sources (and other projects) comes instead of, rather than in addition to, the money wasted on war. The "balanced budget amendment" proposal would be a reasonable place to start. Until then, massive overspending in one area cannot possibly justify spending other areas, no matter how small the latter might be by comparison. Quite the opposite, really; when you're that far over budget to begin with, any additional spending requires more justification, not less.
As long as DARPA's research comes to the public eventually (we got the internet, after all) it's still beneficial. Quite possibly delayed and almost certainly more expensive than it should be, but slow and expensive progress is still progress.
Unless, of course, it comes at the expense of the same result being achieved by someone else at a lower cost and/or in less time, or even other, completely unrelated, results which may be valued more highly. Whether it is beneficial for DARPA to do the research depends on the opportunity cost, even if the research ultimately produces usable results.
Naturally, when your source of funding is taxes, the (externalized) opportunity costs are never given any serious consideration.
Escrow involves trusting a third party. That adds risk.
Regular escrow with physical goods requires a fully-trusted third party, because someone has to hold the goods. Escrow with Bitcoin (or whatever you prefer to call it; it serves the same purpose) can use multisignature transactions for that purpose, preventing unilateral action on the part of the arbiter, and you can stick to pseudonyms if you wish. If you want third-party arbitration you'll obviously need to trust that the arbiter is impartial, and provide evidence to support your case, which may well reduce your anonymity.
Multisignature transactions are being implemented, but I don't see that N of M transactions are.
My understanding is the N-of-M transactions with independent private keys and signatures are supported by the protocol (via the CHECKMULTISIGVERIFY operation), but lack a convenient interface in the GUI of the official client. It is possible to generate them with other programs and upload the transactions manually. The system I described is detailed on the Contracts pages on the Bitcoin Wiki.
If all else fails, standard cryptographic techniques exist for splitting a private key into several parts, a subset of which are required to reconstitute the original key, and these techniques can be applied to Bitcoin private keys.
The system you described, which is commonly known as "escrow", can be trivially implemented on top of Bitcoin. There are, in fact, several Bitcoin escrow providers already. Most people don't need escrow for most transactions, but it's available for those who do.
It is even possible to incorporate escrow and arbitration directly into the existing transaction format using N-of-M multisignature transactions. The simplest example would be a payment to a temporary holding address requiring two of three signatures to spend: the sender's, the recipient's, and/or an arbiter's. The sender and recipient can then agree to the payment (or refund), or the sender or recipient can appeal to the arbiter to refund the payment or complete the transfer. Since two signatures are required to spend the funds, no one can act unilaterally--not even the arbiter.
I've never really understood why it is that something you were going to do anyway becoming mandatory means that you should automatically resist it.
Some people choose to stand up for others' rights, as well as their own. The fact that I would choose to be vaccinated, and thus am not personally affected by the mandate, has zero impact on the fact that I believe it is wrong for anyone to be forced to accept any unwanted, invasive medical procedure, including vaccination.
You've lost nothing except the choice you weren't going to make, and society has benefited.
This is a contradiction. If you weren't going to make the choice to reject vaccination anyway, then society has not benefited from removing that choice. The goal of maximizing vaccination is only served by removing that choice from someone who would have taken it.
If anything, the choice itself (liberty) has positive value to society, even if no one ever chooses to exercise it, which means removing the choice from someone who would have chosen vaccination anyway is strictly detrimental. Is the heath benefit of maximizing vaccination of higher value than the cost in liberty to everyone, and in aggression against those who continue to refuse? There is no objective answer, of course, but from my point of view mandatory vaccination appears to be a very poor trade.
Which is one of many reasons the last story smelled bad, who'd wait 10 minutes for their grocery payment to clear.
Ignoring for the moment the fact that from the merchant's point of view the last story was about a traditional debit card, not actual Bitcoin transfers, ten minutes is still quite an improvement over the current system, which generally takes months to "clear" the payment beyond the risk of a chargeback. With Bitcoin, after about 10 minutes (the time it takes, on average, to incorporate the transaction into the next block in the chain) the payment is effectively final, and within an hour or so (six confirmations) there is no credible risk of reversal whatsoever.
Fundamentally, either you can say anything you want without adverse consequences, or you can't. That is the very essence of freedom of speech. If there can be adverse consequences purely because you said something, however hurtful or unpleasant, then you don't really have freedom of speech at all.
This is perfectly true, within the context of this discussion, which is the law. If there can be adverse legal consequences purely because you said something, then the law fails to recognize your freedom of speech. Adverse social consequences and the use of speech as evidence in relation to other actions have no bearing on one's legal freedom of speech.
And yet, you have also described in great detail numerous ways in which that freedom is effectively constrained, where while the speech itself is "protected" the person speaking may be penalised despite taking no other action themselves as a consequence of what they said.
The only example I gave where there was any kind of legal "penalty" was the one where theft was committed by means of fraud, and the penalty was for the theft. Similarly, self-defense is not a "penalty" for making threats. Your threats are merely evidence supporting the affirmative defense that the action against you was defense, and not aggression. You are not the one on trial here. The fact that your threats led someone to believe they needed to act in self-defense is a social consequence, not a legal one.
Were it not for the exception, anything compiled with the [gcc compiler?] would either be GPL (because of libgcc and libsupc++) or produce a linker error (because the libraries are called and not present).
I think you mean "linked with libgcc/libsupc++". One can compile code with gcc/g++ without linking against the bundled libgcc. For example, the BSD-licensed libcompiler-rt library produced for the LLVM project is said to be a drop-in replacement for libgcc, and as a bonus, it's even a bit more efficient. If the same is not already true for libsupc++, I'm sure it's only a matter of time.
And yet, for example, the Copyright Clause explicitly allows for reserving certain rights to authors. So even the Constitution itself doesn't support your position.
I shouldn't have to point this out, but the First Amendment amends the Constitution, including the Copyright Clause. Where there is a conflict, the First Amendment takes precedence.
No, you cannot legally be penalized for saying certain things. You can legally be penalized for doing certain things: for acting violently against others, as you expressed your credible intent to do by making threats
Ah, now I see. You have absolute freedom of speech, and you are penalised only for your actions, except in cases of thoughtcrime, in which case you can be punished for crimes you have not committed based only on something you said.
First, it's not punishment, it's defense. The standard "imminent risk of irreversible harm" rule for preemptive self-defense still applies, so if the threat is not imminent or the harm would be reversible then self-defense would not be justified. However, if someone believes your threat, and responds accordingly, you have only yourself to blame. If you don't want people to treat you as a threat, don't make threats.
By your argument, is hiring a hitman against the law? After all, you're not expressing any intent to commit any illegal act yourself even in the future.
That's a separate issue. The crime in this case would not be the speech, but rather inducing / conspiring with someone else to commit murder. You would be punished as an accomplice to the murder, which is a non-speech action causing actual harm.
If speech is protected absolutely, there's not much room left to prohibit fraud, so can just anyone declare themselves qualified to offer such advice (whether or not they actually are) and then charge for doing so?
I already addressed this. You entered into a contract to pay them in exchange for advise, under the explicit understanding that they were qualified to offer it. The contract is void for lack of "meeting of the minds" as a direct result of their deception. They knew otherwise, and took your money anyway: that is theft, accomplished through fraud. You would be justified in seeking both restitution and retribution for that theft.
For that matter, what if they make no claim to being qualified and simply charge for giving advice when they have no idea what they're talking about?
I have no problem with that. If they made no claim to expertise, then there is no basis for voiding the contract. Caveat emptor.
Obviously the Constitution is not self-enforcing, and if enough people decide the law doesn't matter then it won't offer any significant protection. That said, given the choice between a government whose formal case for legitimacy rests on adherence to a Constitution with protection for the rights of minorities, and whose officers swear to uphold said Constitution, and a government whose case for legitimacy rests only on maintaining the support of the majority, there is no question of which system I would prefer.
The government depends mainly on the support (or at least acquiescence) of the majority either way. Given that support, a government can get away with violating any legal limits on its power. Lacking it, even a tyrannical dictator will have difficulty staying in power for long. That being the case, I prefer a system where the majority, and especially the majority's political representatives, commit to holding universal rights (including the rights of the minority) in greater esteem than any political "mandate".
The fact that the First Amendment is often ignored by the legislature and the courts does not change the fact that it guarantees the legal right to free speech without specifying any exceptions whatsoever. The government routinely exceeds it constitutional authority; that fact does not change the substance of the law.
All the legitimate damages attributed to speech are actually caused by other actions which, unlike speech, are capable of violating another person's rights of self-ownership. Such damages can be countered without abridging the freedom of speech.
So you have freedom of speech except that you can be penalised if you say certain things and there are negative consequences? I think you and I have different understandings of the word "freedom".
No, you cannot legally be penalized for saying certain things. You can legally be penalized for doing certain things: for acting violently against others, as you expressed your credible intent to do by making threats, or for knowingly taking property which does not belong to you through fraud. The speech itself is irrelevant, except as evidence of intent or the means by which you trick someone into turning over their property. There are other means of gauging intent to cause harm, and other ways to take someone's property, and these would be treated exactly the same in the absence of any speech.
There are of course drawbacks in this system, but many people overlook the benefits - legislation cannot be challenged in court for constitutionality, resulting in a legislative program which can push changes through as long as the government has the mandate politically.
Frankly, I'm not sure any possible benefits can outweigh as major a drawback as this one. The stronger the political mandate becomes, the more need there is for constitutional protection for the rights of the minority.
And more important, you need to define what can be "property". In various times and places, such definitions included people and ideas.
I think those issues have already been addressed. You can't stake a claim to another person because there is already an owner—that person. And as implied by the GP's brief definition, ownership is not fundamentally the right to exclude others from using or benefiting from something, though that often comes with it to some degree as a side-effect of scarcity, but rather the right to freedom from interference in your own use of the property. "Ownership" of an idea is thus meaningless—as an abstract concept, there is no way for others to interfere with your use, at least not without directly interfering with you, so there is no point in staking a claim.
As far as I'm aware, nowhere in the world actually has an absolute right to free speech codified in law.
I'm amazed that you somehow managed to skip the U.S.A. in your research:
Congress shall make no law ... abridging the freedom of speech, or of the press; ....
There you have it, an absolute right to free speech in perfectly plain English. Any law which imposes a fine or punishment, or any other penalty, as a response to any speech on any subject whatsoever is unconstitutional.
That does not eliminate all potential fallout, of course. The First Amendment does not force any private individual to like you, or to associate with you; you can expect social ostracism and other extra-legal consequences should you choose to voice certain opinions publicly. If you declare an intent to take violent action against someone, they have the right to respond, preemptively if necessary, to that action—not the speech. Any contract founded on fraud is void, not as punishment for false speech but because it lacks "meeting of the minds". Property and/or services obtained through a fraudulent contract can be considered deliberately and maliciously stolen since, as the instigator of the fraud, you knew that the contract was void from the start and took possession of the property and/or accepted the services anyway.
All the legitimate damages attributed to speech are actually caused by other actions which, unlike speech, are capable of violating another person's rights of self-ownership. Such damages can be countered without abridging the freedom of speech. Your target is misplaced.
I never said it was reasonable to expect the "up to" speed on every connection. What I said was that the speed should only be less than advertised due to factors outside the ISP's reasonable control. Distance and upstream congestion are obviously outside of their control; throttling and underprovisioning are not.
The AC is correct; Wayland instances can be layered with minimal overhead. The idea is probably to run one system-wide instance for the login manager, which would also deal with compositing separate screens for things like fast user switching, and a separate per-session Wayland instance for each user. That not only allows customization, but also separation of permissions—the per-session instance can run under the user's account.
can Awesome be ported to Wayland itself, so that it manages X clients and native Wayland clients?
Yes, but with a twist. Wayland doesn't have window managers as a separate process. Instead of porting your preferred window manager to work with Wayland, one would implement the Wayland protocol support in the window manager, with help from libwayland for the common parts. Supposedly the Wayland support only requires about as much code as the boilerplate for an X window manager. Of course, X core rendering and XRender will be unavailable. If the WM already uses a portable library like Cairo, GTK+ or Qt for rendering that shouldn't be a problem; otherwise all the drawing code would need to be ported as well.
customers should be able to get that [advertised Internet] speed from any Internet site (network conditions permitting), not just sites which have payed the ISP for priority access.
Then you ARE a fan of so-called "net neutrality". Do you hate the terminology "net neutrality" ? You sure are for the concept behind it.
"Net neutrality" is too blunt; it would prohibit ISP from charging sites for priority access entirely. I'm fine with ISPs doing that; I just feel they shouldn't count that as Internet access for the purposes of advertising.
If an ISP wants to offer 10 Mbps access to normal sites, but charge Netflix extra for 20 Mbps, that's fine; it just can't throttle Netflix down to 5 Mbps for not paying (since that would mean the ISP was only providing 5 Mbps Internet rather than 10 Mbps), or advertise 20 Mbps Internet access if they agree. Advertising 10 Mbps from the Internet and 20 Mbps from Netflix would be fine. I'm only interested in truth in advertising, not complete ISP neutrality.
However, for security reasons, we block the Microsoft file and print sharing ports (which nobody should use directly over the Internet anyway) and outgoing port 25 (SMTP) traffic. The latter makes a huge difference in blocking spam from infected customer computers. If you ask for port 25 to be unblocked on your connection, we will unblock it.
Personally, I think this is exactly how ISPs should behave. Anything I should do differently? Is this an "Internet connection", or does the port blocking disqualify it?
Would you also unblock the file and print sharing ports on request? I don't have a problem with ports being blocked by default for security reasons, so long as they can be unblocked at no additional charge. Some ISPs block incoming HTTP/HTTPS, IRC, (non-ISP) DNS, and other traditional "server" ports, or use DPI to detect and filter specific protocols, unless you pay extra for a "business class" connection—assuming they even offer that at your address. What I really object to is essentially being sold a fancy cable TV package as if it were Internet access. The Internet is a network of peers, not just a way to consume content or upload it to someone else's servers. Far too much effort has been expended on workarounds for "peers" which can only initiate connections, not accept them.
It sounds to me like you're doing everything right, and I wish you the best of luck. I wish more ISPs were so enlightened. My own isn't so bad either (CenturyLink DSL, formerly Qwest, though it can vary by location). It's depressing how many manage to get this wrong, though—even Google's new 1Gbps broadband project has a "no servers" policy.
On the contrary, it is much easier to brute-force the password matching a known salt and hash on your own workstation, cluster, or botnet than it would be to brute-force it through repeated logins to a remote server, particularly if basic security precautions are implemented, such as rate-limiting login attempts and locking the account after several failures.
Salted and (repeatedly, as with bcrypt) hashed passwords are much better than merely hashed passwords, which are in turn somewhat better than plain-text passwords, but you really don't want any of the three out in the open. Actually reversing the hash is unlikely, but if a user with a valuable enough account picks an insecure password, not even salting will prevent it from being brute-forced from the password side.
If you really need all your accounts to be secure in the face of server data leaks, you're looking for public-key cryptography and challenge-response authentication. Server-side password checking against a hash isn't sufficient. However, if you must use passwords, at least generate them randomly on the server rather than letting users pick their own. Humans are really bad at randomness and pick passwords subject to trivial dictionary attacks far too often.
They all say "up to".
Technically, 0 bps is "up to" any speed. That doesn't mean they can get away with a connection that never works. To be meaningful, that "up to" number has to represent their best reasonable effort—in other words, you can actually get the advertised "up to" speed in the absence of factors outside the ISPs control, like upstream network issues or server-side congestion. Active throttling is something the ISP controls, and about as far from "best effort" as it's possible to get.
If you advertise a service "up to" a given amount, you may not be able to guarantee the customer will actually be able to use that amount, but you certainly shouldn't get in the way until they reach the advertised limit.
While we're on the subject, if an ISP restricts incoming connections or servers, or blocks certain services, protocols, or ports, then they aren't selling Internet service and shouldn't advertise as such. Similarly, co-located servers and caches within the ISP's network, and other services granted priority access based on a specific arrangement with the ISP, should not be counted toward the advertised Internet speed. I'm not a fan of so-called "net neutrality", but if you advertise a 10 Mbps Internet connection, customers should be able to get that speed from any Internet site (network conditions permitting), not just sites which have payed the ISP for priority access.
The strength of the concept is to have all owners join one, or perhaps a few, of these sites. 100 sites with 1.000 members is a lot less effective than 1 site with 100.000 members.
There is something to be said for a more federated approach. Sure, you don't want 100 completely isolated sites with their own distinct libraries, so that you have to join all of them to get a decent selection, but those 100 sites could get together and develop an inter-site loan system to pool their resources. Different sites could have different terms (though you would probably want the inter-site loan terms to be standardized), and a variety of sites in different jurisdictions and hosting environments would be much harder to shut down.