Mine is a Sony G500. Checking the power rating on the back it shows only 150W power draw, so sorry, my original post was in error. The figure 700W actually comes from the sum total of all of the components on my computer at their rated (rather than actual) power draw. I misremembered the total as the cost of the monitor which was plainly in error.
Honestly, at ~700W for a typical 19" monitor, your monitor is probably the highest powered device in your computer. CPU power draw varies from about 10W or less when idle, to 70W or so for a rigorous instruction mix (the Intel Itanium is somewhat anomalous at about 100W when fully exercised). So remembering to turn off you monitor, or at least selecting the low-power mode of your monitor for a screen saver rather than animating useless objects will probably have the largest effect on your power bills.
Would they have had any choices other than:
a. Tell the crew the situation, and let the crew decide whether to take a chance.
b. Breaking every single shuttle launch safety rule, try to launch another shuttle with 1-2 crew, and a bunch of space suits, before Columbia ran out of consumables.
Neither choice would look good to me.
Luckily your question is answered by the CAIB report. First, an ad-hoc wing repair using a combination of water (frozen in space), titanium tools on board the shuttle, and miscellanous junk might have held in place long enough to allow the shuttle to reenter without being destroyed. Second, by working around the clock in shifts, the next shuttle launch could have been moved up in time to rescue the Columbia with about 5 days to spare, without skipping any safety checks.
The CAIB report rejected the possibility of tranferring to the ISS (too much delta-V for the fuel left on board), and flying a different reentry pattern that would take load off of the damaged wing (too dangerous). Of course those were just the first four suggestions for approaches that might have been tried had they known that there was something wrong; no doubt there would have been dozens of other ideas floated if the engineers had had the need to do something.
The Lindows webstation PC's that they are giving away are diskless Live-CD based machines. Mouse, Keyboard, Case + CPU + Memory, CD-ROM drive. Sound and video are integrated on to the motherboard. These are really cheap boxes -- still it looks from their normal sales price that Lindows will be losing money on the deal (the machines normally sell for $169, but the settlement terms require you to produce sales receipts for any claims over $100.) Full data on the webstation is located here.
LOL! Now that is the best troll that I have seen in quite some time! I usually do not respond to them, but some people migh fall for the parent comment.
I'm not trolling. I may be in error, but on the same site that you linked to, in the Background section it states that Natural Gas, in its purest form, such as the natural gas that is delivered to your home, is almost pure methane. Methane is a molecule made up of one carbon atom and four hydrogen atoms, and is referred to as CH4. This is exactly what I claimed Natural Gas is, ie: scrubbed methane.
Searching for PG&E documents I'm not able to find any that describe the exact make-up of end-user Natural Gas in California, more exact than the overly broad "mostly methane." I did find CPUC documents that require that any Natural Gas piped into the state must have a statutory minimum of 90.8% methane (the bulk of the remainder is made up of ethane, propane, CO2, and SO2. I know from friends who work at PG&E though that at least the SO2 is removed before it is put into any of the residential supply (my fill-ups come from substations that are attached to the residential supply network.)
Not having anything like that available where I live, it begs the question: how do you dive the car out of state if it doesn't use run-of-the-mill, get-it-at-every-highway-exit gasoline?
This is a real limitation of natural gas vehicles. There are maps of locations where you can find CNG stations, but right now there aren't enough to travel freely across country. But I do a lot more commuting that I do cross country driving. An NGV is useful as a second vehicle for local area commuting, or if you travel rarely enough that you don't mind renting a car for those occasions where you need to travel across country. In my case it is a second car.
There are several commonplace uses for CNG that are tending to extend the infrastructure; many police and city vehicles are running on CNG here in the Bay Area, as are PG&E trucks, and UPS. The infrastructure costs on the order of $10k/ea. for a fast-fill fueler so the economics of distance, number of vehicles, and fuel efficiency are all playing their part in how quickly the infrastructure spreads, but it is spreading albeit at a moderate pace.
Oil is just as "renewable" as natural gas. They both come from the same hole in the ground.
You are confusing natural gas with propane. Natural Gas is just scrubbed methane. Methane is produced from innumerable sources, and is absolutely renewable; while it is often found in the same pockets as oil reserves, there is no reason at all that natural gas has to be obtained by mining; any decaying plant or animal waste will produce ample quantities of methane.
Funny you should mention the Crown Victoria, since it is one of two commercially available alternative fuel vehicles. Both the Ford Crown Victoria, and Honda's Civic are available in Natural Gas models. As alternative fuel vehicles go, these are pretty easy to use -- they handle like gasoline engines, are easy to refill (in California) at many PG&E stations in the area, at the cost of about half your trunk space, and about two-thirds the range between fill-ups of the same gasoline powered vehicle. Or, like an EV, you can fill up at home.
Natural gas is almost entirely domestically produced. It costs less than $1.50 a gallon of gasoline equivalent, and it is renewable and clean. Plus, here in California an alternative fuel and low emission vehicle gets you permission to use the commuter lanes, even over toll-bridges, which can save considerable time and money.
And in case you hadn't guessed, I like mine pretty well. Range and trunk space aren't good, but they aren't embarrasing either. Availability needs to be improved, but I happen to have a PG&E station just along my commute path, so it works out fine for me.
Gotta disagree with the separate islands comment. Software being open, and standards being public are related, but not necessary for one another.
I was writing factually rather than theoretically. Open software happens to link all of those systems together. One can theorize that the same interconnections could/would have happened eventually through some other hypothetical open standard for interconnection that happened to be wholly implemented as proprietary software, but that doesn't reflect what actually happened.
First of all public goods only need to benefit enough people to enjoy a majority support. Does the military benefit everyone? How about the people who disagree with how the military is being used? Does the bus system benefit everyone? Does welfare benefit everyone?
Arguably all of these do support everyone; that is, everyone benefits by living in a country where the destitute don't have to resort to theft to avoid starvation, everyone benefits from living in a society which is well protected from foreign aggressors, and society as a whole benefits from having people who are incapable of passing a driving test, or unable to afford a car, never the less able to hold a job and be productive so that they won't have to live on welfare.
Likewise open software benefits everyone -- if not directly then indirectly -- in lower prices for services, in greater productivity, resulting in greater general prosperity, in better and cheaper communication technologies, and greater efficiency for those areas that open software is able to cover.
Where would we be without open software. Let's see. No email, no Internet (no DNS), no TCP/IP, no world wide web, no interoperable software. Novell, Microsoft, MacOS, and the mainframes would all still be separate islands.
You've connected something that I intended to be unconnected. Debian installation is unpolished because it insists that I remember all of the pieces of hardware that I have installed in my machine. I haven't kept track of the hardware in my chassis since the bad old days of ISA.
Debian is too BSD'ish for my tastes in that I am much more comfortable with the System V style of system management and
system layout having used the AT&T derived systems since '85. I had a SunOS box on my desk for about half a year back in '92 or '93, but was mostly transitioning from my AIX workhorse, to a new HPUX box at the time and never grew to like SunOS before it was replaced with Solaris.
I realize that lots of people disagree with me on this, but then I also prefer 'vi' over 'emacs'. Actually that is too weak -- as far as I can tell
'emacs' sucks utterly. Let the flames begin...
I strongly suspect that what RedHat is worried about is looking good now. It's quite possible, even likely, that SCO's FUD is making some of RedHat's customers worry about their legal position, and if that's true than RedHat really has to do something about it. A willingness to launch a lawsuit to protect their customers' interests is exactly the kind of thing that will reassure those customers.
Actually it has already made Redhat at least a couple of bucks. My RHN demo account was expiring, and I could either renew it by spending a couple of bucks, or by filling out a long questionnaire. At the same time Redhat seemed to be disappearing from my radar having cancelled their desktop linux box set, leaving me in a bit of a quandary for what to do to find a replacement. Debian probably represents the best way of managing a distribution, but the end result is much too unpolished and BSD'ish for my tastes. SuSE might be an alternative although it doesn't have a huge presence in the USA.
Anyway, Redhat filing this lawsuit puts them enough in alignment with my own priorities that I've reconsidered, and signed up with RHN for a couple of Basic accounts. I don't really doubt that RH is primarily concerned about proving the legality of its Enterprise offering, so if the desktop distribution goes seriously out of whack then I'll be looking again, but I'm willing for a while to see what their new mode of operating will put out.
The sad part is that actually used to be a Caldera customer up until Caldera left customers who had purchased 1.0 without an upgrade path. SCO is suffering a terrible and sad loss of judgement.
Why are you so against using the discount card? You say "privacy", but it seems you are blowing it way out of proportion.
Once upon a time there were things called coupons. Manufacturers of various products used to give you these coupons; some in newspaper ads, some on product containers, some in the store. These coupons saved you maybe a quarter, or 40c on a typical purchase of a couple of dollars. Not a lot, but if you were tight on cash it would add up over time; maybe $5-6 on a typical shopping run of $100 or so.
But coupons were pretty expensive in terms of time to process. Expensive for the consumer to clip, expensive for the store to store, and expensive to redeem.
Enter discount cards. At first these were used to track store coupons rather than have to collect slips of paper and turn them in at the register. Ultimately these cards replaced almost every other form of coupon. Now of itself that doesn't bother me very much as I never clipped coupons anyway. What annoys me about discount cards is that to make them more attractive to their customers, stores started raising prices so that they could give artificially increased discounts so that the card holders could feel better about using their cards.
This in turn has led to an increased interest in these cards so that while I used to simply avoid stores with these cards, consumer demand for the cards has forced every single market in my area to supply them (actually the Berkeley Bowl is an exception, but at a 40m drive it is pushing my boundaries for in my area.) These days if I check my register receipt on a $140 of groceries and the register card is used, it is rare to find less than $20 in supposed discounts on my receipt.
Raising my prices by 20% so that a bunch of people can save 2% on their shopping bills (plus the 20% markup) is annoying in itself. The data that is
actually being collected has completely unknown value -- the applications to sort through that kind of data stream and extract any meaningful information from it are only in their infancy. This is the field of data-mining, and is one of the research areas in my lab.
Whether the data collected actually has any noticeable impact on your freedom or any other really objectionable uses is something you'll be finding out in the next five to ten years. To me it just seems like a scam start to finish, and one that annoys me because I end up paying for it.
They give you a temporary one that you then you are supposed to go over to the customer service desk to fill out the forms. I let her swipe it, then throw it in the trash below the register.
Dude! I've tried that, but it doesn't work after the fifteenth time you go in, in one week, to buy jello and hosiery. The checkers start to recognize you...
Seriously, there are many ways around the system depending on what your goals are. My goals are impress upon the businesses that I buy from that one of their values to me as their customer, is that they are able to perform anonymous transactions. So I make it a point to complain, and hope that others will do likewise -- the only way to change business practices is to let them know what it is that you don't like.
Pretending to go along and cheating the system doesn't get that point across nearly as well.
There is already price discrimination. Every time I walk into a grocery store I pay a premium for my food in order to maintain my privacy by refusing to use a discount card. Realistically I represent a small minority of consumers who values privacy over money and the market can charge a premium for selling to me and others like me.
It pisses me off every time I'm in a store, but I only get really angry when the checker says something like 'Sir, you would have saved $15 on this purchase if you had used your discount card. Would you like me to give you one now that I'll use for this purchase.' If I have to pay outrageous fines to maintain my privacy, I'd rather not know how outrageous they are.
Recently (probably as complaints have risen from my demographic), most of upscale markets in our area have started granting the discount anyway if you tell them that you value your privacy, and they swipe a register card instead.
Presumably they now are collecting data on privacy freaks, but at least it is
as a group rather than as individuals.
Database server communicates with clients using ESP/IPSec protected communications. The [audit copy of the] ballot contains: 1: An easy-to-scan bar code
2: A human readable ballot listing for manual verification. 3: The ballot serial number.
I've been thinking about this since the 2002 election on and off. I think that storing duplicate copies of original data is asking for trouble. If the physical ballot slip is needed then those ballots should ideally be the only canonical source for the information. IF backup data is really needed then it should be in the form of a physical backup -- it should be easy to inspect the physical machinery and verify that it can only create backups in a trustworthy
way.
Likewise duplicating the information between bar-code and printed text is asking for trouble IMHO.
Borrowing some ideas from the accounting practices of the retail industry, I would suggest that the balloting data be printed using
physical carbon copy to three layers of continuous roll paper, with the third copy being torn off and returned to the voter. On that paper the voting record
should be printed once in human readable form, using an OCR-able font. The first and second records should then be divided and spooled separately.
The spool containing the carbon copy should be notarized and stored in a fire safe container, while the original should be scanned for summary by the state, and then by any other independent organization who is willing to bring equipment to the polling place.
There are 32k class B addresses so there are about 32000 corporations or organizations which own the range of addresses you give. There are also
127 class A addresses, many of those also controlled by corporations or
organizations. From a quick perusal of the registrations we find:
GE (3.x.x.x), GTEI (4.x.x.x and 8.x.x.x), army.mil (6.x.x.x, and 55.x.x.x),
AT&T (12.x.x.x, 32.x.x.x), Xerox (13.x.x.x), HP (15.x.x.x, 16.x.x.x), Apple (17.x.x.x), MIT (18.x.x.x), Ford (19.x.x.x), CSC (20.x.x.x), ARIN.NET (24, 63-69),
ucl.ac.uk (25), nipr.mil (33), inet-hou.com (34), merit.edu (35), psi.net (38), uu.net (40), v6nic.net (43), ampr.org (44), vt.edu (45), Nortel (46), Dupont (52), debir.de (53), usps.gov (56), equant.net (57), apnic.net (60, 61), ripe.net (62, 80-82).
Those are all of the ones that respond to an in-addr.arpa request. It would be interesting to see how many of those listed actually use their addressable space. ARIN, RIPE, and APNIC provide subdivided blocks of addresses to Europe, Asia, and North America. Net 34 (inet-hou.com) appears to be the personal property of a Houston resident named Richard Harrison. Net 44 (ampr.org) is the amateur packet radio subnet, and there are a few other ISPs there, like 40 (uu.net), 38 (psi.net), and probably one or both of the AT&T class A's. And there are a few universities both in the US and one in the UK. I would suspect that most of the corporate subnets are firewalled anyway, so moving any of those would represent only the inconvenience of renumbering their networks -- but it isn't as if the machines were actually reachable from the 'net.
This pisses me off. OK, so it does not say you have to distribute your code. But it does say you must permit modification [for personal use] of the work and you must permit reverse engineering for debugging the modifications the other user decided to do.
Those are both fair uses. The LGPL does not permit users to link with LGPL'd code if their license explicitly prohibits fair use -- in fact it goes out of its way to ensure that linked LGPL code can be replaced with other linked LGPL code.
In other RFID news today, Wired is reporting that the EU may implant RFID tags into the Euro, basically eliminating the anonymous cash transaction.
To eliminate anonymous transactions they would first have to ask for ID before either giving change, or accepting cash. Identifying the bill doesn't identify the person who holds it (notice that all US notes carry a unique number as well; ooh! they are watching us!) If you want to theorize that the data could all be collected and used for central tracking of the flow of public money then you'll have to admit that the same thing is possible for any serially numbered printed bill (indeed serial numbers have been used in the US to trace criminal money laundering operations.) The EC just wants to make their bills harder to forge.
Reading through the description, I get the idea of a binaries friendly version of Usenet. Unfortunately it is only Internet visible end-points are capable of receiving its broadcasts.
I think a combination of this application, with the BitTorrent idea of using all of the routable hosts as forwarding points to the nonroutable hosts, would much farther reaching.
If a store clerk sold you a new computer, only to find that it wasn't general stock but intended to be a sales terminal, it wouldn't mean that they could find you and take the computer back. (With some small exceptions.)
The difference here is consideration. Unless the downloaders can show that they have given some consideration in exchange for the software license, there is nothing preventing Nullsoft from retroactively invalidating the license. In this case there hasn't been enough time between the 'release' and the revocation for the normal open source consideration of gift in kind to apply.
Seriously, pull up "top" or something and tell me if bittorrent actually uses nontrivial CPU. I could be wrong, but I'd be very surprised.
During a normal download BT isn't a significant CPU hog, but the SHA1 recalculations can take quite a while when restarting a transfer, even on a fast machine.
If the Tracker itself, had this built in, i propose it could do it more efficently, and with less setup hassle. Imagine being able to setup a mirror by simply having the admin place your new "cluster-able" tracker IP:Port on an approved mirror list. The main tracker could refer clients to a mirror after behind-the-scenes communication to determine which mirror has least load.
A step below this, but better than DNS round-robin, would be to give the client an array of tracker addresses. This is better than DNS because you don't get the stalled server mixed with cached DNS record causing inaccessibility. The clients could try connections randomly to the servers in the array, and prevent cached dns records for altering distribution.
How is providing the client an array of tracker addresses any different from DNS round-robin? Isn't it up to the client to pick an IP address from the list of
available A records when using DNS round-robin? In any case all of the BT clients I am aware of will follow HTTP-Redirect headers so it is possible to set up a meta-server for redirect based on some arbitrary load balancing algorithm, so you can already do the kind of cluster load balancing you are requesting.
Slashdot Mirror
I quite like 'Tiny Labor' as an anagram for 'Tony Blair'.
Mine is a Sony G500. Checking the power rating on the back it shows only 150W power draw, so sorry, my original post was in error. The figure 700W actually comes from the sum total of all of the components on my computer at their rated (rather than actual) power draw. I misremembered the total as the cost of the monitor which was plainly in error.
Honestly, at ~700W for a typical 19" monitor, your monitor is probably the highest powered device in your computer. CPU power draw varies from about 10W or less when idle, to 70W or so for a rigorous instruction mix (the Intel Itanium is somewhat anomalous at about 100W when fully exercised). So remembering to turn off you monitor, or at least selecting the low-power mode of your monitor for a screen saver rather than animating useless objects will probably have the largest effect on your power bills.
a. Tell the crew the situation, and let the crew decide whether to take a chance.
b. Breaking every single shuttle launch safety rule, try to launch another shuttle with 1-2 crew, and a bunch of space suits, before Columbia ran out of consumables.
Neither choice would look good to me.
Luckily your question is answered by the CAIB report. First, an ad-hoc wing repair using a combination of water (frozen in space), titanium tools on board the shuttle, and miscellanous junk might have held in place long enough to allow the shuttle to reenter without being destroyed. Second, by working around the clock in shifts, the next shuttle launch could have been moved up in time to rescue the Columbia with about 5 days to spare, without skipping any safety checks.
The CAIB report rejected the possibility of tranferring to the ISS (too much delta-V for the fuel left on board), and flying a different reentry pattern that would take load off of the damaged wing (too dangerous). Of course those were just the first four suggestions for approaches that might have been tried had they known that there was something wrong; no doubt there would have been dozens of other ideas floated if the engineers had had the need to do something.
The Lindows webstation PC's that they are giving away are diskless Live-CD based machines. Mouse, Keyboard, Case + CPU + Memory, CD-ROM drive. Sound and video are integrated on to the motherboard. These are really cheap boxes -- still it looks from their normal sales price that Lindows will be losing money on the deal (the machines normally sell for $169, but the settlement terms require you to produce sales receipts for any claims over $100.) Full data on the webstation is located here.
I'm not trolling. I may be in error, but on the same site that you linked to, in the Background section it states that Natural Gas, in its purest form, such as the natural gas that is delivered to your home, is almost pure methane. Methane is a molecule made up of one carbon atom and four hydrogen atoms, and is referred to as CH4. This is exactly what I claimed Natural Gas is, ie: scrubbed methane.
Searching for PG&E documents I'm not able to find any that describe the exact make-up of end-user Natural Gas in California, more exact than the overly broad "mostly methane." I did find CPUC documents that require that any Natural Gas piped into the state must have a statutory minimum of 90.8% methane (the bulk of the remainder is made up of ethane, propane, CO2, and SO2. I know from friends who work at PG&E though that at least the SO2 is removed before it is put into any of the residential supply (my fill-ups come from substations that are attached to the residential supply network.)
This is a real limitation of natural gas vehicles. There are maps of locations where you can find CNG stations, but right now there aren't enough to travel freely across country. But I do a lot more commuting that I do cross country driving. An NGV is useful as a second vehicle for local area commuting, or if you travel rarely enough that you don't mind renting a car for those occasions where you need to travel across country. In my case it is a second car.
There are several commonplace uses for CNG that are tending to extend the infrastructure; many police and city vehicles are running on CNG here in the Bay Area, as are PG&E trucks, and UPS. The infrastructure costs on the order of $10k/ea. for a fast-fill fueler so the economics of distance, number of vehicles, and fuel efficiency are all playing their part in how quickly the infrastructure spreads, but it is spreading albeit at a moderate pace.
You are confusing natural gas with propane. Natural Gas is just scrubbed methane. Methane is produced from innumerable sources, and is absolutely renewable; while it is often found in the same pockets as oil reserves, there is no reason at all that natural gas has to be obtained by mining; any decaying plant or animal waste will produce ample quantities of methane.
Natural gas is almost entirely domestically produced. It costs less than $1.50 a gallon of gasoline equivalent, and it is renewable and clean. Plus, here in California an alternative fuel and low emission vehicle gets you permission to use the commuter lanes, even over toll-bridges, which can save considerable time and money.
And in case you hadn't guessed, I like mine pretty well. Range and trunk space aren't good, but they aren't embarrasing either. Availability needs to be improved, but I happen to have a PG&E station just along my commute path, so it works out fine for me.
I was writing factually rather than theoretically. Open software happens to link all of those systems together. One can theorize that the same interconnections could/would have happened eventually through some other hypothetical open standard for interconnection that happened to be wholly implemented as proprietary software, but that doesn't reflect what actually happened.
First of all public goods only need to benefit enough people to enjoy a majority support. Does the military benefit everyone? How about the people who disagree with how the military is being used? Does the bus system benefit everyone? Does welfare benefit everyone?
Arguably all of these do support everyone; that is, everyone benefits by living in a country where the destitute don't have to resort to theft to avoid starvation, everyone benefits from living in a society which is well protected from foreign aggressors, and society as a whole benefits from having people who are incapable of passing a driving test, or unable to afford a car, never the less able to hold a job and be productive so that they won't have to live on welfare.
Likewise open software benefits everyone -- if not directly then indirectly -- in lower prices for services, in greater productivity, resulting in greater general prosperity, in better and cheaper communication technologies, and greater efficiency for those areas that open software is able to cover.
Where would we be without open software. Let's see. No email, no Internet (no DNS), no TCP/IP, no world wide web, no interoperable software. Novell, Microsoft, MacOS, and the mainframes would all still be separate islands.
Yeah, I guess that doesn't add up to squat.
Debian is too BSD'ish for my tastes in that I am much more comfortable with the System V style of system management and system layout having used the AT&T derived systems since '85. I had a SunOS box on my desk for about half a year back in '92 or '93, but was mostly transitioning from my AIX workhorse, to a new HPUX box at the time and never grew to like SunOS before it was replaced with Solaris.
I realize that lots of people disagree with me on this, but then I also prefer 'vi' over 'emacs'. Actually that is too weak -- as far as I can tell 'emacs' sucks utterly. Let the flames begin...
Actually it has already made Redhat at least a couple of bucks. My RHN demo account was expiring, and I could either renew it by spending a couple of bucks, or by filling out a long questionnaire. At the same time Redhat seemed to be disappearing from my radar having cancelled their desktop linux box set, leaving me in a bit of a quandary for what to do to find a replacement. Debian probably represents the best way of managing a distribution, but the end result is much too unpolished and BSD'ish for my tastes. SuSE might be an alternative although it doesn't have a huge presence in the USA.
Anyway, Redhat filing this lawsuit puts them enough in alignment with my own priorities that I've reconsidered, and signed up with RHN for a couple of Basic accounts. I don't really doubt that RH is primarily concerned about proving the legality of its Enterprise offering, so if the desktop distribution goes seriously out of whack then I'll be looking again, but I'm willing for a while to see what their new mode of operating will put out.
The sad part is that actually used to be a Caldera customer up until Caldera left customers who had purchased 1.0 without an upgrade path. SCO is suffering a terrible and sad loss of judgement.
Once upon a time there were things called coupons. Manufacturers of various products used to give you these coupons; some in newspaper ads, some on product containers, some in the store. These coupons saved you maybe a quarter, or 40c on a typical purchase of a couple of dollars. Not a lot, but if you were tight on cash it would add up over time; maybe $5-6 on a typical shopping run of $100 or so.
But coupons were pretty expensive in terms of time to process. Expensive for the consumer to clip, expensive for the store to store, and expensive to redeem.
Enter discount cards. At first these were used to track store coupons rather than have to collect slips of paper and turn them in at the register. Ultimately these cards replaced almost every other form of coupon. Now of itself that doesn't bother me very much as I never clipped coupons anyway. What annoys me about discount cards is that to make them more attractive to their customers, stores started raising prices so that they could give artificially increased discounts so that the card holders could feel better about using their cards.
This in turn has led to an increased interest in these cards so that while I used to simply avoid stores with these cards, consumer demand for the cards has forced every single market in my area to supply them (actually the Berkeley Bowl is an exception, but at a 40m drive it is pushing my boundaries for in my area.) These days if I check my register receipt on a $140 of groceries and the register card is used, it is rare to find less than $20 in supposed discounts on my receipt.
Raising my prices by 20% so that a bunch of people can save 2% on their shopping bills (plus the 20% markup) is annoying in itself. The data that is actually being collected has completely unknown value -- the applications to sort through that kind of data stream and extract any meaningful information from it are only in their infancy. This is the field of data-mining, and is one of the research areas in my lab.
Whether the data collected actually has any noticeable impact on your freedom or any other really objectionable uses is something you'll be finding out in the next five to ten years. To me it just seems like a scam start to finish, and one that annoys me because I end up paying for it.
Dude! I've tried that, but it doesn't work after the fifteenth time you go in, in one week, to buy jello and hosiery. The checkers start to recognize you...
Seriously, there are many ways around the system depending on what your goals are. My goals are impress upon the businesses that I buy from that one of their values to me as their customer, is that they are able to perform anonymous transactions. So I make it a point to complain, and hope that others will do likewise -- the only way to change business practices is to let them know what it is that you don't like.
Pretending to go along and cheating the system doesn't get that point across nearly as well.
It pisses me off every time I'm in a store, but I only get really angry when the checker says something like 'Sir, you would have saved $15 on this purchase if you had used your discount card. Would you like me to give you one now that I'll use for this purchase.' If I have to pay outrageous fines to maintain my privacy, I'd rather not know how outrageous they are.
Recently (probably as complaints have risen from my demographic), most of upscale markets in our area have started granting the discount anyway if you tell them that you value your privacy, and they swipe a register card instead. Presumably they now are collecting data on privacy freaks, but at least it is as a group rather than as individuals.
I've been thinking about this since the 2002 election on and off. I think that storing duplicate copies of original data is asking for trouble. If the physical ballot slip is needed then those ballots should ideally be the only canonical source for the information. IF backup data is really needed then it should be in the form of a physical backup -- it should be easy to inspect the physical machinery and verify that it can only create backups in a trustworthy way.
Likewise duplicating the information between bar-code and printed text is asking for trouble IMHO.
Borrowing some ideas from the accounting practices of the retail industry, I would suggest that the balloting data be printed using physical carbon copy to three layers of continuous roll paper, with the third copy being torn off and returned to the voter. On that paper the voting record should be printed once in human readable form, using an OCR-able font. The first and second records should then be divided and spooled separately.
The spool containing the carbon copy should be notarized and stored in a fire safe container, while the original should be scanned for summary by the state, and then by any other independent organization who is willing to bring equipment to the polling place.
GE (3.x.x.x), GTEI (4.x.x.x and 8.x.x.x), army.mil (6.x.x.x, and 55.x.x.x), AT&T (12.x.x.x, 32.x.x.x), Xerox (13.x.x.x), HP (15.x.x.x, 16.x.x.x), Apple (17.x.x.x), MIT (18.x.x.x), Ford (19.x.x.x), CSC (20.x.x.x), ARIN.NET (24, 63-69), ucl.ac.uk (25), nipr.mil (33), inet-hou.com (34), merit.edu (35), psi.net (38), uu.net (40), v6nic.net (43), ampr.org (44), vt.edu (45), Nortel (46), Dupont (52), debir.de (53), usps.gov (56), equant.net (57), apnic.net (60, 61), ripe.net (62, 80-82).
Those are all of the ones that respond to an in-addr.arpa request. It would be interesting to see how many of those listed actually use their addressable space. ARIN, RIPE, and APNIC provide subdivided blocks of addresses to Europe, Asia, and North America. Net 34 (inet-hou.com) appears to be the personal property of a Houston resident named Richard Harrison. Net 44 (ampr.org) is the amateur packet radio subnet, and there are a few other ISPs there, like 40 (uu.net), 38 (psi.net), and probably one or both of the AT&T class A's. And there are a few universities both in the US and one in the UK. I would suspect that most of the corporate subnets are firewalled anyway, so moving any of those would represent only the inconvenience of renumbering their networks -- but it isn't as if the machines were actually reachable from the 'net.
Those are both fair uses. The LGPL does not permit users to link with LGPL'd code if their license explicitly prohibits fair use -- in fact it goes out of its way to ensure that linked LGPL code can be replaced with other linked LGPL code.
To eliminate anonymous transactions they would first have to ask for ID before either giving change, or accepting cash. Identifying the bill doesn't identify the person who holds it (notice that all US notes carry a unique number as well; ooh! they are watching us!) If you want to theorize that the data could all be collected and used for central tracking of the flow of public money then you'll have to admit that the same thing is possible for any serially numbered printed bill (indeed serial numbers have been used in the US to trace criminal money laundering operations.) The EC just wants to make their bills harder to forge.
A poke with my fork
Gelatinous pink cubeness
It twitches slightly
Spam-ku
I think a combination of this application, with the BitTorrent idea of using all of the routable hosts as forwarding points to the nonroutable hosts, would much farther reaching.
The difference here is consideration. Unless the downloaders can show that they have given some consideration in exchange for the software license, there is nothing preventing Nullsoft from retroactively invalidating the license. In this case there hasn't been enough time between the 'release' and the revocation for the normal open source consideration of gift in kind to apply.
During a normal download BT isn't a significant CPU hog, but the SHA1 recalculations can take quite a while when restarting a transfer, even on a fast machine.
A step below this, but better than DNS round-robin, would be to give the client an array of tracker addresses. This is better than DNS because you don't get the stalled server mixed with cached DNS record causing inaccessibility. The clients could try connections randomly to the servers in the array, and prevent cached dns records for altering distribution.
How is providing the client an array of tracker addresses any different from DNS round-robin? Isn't it up to the client to pick an IP address from the list of available A records when using DNS round-robin? In any case all of the BT clients I am aware of will follow HTTP-Redirect headers so it is possible to set up a meta-server for redirect based on some arbitrary load balancing algorithm, so you can already do the kind of cluster load balancing you are requesting.