That's exactly what I said. I said that allowing authentication via SMTP should be a _prerequisite_ for using SPF on your own domain. For example, if r0x0rISP starts using SPF, and they add their single mail server, smtp.r0x0rISP.com to their SPF list, and that mail server does not allow relaying or authentication, then there is a problem since my mails from fryguy r0x0rISP (using whatever mail server I can use right now) will not be on the SPF list, so they will get a negative spamassassin score, or blocked, or whatever.
The only problem with your suggestion is that mail servers don't allow relaying. I have an email address that I don't subscribe to the ISP anymore (paid to keep the email). How the heck am I supposed to send mail through them? I'll get a relaying denied error. I know that there are SMTP authorization methods, but most mail servers don't use them. Until people from any network (that are allowed to use From: email@mailserver.com) can send a mail through your mail server, then that mail server should not use SPF.
Why would they need to be huge? Just a single key in the DNS -- have your public key be signed by their private key (with which their public key is in DNS), and include the encrypted public key in your message. When the recipient receives your message, first they go to dns, downloads yahoo's public key, decrypts your public key, then uses it to verify that the email signature is correct. A side benefit is that this can be cached and not require so many lookups. A downside is that mail may not be as anonymous anymore since your public key may be associated with you.
If you have to contact yahoo to add your key to the dns, then why not instead just contact yahoo to sign your public key? Either way requires contact.
In my opinion, the best way would be to require authorization on SMTP (or a new standard) servers, and then just sign the emails on the server. This would require no change in software on the user's end, and isn't too much more computationally intensive than say using SSL to check your mail.
Their site says that they can't do something for free for the candidates, so they have to charge. The money that Amazon earns by this will be donated to a charity.
I know at least for half-life on steam, that only the core game downloads first, so you may have to wait for the first 20-30 megs before you can play. However, the levels and media are downloaded on demand (download level 2 and all textures that aren't in the cache while you're playing level 1, and so on)
Perhaps I'm wrong, but aren't flourescent bulbs usually sealed glass with a gas inside, and applying current through the bulb is what makes it light up? I would've thought that you using your dremel on it would break the bulb and let all the gas out. Perhaps I'm wrong though. Anyone with some insight?
It's not like they don't have a year from the time the movie goes from script to being produced, to being in theatres to being shipped. The studios can produce the DVD's before the movie starts playing in theatres. The only reason they don't is because more money is made at theatres.
Alternatively, a much more reliable method is using one of the serial cable controls if your box supports it at all. I bought a $5 DB-9 RJ-11(?) converter for my RCA DirecTV tuner, and it was just a matter of setting the tuner up to use the cable instead of "Home Control" mode, and a perl script to open/dev/ttyS0 (search the mythtv-users mailing list for RCA.pl if you have this one)
most people use basic stamps for their little robots. It can't run a "decent" OS, but I believe they have a module that allows them to receive radio data.
Correct me if I'm wrong, but the configuration app in SuSe lets you do this (YaST2 I think). It starts a temporary X server to test out the resolution to make sure it works, otherwise it will return you back in 30 seconds. If you apply the changes it restarts X somehow saving your open programs and getting the new resolution.
Our own Punica-device is powered by a 40MHz Intel i486 CPU with 16MiB RAM Main costs:
* computer was sponsored
I sure hope no company went out and bought these guys such a powerful computer in exchange for their name on it.. It might have run them out of business:)
Maybe they didn't make clean. I sometimes do this and get a false sense of security in how fast my computer when upgrading apps, but only to be sorely disappointed at the 3 hours stuff takes to compile when I clean it out. Maybe someone can buy me one of these things:)
So we should be calling them the Primary Primary drive?
Re:Cool linux desktop commercial
on
Recycling TV Ads
·
· Score: 1
My dad thought this would be a great Star Trek ad. A Klingon and Federation member drive and pick up a Romulan, then after a while they decide the Romulan sticks and throw him out.
"If someone came to us and said, `We want Elvis to sing this new song,' we'd have a lot to contemplate," he said. "We tried to retain the integrity of his original song with the remixes. Now you're talking about a whole new vocal performance of a song he never sang or knew? How do we know he'd want to sing it?"
Why not just ask for a synthetic version of an Elvis impersonator to sing the song? That should sound pretty close.
PHPEdit (phpedit.net) uses a similar method to develop their software. Basically their software is free, and if you want features you donate and they will implement them after reaching a certain commitment level based on the number of hours it will take to add the feature. The work is done by their team presumably, so it's not really for open source projects, just an innovative way to "sell" the product.
Ah, but the problem with this is the fact that different move orders result in the same board configuration. Take for instance: [1. e4 e5 2. d4] vs [1. d4 e5 2. e4]. The single-computer would already know that the first one was score +0.005, so then it wouldn't recalculate the score to be.005 for the second one since it knows they are the same board position. Transmitting all the scores for all the moves would be pointless since that would take too much bandwidth to do.
Admittedly the programmers have to understand chess thoroughly, but Kasparov could wipe the floor with the lot of them
This is not necessarily true. A successful checkers player was built with the creator not knowing much at all about checkers. See One Jump Ahead for more info.
I believe the algorithm traditionally used for chess programs is not good for such incredible parallelism like distributed computing would provide. Distributed computing is best for independant modules that can be solved seperately (i.e. see if any of the numbers 1232412000000000000-1232413000000000000 divide 5454934934923283184138231231231253849348096454). The min-max algorithm (i think that's what it's called) requires information about the scoring of following points on the graph in order to determine how well the current score is going to do. It seems almost counter-intuitive that so many computers doing work couldn't make something more effective, but it's true.
First off--Where at, then, in the IPv4 packet header, do you suggest putting the "differentiator"? Oh, shucks, I guess there isn't much space left in the header. (I'm interpreting you literally, here.)
Maybe the evil bit can be used.. Sure, maybe Russia and China may have it set and some routers might mistake their packets as evil, but they're Communist anyway so the routers shouldn't care!
While you may be correct about IRV having problems and Condorcet method being better, I don't understand why nobody suggests (or at least nobody that I have seen at least) that voters use the IRV method to vote, while the tabulators use the Condorcet method to tally the winners. The ranking system used by IRV is just a logical subset of the possible choices for condorcet.. and I doubt that people will want to sway from configurations other than the ones built by using rankings.
A simple algorithm for converting rankings to condorcet voting would be:
bool table[CANDIDATES][CANDIDATES]; for (int i=0; iCANDIDATES; i++)
for (int j=0; jCANDIDATES; j++)
table[i][j] = rank[i]rank[j];
Slashdot Mirror
He does not still use a typewriter. According to the author's notes at the end of each Xanth book, he converted to using a PC 6-7 years ago.
That's exactly what I said. I said that allowing authentication via SMTP should be a _prerequisite_ for using SPF on your own domain. For example, if r0x0rISP starts using SPF, and they add their single mail server, smtp.r0x0rISP.com to their SPF list, and that mail server does not allow relaying or authentication, then there is a problem since my mails from fryguy r0x0rISP (using whatever mail server I can use right now) will not be on the SPF list, so they will get a negative spamassassin score, or blocked, or whatever.
The only problem with your suggestion is that mail servers don't allow relaying. I have an email address that I don't subscribe to the ISP anymore (paid to keep the email). How the heck am I supposed to send mail through them? I'll get a relaying denied error. I know that there are SMTP authorization methods, but most mail servers don't use them. Until people from any network (that are allowed to use From: email@mailserver.com) can send a mail through your mail server, then that mail server should not use SPF.
Why would they need to be huge? Just a single key in the DNS -- have your public key be signed by their private key (with which their public key is in DNS), and include the encrypted public key in your message. When the recipient receives your message, first they go to dns, downloads yahoo's public key, decrypts your public key, then uses it to verify that the email signature is correct. A side benefit is that this can be cached and not require so many lookups. A downside is that mail may not be as anonymous anymore since your public key may be associated with you.
If you have to contact yahoo to add your key to the dns, then why not instead just contact yahoo to sign your public key? Either way requires contact.
In my opinion, the best way would be to require authorization on SMTP (or a new standard) servers, and then just sign the emails on the server. This would require no change in software on the user's end, and isn't too much more computationally intensive than say using SSL to check your mail.
Their site says that they can't do something for free for the candidates, so they have to charge. The money that Amazon earns by this will be donated to a charity.
I know at least for half-life on steam, that only the core game downloads first, so you may have to wait for the first 20-30 megs before you can play. However, the levels and media are downloaded on demand (download level 2 and all textures that aren't in the cache while you're playing level 1, and so on)
Perhaps I'm wrong, but aren't flourescent bulbs usually sealed glass with a gas inside, and applying current through the bulb is what makes it light up? I would've thought that you using your dremel on it would break the bulb and let all the gas out. Perhaps I'm wrong though. Anyone with some insight?
Am I the only person that ever kicks boxes that are on floors before picking them up? I guess you wouldn't want to work with me then.
It's not like they don't have a year from the time the movie goes from script to being produced, to being in theatres to being shipped. The studios can produce the DVD's before the movie starts playing in theatres. The only reason they don't is because more money is made at theatres.
Alternatively, a much more reliable method is using one of the serial cable controls if your box supports it at all. I bought a $5 DB-9 RJ-11(?) converter for my RCA DirecTV tuner, and it was just a matter of setting the tuner up to use the cable instead of "Home Control" mode, and a perl script to open /dev/ttyS0 (search the mythtv-users mailing list for RCA.pl if you have this one)
most people use basic stamps for their little robots. It can't run a "decent" OS, but I believe they have a module that allows them to receive radio data.
LindowsOS Windows
Correct me if I'm wrong, but the configuration app in SuSe lets you do this (YaST2 I think). It starts a temporary X server to test out the resolution to make sure it works, otherwise it will return you back in 30 seconds. If you apply the changes it restarts X somehow saving your open programs and getting the new resolution.
donated, and it was a joke. hence the smiley.
Our own Punica-device is powered by a 40MHz Intel i486 CPU with 16MiB RAM
:)
Main costs:
* computer was sponsored
I sure hope no company went out and bought these guys such a powerful computer in exchange for their name on it.. It might have run them out of business
Maybe they didn't make clean. I sometimes do this and get a false sense of security in how fast my computer when upgrading apps, but only to be sorely disappointed at the 3 hours stuff takes to compile when I clean it out. Maybe someone can buy me one of these things :)
Except that the primary/secondary notation is already used for the cable arrangment. Currently it's the:
...
Primary Master
Primary Slave
Seconndary Master
Secondary Slave
Tertiary Master
So we should be calling them the Primary Primary drive?
My dad thought this would be a great Star Trek ad. A Klingon and Federation member drive and pick up a Romulan, then after a while they decide the Romulan sticks and throw him out.
"If someone came to us and said, `We want Elvis to sing this new song,' we'd have a lot to contemplate," he said. "We tried to retain the integrity of his original song with the remixes. Now you're talking about a whole new vocal performance of a song he never sang or knew? How do we know he'd want to sing it?"
Why not just ask for a synthetic version of an Elvis impersonator to sing the song? That should sound pretty close.
PHPEdit (phpedit.net) uses a similar method to develop their software. Basically their software is free, and if you want features you donate and they will implement them after reaching a certain commitment level based on the number of hours it will take to add the feature. The work is done by their team presumably, so it's not really for open source projects, just an innovative way to "sell" the product.
Ah, but the problem with this is the fact that different move orders result in the same board configuration. Take for instance: [1. e4 e5 2. d4] vs [1. d4 e5 2. e4]. The single-computer would already know that the first one was score +0.005, so then it wouldn't recalculate the score to be .005 for the second one since it knows they are the same board position. Transmitting all the scores for all the moves would be pointless since that would take too much bandwidth to do.
Admittedly the programmers have to understand chess thoroughly, but Kasparov could wipe the floor with the lot of them
This is not necessarily true. A successful checkers player was built with the creator not knowing much at all about checkers. See One Jump Ahead for more info.
I believe the algorithm traditionally used for chess programs is not good for such incredible parallelism like distributed computing would provide. Distributed computing is best for independant modules that can be solved seperately (i.e. see if any of the numbers 1232412000000000000-1232413000000000000 divide 5454934934923283184138231231231253849348096454). The min-max algorithm (i think that's what it's called) requires information about the scoring of following points on the graph in order to determine how well the current score is going to do. It seems almost counter-intuitive that so many computers doing work couldn't make something more effective, but it's true.
First off--Where at, then, in the IPv4 packet header, do you suggest putting the "differentiator"? Oh, shucks, I guess there isn't much space left in the header. (I'm interpreting you literally, here.)
Maybe the evil bit can be used.. Sure, maybe Russia and China may have it set and some routers might mistake their packets as evil, but they're Communist anyway so the routers shouldn't care!
While you may be correct about IRV having problems and Condorcet method being better, I don't understand why nobody suggests (or at least nobody that I have seen at least) that voters use the IRV method to vote, while the tabulators use the Condorcet method to tally the winners. The ranking system used by IRV is just a logical subset of the possible choices for condorcet.. and I doubt that people will want to sway from configurations other than the ones built by using rankings.
A simple algorithm for converting rankings to condorcet voting would be:
bool table[CANDIDATES][CANDIDATES];
for (int i=0; iCANDIDATES; i++)
for (int j=0; jCANDIDATES; j++)
table[i][j] = rank[i]rank[j];