Ha, I wondered if you were talking about MyRealBox. They're pretty cool. If they go paid tomorrow, I'll simply drop my address. I only use it as a throwaway address mostly, and subscribe to a few mailing lists with it. Now that I've discovered mailinator.com, which is also awesome, I don't have as much of a need for a throwaway address anyway. IMHO emails from free services should always be treated as temporary. I don't give people I know my MyRealBox address. For people I care about communicating with I use my college address, and I expect to move to an ISP address after I graduate. Why don't you use your ISP-provided address for important stuff?
You just don't grasp that the standards are going to have to change. SPF is already on its way to becoming an Internet standard. Having to pay a tiny amount for the privelege of using a mail account at a different ISP is not the end of the world, and it doesn't "cripple" anything.
I wouldn't mind DJB's idea either, but if some big ISPs got frustrated with the slow progress of Internet standards and joined together to replace SMTP, I doubt they would go with DJB. Instead they would invent a new closed protocol that would give them total control over the email system and shut out the little guy entirely. Forget running your own mail server, or having free email accounts. Frankly, I'm surprised it hasn't happened already.
I still fail to see a huge problem here. If you don't want to pay, use an account on your ISP's servers. If you really can't stand using your ISP's address for some reason, then maybe you'll have to pay (Gasp! Paying for something you want? On the Internet? never!). If you're unlucky enough to be on a pay-per-byte plan, that is. However, amount of bandwidth taken by email is very small, and bandwidth is cheap. I doubt very much that email would be a significant percentage of your monthly bill unless you do nothing else on the Internet at all. SPF is the kind of change that the Internet's email architecture needs to see if it is going to survive this spam crisis. If we insist that every part of SMTP works just the way it does now for the rest of eternity, we will drown in the deluge of spam. Either SMTP will change or it will get replaced by a protocol that works better, and if you think this minor change is bad I can guarantee you won't like SMTP's replacement.
In Windows XP at least, not all drivers require a reboot when installed. This page lists various circumstances and whether or not they require a reboot. I'm pretty sure starting Linux could be done without a reboot, even if it required installing some sort of driver. That page also mentions that "extending" an NTFS partition can be done without a reboot, but it doesn't mention shrinking. More investigation would be needed to find out.
Better real-time realistic physics for games (including on-the-fly synthesis of human-like motions for characters), better AI for game characters, real-time video processing/editing, even better real-time sound synthesis, faster compiling, higher-level languages for more productive programmers, voice synthesis and recognition, robot motion control (walking), natural language processing, next-generation smarter desktop interfaces, useful computer vision, AI (as in, a computer that passes the turing test).
There are still plenty of cool applications for general-purpose processors that require more speed. Then once these things can be done by themselves, you'll want to do them all at once. And there are always the crazy applications that nobody's come up with yet.
To me, any setup where mail is sent from B's mailserver with an address from A is not a fairly simple setup. This is not something your average person does; only people who know a lot about email know that this is even possible.
In your situation, migrating from A to B seems to be the best solution. I can't say I sympathize with you, however. You seem to be circumventing A's restrictions in a roundabout way. I'm surprised they even offer a forwarding service for free, and I'm sure they only do it because most people don't know they can have a setup like yours. If A stopped offering that service, you'd be out of luck regardless of SPF. A has the right to charge for POP/SMTP, and you don't have to use their service. There are plenty of free email places still out there (I use myrealbox.com).
I think it has to do with the history of SMTP being unsecured and unauthenticated. A lot of people (and ISPs) are stuck in old ways of thinking about securing SMTP. Either Mindspring needs to get educated and allow you to connect how you like, or you need to migrate away from them so you can stop paying for their sub-par service.
Yes, I still say it is no big deal. A 50 MB attatchment is extremely rare, and vacations to far-away countries where you email people 50 MB attachments are even rarer. Even in this worst-case scenario it will only take a minute or three longer to transfer to the US than to a local mailserver (assuming you have broadband, otherwise your local connection will be the bottleneck anyway).
Using a local mailserver is a pointless optimization, adding needless complexity and vulnerability to the email system. Globally, the extra resources used would be negligable. Actually, since most people either don't bother or don't know how to configure their mail client to do what you describe, everyone *already* sends all their mail through their ISP's servers. It hasn't been a tremendous problem so far.
If you want to transfer 50 MB, and you just can't stand the thought of wasting a little precious bandwidth, then you can use another transfer method. Most service providers won't allow 50 MB emails anyway. Use an instant messaging program to transfer it directly, or set up an http server and host it yourself. If your ISP doesn't allow you to do that, that's much worse than requiring you to use their mail servers.
Why would you use your ISP's mail server? You should use AOL's mail servers to send mail from an AOL account. Using your ISP's mail server is a retarded way to do it, and the fact that it is possible now is the *whole problem* that SPF was designed to solve.
I think the promise of this approach may strech beyond people trying out Linux to helping people install Linux alongside Windows. I can imagine a new Linux install process that doesn't require booting from a disk or CD. Instead you download a giant executable which starts a coLinux system. Once it is running it can cooperate with part of the installer still running as a Windows process to resize the main Windows partition, create a Linux partition, and install Linux there. (I think it's possible in Windows to resize an NTFS partition online, correct me if I'm wrong though...) It could also take network and other hardware settings directly from the running Windows installation. After it's done, it could simply install its own bootloader, and then reinitialize all the devices and take over the system. That would be really cool.
You're right; you can't send email from your AOL address without being logged into AOL, but it's not because of SPF. AOL's email system is, and has always been, proprietary. You can't use any Outlook or mutt or any other email client with AOL. That's just because AOL sucks. But SPF records don't prevent you from using your ISP's mail servers to send mail from anywhere, as long as you authenticate with them. The only thing SPF prevents you from doing is setting up your own mail server to send mail claiming to be from your ISP. So what is the problem exactly?
You don't claim copyright on the information, you claim copyright on the collection as an entity. If someone else collects data from sources other than your database, they cannot possibly infringe on your copyright. So this won't work.
The whole trick is the bit "~r&c" at the end. In my discrete math class we learned that would generate a sirpinski triangle if r is the row number and c is the column number in the triangle. I don't really remember why it does now. The rest of the code is basically just a wrapper for that, producing nice formatting.
Another Slashdotter was able to reduce it to 74 characters, but I can't seem to find his post anymore. Now I've been able to reduce it to 75 characters myself, and if you have any ideas on how to make it smaller I'd love to hear them.
Thanks! That code there is actually longer than it needs to be because the lameness filter rejects the short version. One other slashdotter actually reduced it to 74 characters through clever use of ?:, but I can't seem to find his post now. I've recreated it here from vague memories but can't seem to get it lower than 75 characters.
it woul dhave the benefit of using google's database to eliminate things relating to porn and france, not just things that have those specific keywords.
This is intelligence, but it has nothing to do with being an agent. We're talking about intelligent agents here, not improved search algorithms. People often get distracted by the term intelligent and ascribe all these cool capabilities to "intelligent agents." They ignore the fact that making something an agent doesn't automatically make it more intelligent. You still have to design an algorithm to make it intelligent. And once you do that, Google can use it too and serve you the results instantly, making an agent redundant.
The problem with intelligent agents is they take too much time. People like instant results, so we need something faster. Also, Google's algorithms can work better than any intelligent agent because Google has total knowledge of the entire web. An agent can only look at a tiny subset of the web as it surfs for you.
instead of having to make up new combinations of terms to find new stuff your agent learns what you likes and finds things for you using related keywords.
A program to suggest better search terms based on past searches isn't an intelligent agent. An intelligent agent is a software robot carrying out a task for you and reporting the results. Suggesting better search terms is already done at Google on a limited basis (spelling). Suggesting future terms based on past ones is a moderately good idea, but it's not a job for an intelligent agent. Suggesting sites based on sites you liked is already done by the "Similar Pages" feature Google has. No intelligent agents required.
Windows is stable for people who know what they're doing. For the majority of comptuer users, Windows is still very unstable. Yes, the core OS has gotten more stable, but the amount of crap that is running on a typical Windows system has increased tenfold. Once you have Comet Cursor, SaveNow, Gator, New.net, Yahoo toolbar, Alexa toolbar, Realplayer/Quicktime/Winamp system tray applets, and 10 different auto-updater programs all sticking their tentacles into the bowels of Windows Explorer, Windows becomes just as unstable as it ever was in the days of 98, if not more. Plus you get flashing advertisements popping up out of nowhere every five minutes and whenever you start a program. I'm not exaggerating about the number of programs either; there are an incredible number of crapware programs out there preying on innocent users who don't know any better. Even when you buy a new PC now it usually comes with at least 5-10 unnecessary programs that start automatically and run all the time.
The core OS doesn't crash hard very much any more, but the Explorer shell is still full of holes and you can knock IE over with a feather once all that crap is installed.
MS is eventually going to have to do something, because it's not getting any better. The line between virus and legitimate program is getting blurred. Maybe this trend will push people to Linux. If companies tried to make spyware for KDE/GNOME, the open-source community would come out with a spyware remover program within three days, and it would be installed in the next release of every major distro.
No, I don't think that's useful. The push model of information delivery on the web failed for a good reason. People like to get their information when they want it, not when someone else decides to send it to them. You won't save any time by having an intelligent agent do a google search for you, because the search takes 1.2 seconds. The long part is sifting through the results, and an intelligent agent isn't going to do any better at that than Google, so you'll still end up doing it. Intelligent agents won't magically be able to sort sites more relevantly than Google can.
Of course Google could be more intelligent, but my point is that the concept of an "intelligent agent" is just dumb. Google should of course work on improving their search algorithms, which could mean adding a bayesian filter or checking for bad taste in colors or whatever. That's separate from the idea of a personal "intelligent agent."
What about K3B, Quanta+, eric3, and scribus? There are tons more great KDE apps at the new KDE-Apps.org. The future of KDE application development looks bright. Remember that you're comparing KDE apps against the complete set of all other open-source applications. I think KDE is doing pretty well, myself.
You mean like KDE-Cygwin, or QKW? I guess in the future if these get mature enough you could replace Windows Explorer with Konqueror/KDesktop/Kicker...
The idea of an agent that crawls the net searching for stuff for you is kinda dumb. Google has already done the crawling, and can serve you the results instantly, whenever you want. The biggest problem is *expressing what you're searching for* in terms that a computer can understand. Without a solution to that problem, an intelligent agent can't be any better than a Google search. If that problem is solved, then Google can still serve you results instantly, without any "intelligent agent" crawling the web specifically for you. The intelligence is all in Google's algorithms, and there's no need for any agents.
To me, the whole idea of intelligent agents sounds too much like Clippy. I don't want software giving me suggestions and telling me what I would like. OTOH, software presenting a list of information that might be useful is OK. It's kind of a psychological thing. Amazon.com doesn't have an "intelligent agent" that tells you what products you would like; instead it has a page with a list of things that are similar or related to products you've shopped for. The end result is the same, and the difference is subtle, but I think it's an important psychological one. The computer shouldn't display intelligence and boss you around; instead it should act like a mechanical device that simply responds to input that you give it. Intelligent agents don't allow you to actually do anything you can't do with passive, subservient software. They're just more obnoxious and annoying.
They say a lot of that stuff to men too. One time a Best Buy salesidiot told me that I should buy the gold-plated USB cable because it would give me "better quality printing." They also tried to press tons of useless accessories onto us when we bought our last computer. The vast majority don't really know anything about computers, they just spout the spiel given to them by their managers, the same to every customer. You know, of course, that the only difference between used-car salesmen and computer salesmen is that the used-car salesmen know when they're lying to you.
Slashdot Mirror
Ha, I wondered if you were talking about MyRealBox. They're pretty cool. If they go paid tomorrow, I'll simply drop my address. I only use it as a throwaway address mostly, and subscribe to a few mailing lists with it. Now that I've discovered mailinator.com, which is also awesome, I don't have as much of a need for a throwaway address anyway. IMHO emails from free services should always be treated as temporary. I don't give people I know my MyRealBox address. For people I care about communicating with I use my college address, and I expect to move to an ISP address after I graduate. Why don't you use your ISP-provided address for important stuff?
I wouldn't mind DJB's idea either, but if some big ISPs got frustrated with the slow progress of Internet standards and joined together to replace SMTP, I doubt they would go with DJB. Instead they would invent a new closed protocol that would give them total control over the email system and shut out the little guy entirely. Forget running your own mail server, or having free email accounts. Frankly, I'm surprised it hasn't happened already.
Speed isn't the only issue in AI, of course, but I think more speed will be needed when the solutions are finally found.
I still fail to see a huge problem here. If you don't want to pay, use an account on your ISP's servers. If you really can't stand using your ISP's address for some reason, then maybe you'll have to pay (Gasp! Paying for something you want? On the Internet? never!). If you're unlucky enough to be on a pay-per-byte plan, that is. However, amount of bandwidth taken by email is very small, and bandwidth is cheap. I doubt very much that email would be a significant percentage of your monthly bill unless you do nothing else on the Internet at all. SPF is the kind of change that the Internet's email architecture needs to see if it is going to survive this spam crisis. If we insist that every part of SMTP works just the way it does now for the rest of eternity, we will drown in the deluge of spam. Either SMTP will change or it will get replaced by a protocol that works better, and if you think this minor change is bad I can guarantee you won't like SMTP's replacement.
In Windows XP at least, not all drivers require a reboot when installed. This page lists various circumstances and whether or not they require a reboot. I'm pretty sure starting Linux could be done without a reboot, even if it required installing some sort of driver. That page also mentions that "extending" an NTFS partition can be done without a reboot, but it doesn't mention shrinking. More investigation would be needed to find out.
There are still plenty of cool applications for general-purpose processors that require more speed. Then once these things can be done by themselves, you'll want to do them all at once. And there are always the crazy applications that nobody's come up with yet.
In your situation, migrating from A to B seems to be the best solution. I can't say I sympathize with you, however. You seem to be circumventing A's restrictions in a roundabout way. I'm surprised they even offer a forwarding service for free, and I'm sure they only do it because most people don't know they can have a setup like yours. If A stopped offering that service, you'd be out of luck regardless of SPF. A has the right to charge for POP/SMTP, and you don't have to use their service. There are plenty of free email places still out there (I use myrealbox.com).
I think it has to do with the history of SMTP being unsecured and unauthenticated. A lot of people (and ISPs) are stuck in old ways of thinking about securing SMTP. Either Mindspring needs to get educated and allow you to connect how you like, or you need to migrate away from them so you can stop paying for their sub-par service.
Using a local mailserver is a pointless optimization, adding needless complexity and vulnerability to the email system. Globally, the extra resources used would be negligable. Actually, since most people either don't bother or don't know how to configure their mail client to do what you describe, everyone *already* sends all their mail through their ISP's servers. It hasn't been a tremendous problem so far.
If you want to transfer 50 MB, and you just can't stand the thought of wasting a little precious bandwidth, then you can use another transfer method. Most service providers won't allow 50 MB emails anyway. Use an instant messaging program to transfer it directly, or set up an http server and host it yourself. If your ISP doesn't allow you to do that, that's much worse than requiring you to use their mail servers.
Why would you use your ISP's mail server? You should use AOL's mail servers to send mail from an AOL account. Using your ISP's mail server is a retarded way to do it, and the fact that it is possible now is the *whole problem* that SPF was designed to solve.
I think the promise of this approach may strech beyond people trying out Linux to helping people install Linux alongside Windows. I can imagine a new Linux install process that doesn't require booting from a disk or CD. Instead you download a giant executable which starts a coLinux system. Once it is running it can cooperate with part of the installer still running as a Windows process to resize the main Windows partition, create a Linux partition, and install Linux there. (I think it's possible in Windows to resize an NTFS partition online, correct me if I'm wrong though...) It could also take network and other hardware settings directly from the running Windows installation. After it's done, it could simply install its own bootloader, and then reinitialize all the devices and take over the system. That would be really cool.
You're right; you can't send email from your AOL address without being logged into AOL, but it's not because of SPF. AOL's email system is, and has always been, proprietary. You can't use any Outlook or mutt or any other email client with AOL. That's just because AOL sucks. But SPF records don't prevent you from using your ISP's mail servers to send mail from anywhere, as long as you authenticate with them. The only thing SPF prevents you from doing is setting up your own mail server to send mail claiming to be from your ISP. So what is the problem exactly?
You don't claim copyright on the information, you claim copyright on the collection as an entity. If someone else collects data from sources other than your database, they cannot possibly infringe on your copyright. So this won't work.
Another Slashdotter was able to reduce it to 74 characters, but I can't seem to find his post anymore. Now I've been able to reduce it to 75 characters myself, and if you have any ideas on how to make it smaller I'd love to hear them.
for(int r=-1,c=0;r<39;c++)printf(c<0?" ":c>r?c=r++-38,"\n":~r&c?" `":" #");
Unfortunately the smaller version can't be posted as a sig due to the *$#*& lameness filter.
for(int r=-1,c=0;r<39;c++)printf(c<0?" ":c>r?c=r++-38,"\n":~r&c?" `":" #");
If you can get it any smaller, that would be cool. I've run out of ideas for making it smaller, but that doesn't mean it's as small as it can get.
This is intelligence, but it has nothing to do with being an agent. We're talking about intelligent agents here, not improved search algorithms. People often get distracted by the term intelligent and ascribe all these cool capabilities to "intelligent agents." They ignore the fact that making something an agent doesn't automatically make it more intelligent. You still have to design an algorithm to make it intelligent. And once you do that, Google can use it too and serve you the results instantly, making an agent redundant.
The problem with intelligent agents is they take too much time. People like instant results, so we need something faster. Also, Google's algorithms can work better than any intelligent agent because Google has total knowledge of the entire web. An agent can only look at a tiny subset of the web as it surfs for you.
instead of having to make up new combinations of terms to find new stuff your agent learns what you likes and finds things for you using related keywords.
A program to suggest better search terms based on past searches isn't an intelligent agent. An intelligent agent is a software robot carrying out a task for you and reporting the results. Suggesting better search terms is already done at Google on a limited basis (spelling). Suggesting future terms based on past ones is a moderately good idea, but it's not a job for an intelligent agent. Suggesting sites based on sites you liked is already done by the "Similar Pages" feature Google has. No intelligent agents required.
MS is eventually going to have to do something, because it's not getting any better. The line between virus and legitimate program is getting blurred. Maybe this trend will push people to Linux. If companies tried to make spyware for KDE/GNOME, the open-source community would come out with a spyware remover program within three days, and it would be installed in the next release of every major distro.
People should be able to go after the *clients* of Fax.com. That would put them out of business REAL fast.
No, I don't think that's useful. The push model of information delivery on the web failed for a good reason. People like to get their information when they want it, not when someone else decides to send it to them. You won't save any time by having an intelligent agent do a google search for you, because the search takes 1.2 seconds. The long part is sifting through the results, and an intelligent agent isn't going to do any better at that than Google, so you'll still end up doing it. Intelligent agents won't magically be able to sort sites more relevantly than Google can.
Of course Google could be more intelligent, but my point is that the concept of an "intelligent agent" is just dumb. Google should of course work on improving their search algorithms, which could mean adding a bayesian filter or checking for bad taste in colors or whatever. That's separate from the idea of a personal "intelligent agent."
What about K3B, Quanta+, eric3, and scribus? There are tons more great KDE apps at the new KDE-Apps.org. The future of KDE application development looks bright. Remember that you're comparing KDE apps against the complete set of all other open-source applications. I think KDE is doing pretty well, myself.
KDE 3.2 Feature Plan
To me, the whole idea of intelligent agents sounds too much like Clippy. I don't want software giving me suggestions and telling me what I would like. OTOH, software presenting a list of information that might be useful is OK. It's kind of a psychological thing. Amazon.com doesn't have an "intelligent agent" that tells you what products you would like; instead it has a page with a list of things that are similar or related to products you've shopped for. The end result is the same, and the difference is subtle, but I think it's an important psychological one. The computer shouldn't display intelligence and boss you around; instead it should act like a mechanical device that simply responds to input that you give it. Intelligent agents don't allow you to actually do anything you can't do with passive, subservient software. They're just more obnoxious and annoying.
They say a lot of that stuff to men too. One time a Best Buy salesidiot told me that I should buy the gold-plated USB cable because it would give me "better quality printing." They also tried to press tons of useless accessories onto us when we bought our last computer. The vast majority don't really know anything about computers, they just spout the spiel given to them by their managers, the same to every customer. You know, of course, that the only difference between used-car salesmen and computer salesmen is that the used-car salesmen know when they're lying to you.