Someone (the author or some editor) added that comma to my sentence. My original email had no comma there. A clearer phrasing that would not tempt someone into adding punctuation would be:
[The least effective technique is] Any technique that tries to identify "good" mail with neither authentication backing it up nor some form of personalized training.
They also removed the name of the company where I work (IronPort), which struck me as a bit odd considering how my job allows me to do open source was part of the article. I think my employer deserves some kudos for that. Not to mention implying that I'm more than just one of the developers. There are eight commiters, six of them on the Project Management Committee and two of them (Justin Mason and Theo Van Dinter) write at least as much code as me.
I don't know about DSPAM either, but the paper
does note about CRM-114:
In an effort to ensure that we had not misinterpreted the installation instructions, we ran CRM-114 in a train-on-everything configuration and, as predicted by the author, the result was substantially worse.
and earlier:
As with CRM114, we trained DSPAM only on misclassifications, as suggested in the documentation.
CRM-114 and DSPAM exhibit substantially inferior performance to the other
filters, regardless of threshold setting. Both exhibit substantial learning through
outthe email stream, leading us to conjecture that their performance might asymptotically approach that of the other filters. From a practical standpoint, this learning
rate would be too slow for personal email filtering as it would take several years atthe observed rate to achieve the same misclassification rates as the other systems.
This is interesting considering the harsh words the DSPAM author directs towards SpamAssassin in the DSPAM FAQ. In contrast, I think, the SpamAssassin developers say they are interested in testing the "dobly" noise reduction technique that DSPAM employs, see
SpamAssassin bug 3078.
I'm surprised he doesn't talk about orbital stability (around a star) as a potentially useful criteria. Maybe this only seems useful to me because I'm not a professional astronomer, but if an object has a significant chance of being captured as a moon or flung out of the solar system (from another object in the solar system), I don't think it should be called a planet.
Perhaps he didn't mention it because all objects meeting his "gravity rules" requirement happen to have stable orbits.
All These Worlds Are Yours Except Europa. Attempt No Landing There. Use Them Together. Use Them in Peace.
Why bother pretending this story post is news?
on
Melting Europa
·
· Score: 5, Insightful
"After having contaminated Earth's Oceans"
"But I wonder how long the time lag will be between the probe finding life, and a leak in the radioactive heater wiping all of it out. What next? Drill Sedna for oil?"
I wish the Slashdot editors could maintain at least the pretense of objectivity in which stories they post. I'm sure someone else submitted the story without the loaded commentary. I mean, even the sexing-up BBC managed to write a decent article about this.
If not that, perhaps it would be helpful for less frequent readers if editors disclosed their obvious biases: Green Party member, voting for ABB, never tires of SCO stories, Microsoft-hater, whatever.
Another option would be sub-sites for News for [insert political bent]-leaning nerds, stuff that confirms your beliefs.
that almost nobody is really taking this seriously, so the lack of interest in space defense seems about right to me. The human species has survived 2 million years without going the way of the dinosaur. It seems like there are many reasons to not stress out about this:
Low risk/reward ratio, public money is much better spent elsewhere. If someone else wants to spend their money on this, more power to them.
Our technology is very rapidly advancing, especially relative to the amount of time that passes (on average) between significant asteroid hits. 100 years ago we were completely helpless. 50 years ago, we had nukes, but no missles that were even close to being able to deliver them, in another 50 or 100 years, this may be a yawner due to general technology advances.
To be completely flippant (and yes, I do realize there is a risk, I just think it is relatively low)... boring! I just hope this doesn't turn into another cause where misguided celebrities drive us into spending money on it disproportionally like certain trendy diseases.
give up one digit or four?
on
The Universal Card
·
· Score: 4, Insightful
and is unlocked with the user's fingerprint
I don't know about you, but I'd much rather have it use a password. I think most people would happily give a sufficiently threatening criminal their 4 digit PIN number (or any style of password) without too much of a fuss, but I'd rather avoid giving anyone any incentive whatsoever to leave me short one digit. It would be a very small consolation to cancel my credit cards after such an incident.
I'm surprised you're upset about share-alike and not at all annoyed by the non-commercial restrictions on the movies. Given the large number of activities that can be interpreted as commercial, it seems like the far more restrictive attribute, one not even found in the GPL.
I think the true genius of the GPL is that it helped move hordes of developers who were formerly in the "non-commercial shareware" camp into the far more productive and revolutionary "share-alike" camp (with no non-commercial attribute). That is, I believe a large number of developers who were not and would not be satisfied with a BSD-style license fell back on simplistic "non-commercial use" licenses and often did not share source code. The GPL gave them an outlet which at the end of the day is a very fair compromise and improves the commons much more.
Finally, the thing that makes GNU, Linux, and other GPL projects (along with other open source licenses, to be sure) competitive with entirely proprietary projects is that commercial use is not restricted. When commercial use is disallowed, then companies (and their employees who often work on open source in their free time so they can apply it at work) have no incentive to invest time and energy in a project.
Since 2 of the 3 winning images are provided as an Apple QuickTime movie file, does that mean there is now an unencumbered open source QuickTime player?
Did I miss something?
Re:Easier way to lower the electricity bill
on
DIY HVAC
·
· Score: 4, Funny
Put the real thermostat somewhere hidden and place a dummy one in the hall for the wife and kids.
Yeah, because divorce is always cheaper than paying higher electrical bills, right?
The Slashdot story is perhaps interesting rhetoric and umm... advocacy, but when I read the story, the thing that jumped out at me is that the questions are generally vague and there are very few specific questions about the substance of the Act. It falls prey to the typical media script of "Some critics say [insult character of interview subject]" or "It has been written that [insert oversweeping charge]". Honestly, it seems like Viet Dinh's replies are very matter of fact and he answers most of the questions as best and as openly as can be expected given the questions. If he does not come out seeming like a fascist, it is either because the interviewer did a poor job, because he's not a fascist, or both. I'm sure that won't stop people from making their own conclusions based on little or no evidence, though. (Oh my god, he gave detailed answers!)
I'd be very interested in someone asking Viet Dinh substantive questions about specific concerns raised in the Patriot Act, but I'm unable to draw much of any conclusion from reading this article, especially not the same alarmist conclusion that the story submitter has drawn.
Another interpretation I could make, especially based on the story submitter's comments, is that the critics of the Patriot Act are equally incapable of discussing the ramifications of the Act as are its supporters. Unfortunately, it's the job of the critics to do a good job criticizing and they get far too hung up in rhetoric and name-calling to take most of them very seriously and given that the law is now on the books, I think they're going to need to change their tactics if they want to have any substantive effect.
Oh crap, I seem to be falling prey to the standard media script of analyzing process rather than issues.
On the author's mail (where all he does is
probably talk about CRM114 and probably does not
subscribe to many newsletters or non-technical mailing lists).
A pre-trained filter. It can't be compared apples-to-apples with any filter that doesn't require training.
Using his own filter on his own mail! Of course it does well.
... to mention a few of the problems. The statistics and methodology behind these claims are really questionable. I think both Consumer Reports and PC Magazine have both done better evaluations of spam filters (read that however you want).
Also, I wonder how many people have actually looked at CRM114 and tried to use it.
The really interesting thing about CRM114 is the windowed polynomial hashing technique used although there's some evidence that it can work just as well (if not better) on a much smaller window of only two tokens. I'm hoping someone will do a full exploration of the idea for SpamAssassin's Bayes module.
This is the deal with Coppola's wines--they're just ok wines for better-than-ok prices.
Actually, the Niebaum-Coppola Rosso California Red Table Wine (pretty consistent from year to year so far), is my favorite inexpensive red wine. It's much better than most $25 wines and I've found it as cheap as $7 or $8 a bottle.
If you joined the Imperial Forces, chances are you did it because of a slick recruiting plan and promises of good wages / adventure -- hey, Luke was about to join up before Obi-Wan came along, wasn't he?
Sheesh. Luke was going to join the Rebel Alliance like his friend Biggs had.
Daniel
I've done it a few times, it's easy. I decided in the end that I like coffee too much to give it up for good, though.
Basically, just stop drinking caffeinated beverages (cold turkey). If you're tired, just take a nap. If you get a headache, take an ibuprofen or something (not Excedrin since it contains caffeine), following the directions on the bottle, of course.
Also drink fluids and make sure you're eating. It never hurts to be well-hydrated.
For a moment there, I thought the power grid was going to fail or perhaps airplanes would be unable to land safely. No, it's just that retailers can't order liquor from the state government.
I think a better question, although perhaps off-topic for Slashdot, is why the hell a lot of states hold this much control over alcohol distribution and keep it centralized like this (using some Stalinist organizational theory that centralized government control is always the best way to handle any sector of the economy).
Yeah, I know when I'm trying to summarize performance, I always include 1871 different numbers in a set of tables. The boss loves that. Readers love that.
Sorry, I looked at the benchmark results and quickly reached total eye glaze over. I think it
could be presented a lot more clearly.
I'll be honest. I do not want any data recording devices in my car because the information gathered by the box could be held against me in any number of ways. Maybe it can improve car safety over time, but with vehicle laws and civil lawsuits being the way they are, I don't want the equivalent of the permanent wire-tap on my driving. If the data is there and any legal situation comes up where it could be useful, it will be used and if these devices are regulated into cars, you won't have any choice about it (obstruction of justice, destroying evidence, anti-tampering laws, etc.).
There might be some number of times where the devices could be used to prove your innocence or lack of liability, but I'd rather take my chances without the devices. I mean, how often does anyone really drive the speed limit on the highway?
Of course, my 2002 probably already has something of the sort and I'm probably just ignorant about it. Anyway, I think car safety can be improved over time almost as well without the boxes and the adjoining less of privacy.
I think I first heard about this concept from Richard Gooch who, not surprisingly, wrote an implementation based on the util-linux init implementation. See Linux Boot Scripts, Richard Gooch for more information.
Midori Linux also has parallel boot scripts which really helps speed up the boot process on the devices that use it. The design is very nice (and based on Richard Gooch's) and doesn't rely on "make". It's named quickinit (it's just a.tar.gz despite the.mlz extension).
I'm not sure using "make" is the best way to implement this although it is an interesting idea. "init" can manage dependencies without too much additional code. The Gooch/Midori systems are nice because dependencies are implemented as simple commands in the init scripts. You just add a "need <service>" command and it will block until the service is available. If I recall correctly, quickinit basically managed a full dependency tree whereas the Gooch version maintained a simple list (which works fine until you want to do something more complicated like restarting X windows or handle resume/suspend using the init system). It's too bad no major distributions have picked up this idea.
Of course, rebooting needs to be first stable and second fast. It's not like we want to reboot all that often.:-)
It makes me hope the other giants will follow suit, and wonder if the music industry is finally listening to some of the consumer's complaints.
So, if theft is a protest, what's next?
Books are too expensive or otherwise worthy of protest, so people will start scanning them and putting up copies for others on the net.
Prescription drugs are too expensive, so we'll import cloned drugs from certain countries or underground labs.
Software licenses are not to our liking, so we'll copy software and put it under a license of our own choosing (note that this can cut against open source as well).
Cars are too expensive (and pollute!), so let's go steal us a nice one.
I don't understand where some of the current thinking about music theft came from. Artists have other choices to get their music heard, so most must believe that they have something to gain from big music companies. If you don't like it, don't buy it or create your own.
If you're stealing it, you obviously do like it or want it, so what exactly makes music downloaders anything other than thieves? Slashdot posters often lament consumerism and greed. Wanting more music than you can afford and resorting to theft seems pretty high on the greed scale to me.
Regarding the article, placing the lowering of CD prices in the category of "victory" presumes that copying music is a moral activity. Yeah, right...
A lot of people seem to be really missing the point of this idea and seem to be making what I would call the "waaah! you can't be a real perl hacker if you like this idea" response.
First, a number of Perl packages already ship in RPM format and are distributed with Red Hat and other RPM-based distributions. This will integrate much more nicely and if package names are chosen well, it will avoid conflicts between CPAN-installed packages and RPM-installed packages. I myself use Debian and always use the Debian package instead of CPAN if it's available. It's easier, it's faster, and (in the case of Debian) you get various fixes that might not have made it into the upstream.
Once you're using some packages for Perl stuff, you might as well try to get everything in the distribution package format. In addition, I think the more appropriate question is why it's taken so long for someone to think of and then implement this idea. Why on earth would you want to use more than one package management system on your box, especially when CPAN is just for Perl packages. Using CPAN when RPM packages are available might even be silly (provided the packages are well-constructed, I don't know that they will be).
Plus, the truth is, CPAN is cool and all, but it's not all that great. The usage is bizarre and if an average user isn't a Perl person, then why should they need to learn perl to install some perl-based package that depends on some CPAN modules? CPAN also is just plain inferior to RPM and.deb when it comes to installation, deinstallation, configuration mangagement, upgrades, and just about everything else.
Now, I just hope they decide to do the same for Debian packages. (Debian provides quite a few already, but more is better.)
Seems like this fool was trying to go after EBay by filing patents that were VERY similar to what EBay had already been doing. Nevermind "Buy it Now", he wanted it all.
Fool? The guy just won a $29.5 MILLION DOLLAR verdict. That's much better than most scams and it's completely legal, so to speak.
Maybe I'm too competitive, but I would have made sure a few really skinny/short people were available and eligible to be drivers on my
team. I weigh 152 pounds and I'm 6 feet tall and I weigh more than a fair number of the skinny geek set. How hard can it be to find a few 115 pound women (they may need to be part of the engineering team, but that too should be no problem, especially in mechanical/electrical engineering which has less of a male-female imbalance than computer engineering) to do the race-time driving? 176 pounds is a fairly mid-sized guy.
I tried downloading the rules to check, but the PDF was a blank document. Does anyone know if 176 pounds is just that team's average or are teams with lighter drivers required to use weights to bring them up to 176 pounds (like some horse races)?
Open source software is generally written by much more than one person. Would the winner have to split her winnings with hundreds of others, or would the award go to whoever led the project?
A lot of the larger open source projects have non-profit corporations behind them, so any rewards could go there. Apache, Mozilla, Free Standards Group, GNU, SPI (Debian's corporation), and quite a few more.
I'd prefer if such a large reward could go somewhere with financial oversight rather than a founder. If it's a one-man project, though, I'd be happy to see it go into his pocket.
Slashdot Mirror
Someone (the author or some editor) added that comma to my sentence. My original email had no comma there. A clearer phrasing that would not tempt someone into adding punctuation would be:
They also removed the name of the company where I work (IronPort), which struck me as a bit odd considering how my job allows me to do open source was part of the article. I think my employer deserves some kudos for that. Not to mention implying that I'm more than just one of the developers. There are eight commiters, six of them on the Project Management Committee and two of them (Justin Mason and Theo Van Dinter) write at least as much code as me.
This is interesting considering the harsh words the DSPAM author directs towards SpamAssassin in the DSPAM FAQ. In contrast, I think, the SpamAssassin developers say they are interested in testing the "dobly" noise reduction technique that DSPAM employs, see SpamAssassin bug 3078.
Perhaps he didn't mention it because all objects meeting his "gravity rules" requirement happen to have stable orbits.
All These Worlds Are Yours Except Europa. Attempt No Landing There. Use Them Together. Use Them in Peace.
"After having contaminated Earth's Oceans"
"But I wonder how long the time lag will be between the probe finding life, and a leak in the radioactive heater wiping all of it out. What next? Drill Sedna for oil?"
I wish the Slashdot editors could maintain at least the pretense of objectivity in which stories they post. I'm sure someone else submitted the story without the loaded commentary. I mean, even the sexing-up BBC managed to write a decent article about this.
If not that, perhaps it would be helpful for less frequent readers if editors disclosed their obvious biases: Green Party member, voting for ABB, never tires of SCO stories, Microsoft-hater, whatever.
Another option would be sub-sites for News for [insert political bent]-leaning nerds, stuff that confirms your beliefs.
To be completely flippant (and yes, I do realize there is a risk, I just think it is relatively low) ... boring! I just hope this doesn't turn into another cause where misguided celebrities drive us into spending money on it disproportionally like certain trendy diseases.
I don't know about you, but I'd much rather have it use a password. I think most people would happily give a sufficiently threatening criminal their 4 digit PIN number (or any style of password) without too much of a fuss, but I'd rather avoid giving anyone any incentive whatsoever to leave me short one digit. It would be a very small consolation to cancel my credit cards after such an incident.
I think the true genius of the GPL is that it helped move hordes of developers who were formerly in the "non-commercial shareware" camp into the far more productive and revolutionary "share-alike" camp (with no non-commercial attribute). That is, I believe a large number of developers who were not and would not be satisfied with a BSD-style license fell back on simplistic "non-commercial use" licenses and often did not share source code. The GPL gave them an outlet which at the end of the day is a very fair compromise and improves the commons much more.
Finally, the thing that makes GNU, Linux, and other GPL projects (along with other open source licenses, to be sure) competitive with entirely proprietary projects is that commercial use is not restricted. When commercial use is disallowed, then companies (and their employees who often work on open source in their free time so they can apply it at work) have no incentive to invest time and energy in a project.
Did I miss something?
Yeah, because divorce is always cheaper than paying higher electrical bills, right?
I'd be very interested in someone asking Viet Dinh substantive questions about specific concerns raised in the Patriot Act, but I'm unable to draw much of any conclusion from reading this article, especially not the same alarmist conclusion that the story submitter has drawn.
Another interpretation I could make, especially based on the story submitter's comments, is that the critics of the Patriot Act are equally incapable of discussing the ramifications of the Act as are its supporters. Unfortunately, it's the job of the critics to do a good job criticizing and they get far too hung up in rhetoric and name-calling to take most of them very seriously and given that the law is now on the books, I think they're going to need to change their tactics if they want to have any substantive effect.
Oh crap, I seem to be falling prey to the standard media script of analyzing process rather than issues.
Also, I wonder how many people have actually looked at CRM114 and tried to use it.
The really interesting thing about CRM114 is the windowed polynomial hashing technique used although there's some evidence that it can work just as well (if not better) on a much smaller window of only two tokens. I'm hoping someone will do a full exploration of the idea for SpamAssassin's Bayes module.
This is the deal with Coppola's wines--they're just ok wines for better-than-ok prices.
Actually, the Niebaum-Coppola Rosso California Red Table Wine (pretty consistent from year to year so far), is my favorite inexpensive red wine. It's much better than most $25 wines and I've found it as cheap as $7 or $8 a bottle.
If you joined the Imperial Forces, chances are you did it because of a slick recruiting plan and promises of good wages / adventure -- hey, Luke was about to join up before Obi-Wan came along, wasn't he? Sheesh. Luke was going to join the Rebel Alliance like his friend Biggs had. Daniel
Basically, just stop drinking caffeinated beverages (cold turkey). If you're tired, just take a nap. If you get a headache, take an ibuprofen or something (not Excedrin since it contains caffeine), following the directions on the bottle, of course.
Also drink fluids and make sure you're eating. It never hurts to be well-hydrated.
I think a better question, although perhaps off-topic for Slashdot, is why the hell a lot of states hold this much control over alcohol distribution and keep it centralized like this (using some Stalinist organizational theory that centralized government control is always the best way to handle any sector of the economy).
Sorry, I looked at the benchmark results and quickly reached total eye glaze over. I think it could be presented a lot more clearly.
There might be some number of times where the devices could be used to prove your innocence or lack of liability, but I'd rather take my chances without the devices. I mean, how often does anyone really drive the speed limit on the highway?
Of course, my 2002 probably already has something of the sort and I'm probably just ignorant about it. Anyway, I think car safety can be improved over time almost as well without the boxes and the adjoining less of privacy.
Midori Linux also has parallel boot scripts which really helps speed up the boot process on the devices that use it. The design is very nice (and based on Richard Gooch's) and doesn't rely on "make". It's named quickinit (it's just a .tar.gz despite the .mlz extension).
I'm not sure using "make" is the best way to implement this although it is an interesting idea. "init" can manage dependencies without too much additional code. The Gooch/Midori systems are nice because dependencies are implemented as simple commands in the init scripts. You just add a "need <service>" command and it will block until the service is available. If I recall correctly, quickinit basically managed a full dependency tree whereas the Gooch version maintained a simple list (which works fine until you want to do something more complicated like restarting X windows or handle resume/suspend using the init system). It's too bad no major distributions have picked up this idea.
Of course, rebooting needs to be first stable and second fast. It's not like we want to reboot all that often. :-)
So, if theft is a protest, what's next?
I don't understand where some of the current thinking about music theft came from. Artists have other choices to get their music heard, so most must believe that they have something to gain from big music companies. If you don't like it, don't buy it or create your own. If you're stealing it, you obviously do like it or want it, so what exactly makes music downloaders anything other than thieves? Slashdot posters often lament consumerism and greed. Wanting more music than you can afford and resorting to theft seems pretty high on the greed scale to me.
Regarding the article, placing the lowering of CD prices in the category of "victory" presumes that copying music is a moral activity. Yeah, right...
First, a number of Perl packages already ship in RPM format and are distributed with Red Hat and other RPM-based distributions. This will integrate much more nicely and if package names are chosen well, it will avoid conflicts between CPAN-installed packages and RPM-installed packages. I myself use Debian and always use the Debian package instead of CPAN if it's available. It's easier, it's faster, and (in the case of Debian) you get various fixes that might not have made it into the upstream.
Once you're using some packages for Perl stuff, you might as well try to get everything in the distribution package format. In addition, I think the more appropriate question is why it's taken so long for someone to think of and then implement this idea. Why on earth would you want to use more than one package management system on your box, especially when CPAN is just for Perl packages. Using CPAN when RPM packages are available might even be silly (provided the packages are well-constructed, I don't know that they will be).
Plus, the truth is, CPAN is cool and all, but it's not all that great. The usage is bizarre and if an average user isn't a Perl person, then why should they need to learn perl to install some perl-based package that depends on some CPAN modules? CPAN also is just plain inferior to RPM and .deb when it comes to installation, deinstallation, configuration mangagement, upgrades, and just about everything else.
Now, I just hope they decide to do the same for Debian packages. (Debian provides quite a few already, but more is better.)
Fool? The guy just won a $29.5 MILLION DOLLAR verdict. That's much better than most scams and it's completely legal, so to speak.
I tried downloading the rules to check, but the PDF was a blank document. Does anyone know if 176 pounds is just that team's average or are teams with lighter drivers required to use weights to bring them up to 176 pounds (like some horse races)?
Daniel
A lot of the larger open source projects have non-profit corporations behind them, so any rewards could go there. Apache, Mozilla, Free Standards Group, GNU, SPI (Debian's corporation), and quite a few more.
I'd prefer if such a large reward could go somewhere with financial oversight rather than a founder. If it's a one-man project, though, I'd be happy to see it go into his pocket.
Daniel