The lameness filters are really getting on my nerves. More than half of my comments over the last few weeks hit some filter or another before I managed to get the comment through.
Every time I screw up in a comment -- trigger some other lameness filter, forget to put a subject, anything -- I get a "it hasn't been 20 seconds since you hit reply".
I like to read the entire article before posting to make sure I'm not putting up something redundant with an existing score:3 post, but if I use that strategy and then end up deciding to post two comments, I invariably get "it's been less than two minutes since your last comment post."
I got a cryptic message about "postersubj compression" when I tried to reply to this comment. Changing the subject of my reply made the error message go away, but it took me 5 minutes to figure out how to get aroudn the filter (mostly because of the 20 second rule).
Others have complained about shell scripts being rejected as "junk posts", and having comments rejected as having been posted before slashdot existed. Some of these are just Bugs, but many seem to be intentional. Does the lameness filter serve any purpose that moderation doesn't?
IMO, there should be two filters:
No duplicate posts. Not so much to filter out "lameness", but to help people who accidentally hit "Submit" twice or hit it again when slashdot forgets to respond to the first post.
No more than five comments from one poster within two minutes.
fake e-mail address vs auto spamproofing
on
Welcome to Slashdot 2.2
·
· Score: 3, Interesting
There are several new user options for comment display. I'll let you poke around them yourself and see how they work. Some you'll like, some you won't. You'll notice that the old 'Fake Email Address' is now gone. We decided that it simply was to easy to abuse, so instead we offer auto spam proofing of your real email address. Of course this is purely optional for you paranoid spam-free types. This is Jamie's, as was the new domain linking stuff, and countless other usability features.
How was the fake e-mail address field "abused"? And how can automatic spam-proofing possibly be good enough that a clever spammer wouldn't be able to get around (especially with the source code visible)?
Is anyone else getting spam with the body "hey. i got the file thanx."? The message doesn't seem to have an attachment, which makes it even more disconcerting.
IIRC, The DMCA makes it illegal to break encryption used as part of a copy-protection system.
But if it were illegal to circumvent Freenet's encryption for filtering purposes, and if it were also illegal to passively allow people to transfer copyrighted data through your computer, those two laws wouldn't be contradictory -- instead, they would together make it illegal to run a Freenet node.
While the article says that Freenet is safe under current US lays, it doesn't address a large potential problem: Congress could pass a law making "anonymous, encrypted, peer-to-peer networks" illegal on the grounds that the only reason you would choose to use that kind of P2P network would be to trade illegal material (especially since Freenet tends to be slower than a P2P network that doesn't involve encryption).
I think we should be thinking about ways to shift public opinion in favor of Freenet, so that such a law doesn't get passed, instead of trying to work around current laws. One thing that might help would be if official videos (movie trailers, convention broadcasts) were officially distributed through Freenet, saving the content providers money on bandwidth. That doesn't help to argue that anonymity and encryption are important, however. Can Freenet be defended, or is it truly only useful for trading kiddie porn and bootleg music?
Mozilla has a similar blurb on their releases page. But they don't give you the option of downloading binaries without crypto, and you don't even see the blurb if you opt to download a nightly build instead of a release build. I can't imagine someone in an embargoed country downloading and building from source just to avoid breaking a US law.
The strange thing is that the default build options don't include crypto. I found this out when I copied someone else's build script, which included the BUILD_PSM2 (build with crypto) option, and then tried to build Mozilla. It turned out that I had to install an extra program in order to make the PSM2 build work. (Why don't they link to the how to build Mozilla with crpyto page from the how to build Mozilla on Windows page?)
Akamai's network can help to smooth out huge fluctuations in traffic. A further benefit is that the customer's web server does not have to deliver the heavy items, which reduces the load on it dramatically and makes it less likely to collapse when faced with a sudden surge of visitors.
In other wordrs, Akamai has almost finished developing a complete defense against our strongest tactical weapon, the Slashdot effect. They must be stopped at all costs.
The difference in the cases you cite is that you are already dead before we do research with your corpse. The child is already dead before we do research with its corpse. We have to kill the embryo before we can do research on its corpse.
Many stem cell lines come from the extra embryos created during fertility treatments. Those embryos probably would have been killed anyway.
Here's another alternative: let an embryonic cell divide once, and then separate the two cells from each other. Take one of the cells for stem cell research, and then do whatever you were originally planning to do with the other one (implant it in the woman, etc). Would that be considered ok, or would it be considered cloning followed by murder?
Second, those bands which do invariably have vastly better organized websites and ftp sites dedicated exclusively to that pursuit.
Outgoing bandwidth costs money.
I am a DMB fan, and I would far prefer to go to from the dedicated ftp/www sites, where I can download entire/full/non-corrupt albums, than a disorganized system like napster, where anything I searched for (any time during its existence) would result largely in his COMMERCIAL recordings.
It sounds like the best solution would be for the DMB site to post freenet keys for non-commerical recordings.
Kent Thuresson, the creator of the 6.0 Orbit theme, has released post-6.0 versions at http://www.ninelineman.com/orbit/. It looks like he had it working in builds as recent as Feb 15, 2001, which was around Mozilla 0.8.
Re:How much deeper does this hole get?
on
Netscape 6.1
·
· Score: 1
The more Netscape keeps releasing beta code, the more users it's alienating. I know it's tough not having the latest and greatest 5.x (err 6.x) browser to market, but come on. By the time we get to 6.2 (i.e. Mozilla 1.0 stable), there will be five Netscape users left.
"Mozilla 1.0" is more likely to be an API freeze than a "this is no longer a beta-quality browser" release. Netscape's version numbers are more reflective of the state of the product: 6.0 was new features and major rewrites, like most x.0 releases, and 6.1 is an improved version with more emphasis on stability and not many new features since 6.0.
GeoCities doesn't allow you to link to images hosted on GeoCities from other sites. To view the image, click on the link and then add a ? to the end of the URL. (If you're using Mozilla, pressing enter in the location bar is sufficient.)
The real solution would be to create a web page in geocities with an tag pointing to the image, and then to link to that page instead of the image. I'm not sure if the page would have to be in the same geocities account, or if it just has to be on geocities.
Mozilla is way too slow and the file save dialogs take a second or two to disappear. Sucks when you're trying to grab images fast and spank it at the same time!
I'm glad I'm not the only person who ran into this problem. Please vote for bug 66723, "Download window should not appear when saving from cache". The bug is currently marked as INVALID, but it's likely to be reconsidered if several people vote for it, especially since fixing the bug would be matching IE's behavior.
Re:great features, too late
on
Netscape 6.1
·
· Score: 1
Mozilla also loads and displays pages slightly faster on Windows than it does on X. You can see weekly graphs on the netscape.public.mozilla.performance newsgroup. I don't know if this is because more effort is put into optimization on Windows or if it's because of inefficiencies in X, because I just read netscape.public.mozilla.performance for the pictures.
Data formats (eg txt, doc, html, pdf) often embed files or scripts. The problem is when the script isn't sandboxed well and so it can do something you didn't think it could do: eg html scripts can read the contents of other html files in the same directory if they know the name of the html files, and doc macros / pdf embeds used to be able to do anything the user could do.
But a hand-made index is better at finding a subject than searching the text; it doesn't skip synonmys and stop at words in the wrong context. Any decent technical book will hopefully have a handmade index that beats using grep.
Good point. I can usually find what I'm looking for in non-fiction; it's fiction books that I find hard to search. They're hard to search because (a) they often don't have indices, and (b) they're more likely to use the kind of language such that I'll remember a passage by some rare word it happens to contain.
I don't mind reading a book in dead tree format, but I'm much more comfortable reading a book that I can get an online copy of for one reason: computers are much better at O(n) tasks like scanning a large amount of text for a specific word than humans are. Dead-tree publishers should be happy each time they see a recent book show up on Freenet, because it means there's one more book I can purchase and read without fear of not being able to find a passage I want to quote/rememeber/reread.
Re:It is only Medium DAMAGE!
on
Code Redux
·
· Score: 2
I agree. <imo>Anti-virus software companies are in the business of protecting against viruses; of preventing a large number of users from being compromised by the same code. They are not interested in the kind of security that would prevent script kiddies or social engineers from gaining access to your computer, and so they rate viruses by the amount of damage they cause, rather than rating security holes by the amount of damage they allow. I suppose they do this to be consistent with their stance that "the viruses are the enemy".</imo>
By the way, did anyone else think it was strange that CERT listed anti-virus software companies, and only anti-virus software companies, in the "vendor information" section of their advisory about SirCam? They could have easily targeted
E-mail client vendors, for having poor user interface surrounding attachments. (Especially Microsoft, for releasing at least one version of OE that shows a very similar dialog when you double-click a.jpg attachment as it does when you double-click a.exe attachment.)
Microsoft, for relying on extensions as the only way for a user to tell the difference between a document and a program, rather than doing one or more of the following:
Giving users and programs a way to flag files as "executable" (or as "not executable"), like linux does with the +x mode.
Using a single, special extension for executable files. For example, foo.vbs would have to be renamed to foo.vbs.exe before it would run.
Using a special type of icon, or icon overlay, to indicate that something is a document. For example, always show documents as a piece of paper, and show an icon chosen by the associated application in the middle of the paper.
Microsoft, for not providing a function in Windows for "is a file with extension.foo a document or a program?".
Slashdot Mirror
The lameness filters are really getting on my nerves. More than half of my comments over the last few weeks hit some filter or another before I managed to get the comment through.
Others have complained about shell scripts being rejected as "junk posts", and having comments rejected as having been posted before slashdot existed. Some of these are just Bugs, but many seem to be intentional. Does the lameness filter serve any purpose that moderation doesn't?
IMO, there should be two filters:
There are several new user options for comment display. I'll let you poke around them yourself and see how they work. Some you'll like, some you won't. You'll notice that the old 'Fake Email Address' is now gone. We decided that it simply was to easy to abuse, so instead we offer auto spam proofing of your real email address. Of course this is purely optional for you paranoid spam-free types. This is Jamie's, as was the new domain linking stuff, and countless other usability features.
How was the fake e-mail address field "abused"? And how can automatic spam-proofing possibly be good enough that a clever spammer wouldn't be able to get around (especially with the source code visible)?
I had to change the subject line of my reply before slashdot would let it through. It kept complaining about "postersubj compression". What the fuck?
Would the book have a rabbit on the cover?
Loki isn't on fuckedcompany.com's front page yet. Let's slashdot their submission queue!
Is anyone else getting spam with the body "hey. i got the file thanx."? The message doesn't seem to have an attachment, which makes it even more disconcerting.
IIRC, The DMCA makes it illegal to break encryption used as part of a copy-protection system.
But if it were illegal to circumvent Freenet's encryption for filtering purposes, and if it were also illegal to passively allow people to transfer copyrighted data through your computer, those two laws wouldn't be contradictory -- instead, they would together make it illegal to run a Freenet node.
While the article says that Freenet is safe under current US lays, it doesn't address a large potential problem: Congress could pass a law making "anonymous, encrypted, peer-to-peer networks" illegal on the grounds that the only reason you would choose to use that kind of P2P network would be to trade illegal material (especially since Freenet tends to be slower than a P2P network that doesn't involve encryption).
I think we should be thinking about ways to shift public opinion in favor of Freenet, so that such a law doesn't get passed, instead of trying to work around current laws. One thing that might help would be if official videos (movie trailers, convention broadcasts) were officially distributed through Freenet, saving the content providers money on bandwidth. That doesn't help to argue that anonymity and encryption are important, however. Can Freenet be defended, or is it truly only useful for trading kiddie porn and bootleg music?
Mozilla has a similar blurb on their releases page. But they don't give you the option of downloading binaries without crypto, and you don't even see the blurb if you opt to download a nightly build instead of a release build. I can't imagine someone in an embargoed country downloading and building from source just to avoid breaking a US law.
The strange thing is that the default build options don't include crypto. I found this out when I copied someone else's build script, which included the BUILD_PSM2 (build with crypto) option, and then tried to build Mozilla. It turned out that I had to install an extra program in order to make the PSM2 build work. (Why don't they link to the how to build Mozilla with crpyto page from the how to build Mozilla on Windows page?)
Akamai's network can help to smooth out huge fluctuations in traffic. A further benefit is that the customer's web server does not have to deliver the heavy items, which reduces the load on it dramatically and makes it less likely to collapse when faced with a sudden surge of visitors.
In other wordrs, Akamai has almost finished developing a complete defense against our strongest tactical weapon, the Slashdot effect. They must be stopped at all costs.
Your post worked fine without the tags. Good post, by the way.
The difference in the cases you cite is that you are already dead before we do research with your corpse. The child is already dead before we do research with its corpse. We have to kill the embryo before we can do research on its corpse.
Many stem cell lines come from the extra embryos created during fertility treatments. Those embryos probably would have been killed anyway.
Here's another alternative: let an embryonic cell divide once, and then separate the two cells from each other. Take one of the cells for stem cell research, and then do whatever you were originally planning to do with the other one (implant it in the woman, etc). Would that be considered ok, or would it be considered cloning followed by murder?
Does that mean Konq doesn't send referrers across servers?
Second, those bands which do invariably have vastly better organized websites and ftp sites dedicated exclusively to that pursuit.
Outgoing bandwidth costs money.
I am a DMB fan, and I would far prefer to go to from the dedicated ftp/www sites, where I can download entire/full/non-corrupt albums, than a disorganized system like napster, where anything I searched for (any time during its existence) would result largely in his COMMERCIAL recordings.
It sounds like the best solution would be for the DMB site to post freenet keys for non-commerical recordings.
Kent Thuresson, the creator of the 6.0 Orbit theme, has released post-6.0 versions at http://www.ninelineman.com/orbit/. It looks like he had it working in builds as recent as Feb 15, 2001, which was around Mozilla 0.8.
The more Netscape keeps releasing beta code, the more users it's alienating. I know it's tough not having the latest and greatest 5.x (err 6.x) browser to market, but come on. By the time we get to 6.2 (i.e. Mozilla 1.0 stable), there will be five Netscape users left.
"Mozilla 1.0" is more likely to be an API freeze than a "this is no longer a beta-quality browser" release. Netscape's version numbers are more reflective of the state of the product: 6.0 was new features and major rewrites, like most x.0 releases, and 6.1 is an improved version with more emphasis on stability and not many new features since 6.0.
screenshot of Mozilla w/Aqua theme
GeoCities doesn't allow you to link to images hosted on GeoCities from other sites. To view the image, click on the link and then add a ? to the end of the URL. (If you're using Mozilla, pressing enter in the location bar is sufficient.)
The real solution would be to create a web page in geocities with an tag pointing to the image, and then to link to that page instead of the image. I'm not sure if the page would have to be in the same geocities account, or if it just has to be on geocities.
Mozilla is way too slow and the file save dialogs take a second or two to disappear. Sucks when you're trying to grab images fast and spank it at the same time!
I'm glad I'm not the only person who ran into this problem. Please vote for bug 66723, "Download window should not appear when saving from cache". The bug is currently marked as INVALID, but it's likely to be reconsidered if several people vote for it, especially since fixing the bug would be matching IE's behavior.
Mozilla also loads and displays pages slightly faster on Windows than it does on X. You can see weekly graphs on the netscape.public.mozilla.performance newsgroup. I don't know if this is because more effort is put into optimization on Windows or if it's because of inefficiencies in X, because I just read netscape.public.mozilla.performance for the pictures.
lol
Do outlook users still have the option of having mail autoexecuted upon downloading? (the "preview feature") Still broken, IMO.
Is that kind of like having a web page "autoexecute" when you view it?
Fix the security hole, not the fact that you can view a message by left-clicking on it once.
Data formats (eg txt, doc, html, pdf) often embed files or scripts. The problem is when the script isn't sandboxed well and so it can do something you didn't think it could do: eg html scripts can read the contents of other html files in the same directory if they know the name of the html files, and doc macros / pdf embeds used to be able to do anything the user could do.
But a hand-made index is better at finding a subject than searching the text; it doesn't skip synonmys and stop at words in the wrong context. Any decent technical book will hopefully have a handmade index that beats using grep.
Good point. I can usually find what I'm looking for in non-fiction; it's fiction books that I find hard to search. They're hard to search because (a) they often don't have indices, and (b) they're more likely to use the kind of language such that I'll remember a passage by some rare word it happens to contain.
I don't mind reading a book in dead tree format, but I'm much more comfortable reading a book that I can get an online copy of for one reason: computers are much better at O(n) tasks like scanning a large amount of text for a specific word than humans are. Dead-tree publishers should be happy each time they see a recent book show up on Freenet, because it means there's one more book I can purchase and read without fear of not being able to find a passage I want to quote/rememeber/reread.
(*grumble* Mozilla find in page O(n^2) algorithm *grumble*)
I agree. <imo>Anti-virus software companies are in the business of protecting against viruses; of preventing a large number of users from being compromised by the same code. They are not interested in the kind of security that would prevent script kiddies or social engineers from gaining access to your computer, and so they rate viruses by the amount of damage they cause, rather than rating security holes by the amount of damage they allow. I suppose they do this to be consistent with their stance that "the viruses are the enemy".</imo>
By the way, did anyone else think it was strange that CERT listed anti-virus software companies, and only anti-virus software companies, in the "vendor information" section of their advisory about SirCam? They could have easily targeted