Does advice that crosses the TLDR threshold score well with CBR but poorly with WABR? From TFA, [brackets added]:
> (if you make your advice hard to follow [read], that reduces the chance of somebody actually climbing that mountain [reading it] >and then pointing out to you if your suggestion didn't work). So it's not just that the advice-giver is being unhelpful, it's that they're being a dick.
what is the TLDR threshold anyway? I'd love to see a quantification of the amount of information that can fit inside it
My phone is about to catch on fire! It is actually uncomfortably hot to the touch running the twister server -- some sort of CPU usage regulation would be nice. Love the idea.
btw the interface just popped up in my gmail account, so I tried calling my cell phone, and the caller ID on my cell phone showed my Google Voice number, so perhaps these services aren't so distinct. Maybe a gmail user who doesn't have a Google Voice account could chime in.
but they can't support SIP - a protocol specifically designed to handle these kinds of situations
Can't, don't, or don't yet? With google voice, they're two trivial steps away from letting wifi capable phones call and receive calls anywhere in the US (and probably elsewhere) for free, with no SIM card installed, even - 1) set up a SIP gateway and allow (android/iOS,etc. or computer based) softphones to connect it to it using Google's existing authentication bindery, 2) allow Google Voice users to direct calls to the gateway (and 3), I guess, create user interfaces (Android, etc.) for making calls that are more convenient than their website tool). I don't see a business model there, but it would be really cool
You can string this stuff together with GV and other existing services, so I know it works:)
... a AMD Athlon XP 2800+, 1GB RAM running Ubuntu 8.10 back then... Haven't gotten any news since).
-- you actually managed to give away equipment without getting tech support calls about it every week for the next 5 years? Please provide more details.
There's a long-running thread on the bbs for spamgourmet discussing a bunch of events like this -- spamgourmet users generally use a unique email address for each of their accounts, and so can quickly identify a problem (unless it was with spamgourmet itself, of course, but records so far show that hasn't happened). The response of the companies varies from complete denial and reticence to surprising accountability. None of it ever ended up in court, afaik.
One thing I've tried recently is to require some information that is contextually relevant, but not obvious from the information surrounding the challenge (which is not captcha, just a form input). For instance, on my blog, I'm requiring that a comment poster supply the name of the blog (which is in bold letters at the top of the page). For real posters, this is no doubt annoying, but the name of the blog is somewhere near the top of the stack in their brains. For a spammer, who's racing through a bunch of blogs to post comment spam, this likely is completely out-of-band. So far (about 3 months) it has completely stopped comment spam. Of course, I don't have info on how many real posters have clicked away from the page in frustration, but I have continued to get real comments at about the same rate as before.
If this sounds like a good idea, do something else, so that there's no pattern:)
What you're suggesting may be trivial to implement on a few machines, but it's anything but trivial to implement on any meaningful percentage of the world's mail servers (and clients!). Additionally, as mentioned above, a great deal of spam does currently come from zombies, and it actually is trivial to configure a zombie to send spam to one recipient per message.
Spamgourmet ( http://www.spamgourmet.com/ ) uses sendmail for the mail server and is handling 1,741,490 disposable email addresses as I write, on tinker toy hardware.
It's a pain in the ass to configure, etc., etc., but it works very well and probably has more flight time than any other choice.
For me, it's primarily about availability of the device - I hate carrying around stuff. I have a treo 600 that's taken over a great deal of my PC tasks. I never started reading e-books until I got it. I have two different reader programs installed -- palmreader (for "secure" files) and tibr (which I prefer, but it doesn't handle the "secure" files). The treo 650 has better resolution and so would be be easier on the eyes, but I haven't got one yet.
Anyway, I've done so much reading with the treo (wherever I find myself waiting I can read) and gotten used to - the convenience of not turning pages - having my place saved automatically in multiple books - having a backlight so that ambient lighting is irrelevant - having all the books I've read on the device present for reference and electronic searches - etc. - that I have a hard time picking up a paper book now.
Seriously, choose the right tool for the job. When we've got a sys-admin level scripting task, and someone can go in and knock it out in a half hour (or less) with a few lines of Perl, who can say that's bad? I'm currently wading through a bunch of heavily patternized java that pulls checkin logs from a scm system and updates an issue tracking system as part of the build process. It's taken me *days* to begin to "grok" what's going on in the many associated xml config files and bizarre string handling approaches that were used in this undocumented hack. I'll replace it with probably less than 150 lines of Perl, and someone else will happily (and much more easily) maintain it. So there!
At the same time, we've got > 25 developers distributed around the world working on a big commodities trading app -- java works pretty well for that.
think of it as a system for machine to machine communication (or process to process communication). Two major topologies are 1) "publish/subscribe", where a message is broadcast to a particular subject or topic, and zero to many processes are listening for it (all of them that are listening will receive the message) and 2) "point to point", where a process sends a message to a queue and (hopefully) one or more processes are checking for it, but only the first one to check will get the message. In either case, the publishing process can continue to go about its business after sending the message, without worrying about whether it was received (or it can wait for acknowledgment if need be). There are many subtle variations on this.
One common use for messaging is to distribute load accross many servers.
At least some of the complaints are misguided. We run the anti-spam disposable email address spamgourmet.com out of HE. Some enthusiastic sg users regularly report spam that comes to unexpired disposable email addresses, and the reports often wrongly implicate the spamgourmet server (which, of course, sits in HE's IP range) as being involved.
Ironically, one of the most time-consuming tasks involved with running sg is dealing with these false reports.
but would they sign a www.bankofslushdot.com cert or a www.wollsforgo.com cert? If so, the man in the middle can put a redirect from the main website address to the typo-like address, then present a valid cert -- it would be up to the user to catch the subtle change in the displayed URL.
The man in the middle can set up an http proxy, route all IP addresses to it, and present their own certificate. Sure, it won't be from the bank, and it may be difficult to come up with one that's signed by a trusted CA, but the little lock in the browser status bar will look the same... Then, they can happily decrypt all transmissions both ways and see what's going back and forth.
On a project I worked on last year, we used a traffic light kind of thing that somebody picked up from a novelty store. QA would light the green light if testing was going well, they'd light the yellow light if there were performance problems or if the defect rate was climbing, and they'd light the red light if the test system crashed.
Slashdot Mirror
... ... ...
But why male models?
Does advice that crosses the TLDR threshold score well with CBR but poorly with WABR? From TFA, [brackets added]:
> (if you make your advice hard to follow [read], that reduces the chance of somebody actually climbing that mountain [reading it]
>and then pointing out to you if your suggestion didn't work). So it's not just that the advice-giver is being unhelpful, it's that they're being a dick.
what is the TLDR threshold anyway? I'd love to see a quantification of the amount of information that can fit inside it
My phone is about to catch on fire! It is actually uncomfortably hot to the touch running the twister server -- some sort of CPU usage regulation would be nice. Love the idea.
Yup - that was me - 13 years old, with a stack of graph paper, a bunch of pencils, and a bag of funny shaped dice. 100% pure gangsta!
btw the interface just popped up in my gmail account, so I tried calling my cell phone, and the caller ID on my cell phone showed my Google Voice number, so perhaps these services aren't so distinct. Maybe a gmail user who doesn't have a Google Voice account could chime in.
but they can't support SIP - a protocol specifically designed to handle these kinds of situations
Can't, don't, or don't yet? With google voice, they're two trivial steps away from letting wifi capable phones call and receive calls anywhere in the US (and probably elsewhere) for free, with no SIM card installed, even - 1) set up a SIP gateway and allow (android/iOS,etc. or computer based) softphones to connect it to it using Google's existing authentication bindery, 2) allow Google Voice users to direct calls to the gateway (and 3), I guess, create user interfaces (Android, etc.) for making calls that are more convenient than their website tool). I don't see a business model there, but it would be really cool
You can string this stuff together with GV and other existing services, so I know it works :)
... a AMD Athlon XP 2800+, 1GB RAM running Ubuntu 8.10 back then... Haven't gotten any news since).
-- you actually managed to give away equipment without getting tech support calls about it every week for the next 5 years? Please provide more details.
what's the impact on my power bill of a bunch of these little things?
There's a long-running thread on the bbs for spamgourmet discussing a bunch of events like this -- spamgourmet users generally use a unique email address for each of their accounts, and so can quickly identify a problem (unless it was with spamgourmet itself, of course, but records so far show that hasn't happened). The response of the companies varies from complete denial and reticence to surprising accountability. None of it ever ended up in court, afaik.
One thing I've tried recently is to require some information that is contextually relevant, but not obvious from the information surrounding the challenge (which is not captcha, just a form input). For instance, on my blog, I'm requiring that a comment poster supply the name of the blog (which is in bold letters at the top of the page). For real posters, this is no doubt annoying, but the name of the blog is somewhere near the top of the stack in their brains. For a spammer, who's racing through a bunch of blogs to post comment spam, this likely is completely out-of-band. So far (about 3 months) it has completely stopped comment spam. Of course, I don't have info on how many real posters have clicked away from the page in frustration, but I have continued to get real comments at about the same rate as before.
:)
If this sounds like a good idea, do something else, so that there's no pattern
heh, yeah. After a few years of repeatedly trying the URL to no avail, you give up, you move on.
BTW - read the "news" section from front to back.
check out:
o ldmanmurray.com/
r ay.com
http://web.archive.org/web/20020528011342/http://
or if that doesn't work, pick a link from:
http://web.archive.org/web/*/http://www.oldmanmur
you never have to give out your address: spamgourmet.com
What you're suggesting may be trivial to implement on a few machines, but it's anything but trivial to implement on any meaningful percentage of the world's mail servers (and clients!). Additionally, as mentioned above, a great deal of spam does currently come from zombies, and it actually is trivial to configure a zombie to send spam to one recipient per message.
you forgot to
use Post::Troll;
I don't think your post will pass taint checking.
yeah, yeah, sendmail.
Spamgourmet ( http://www.spamgourmet.com/ ) uses sendmail for the mail server and is handling 1,741,490 disposable email addresses as I write, on tinker toy hardware.
It's a pain in the ass to configure, etc., etc., but it works very well and probably has more flight time than any other choice.
down in Houston, all they got is steers, queers, and engineers, and you don't look much like a steer or a queer to me, boy -- heh heh .
For me, it's primarily about availability of the device - I hate carrying around stuff. I have a treo 600 that's taken over a great deal of my PC tasks. I never started reading e-books until I got it. I have two different reader programs installed -- palmreader (for "secure" files) and tibr (which I prefer, but it doesn't handle the "secure" files). The treo 650 has better resolution and so would be be easier on the eyes, but I haven't got one yet.
Anyway, I've done so much reading with the treo (wherever I find myself waiting I can read) and gotten used to - the convenience of not turning pages - having my place saved automatically in multiple books - having a backlight so that ambient lighting is irrelevant - having all the books I've read on the device present for reference and electronic searches - etc. - that I have a hard time picking up a paper book now.
People kill readability (with Perl).
Seriously, choose the right tool for the job. When we've got a sys-admin level scripting task, and someone can go in and knock it out in a half hour (or less) with a few lines of Perl, who can say that's bad? I'm currently wading through a bunch of heavily patternized java that pulls checkin logs from a scm system and updates an issue tracking system as part of the build process. It's taken me *days* to begin to "grok" what's going on in the many associated xml config files and bizarre string handling approaches that were used in this undocumented hack. I'll replace it with probably less than 150 lines of Perl, and someone else will happily (and much more easily) maintain it. So there!
At the same time, we've got > 25 developers distributed around the world working on a big commodities trading app -- java works pretty well for that.
think of it as a system for machine to machine communication (or process to process communication). Two major topologies are 1) "publish/subscribe", where a message is broadcast to a particular subject or topic, and zero to many processes are listening for it (all of them that are listening will receive the message) and 2) "point to point", where a process sends a message to a queue and (hopefully) one or more processes are checking for it, but only the first one to check will get the message. In either case, the publishing process can continue to go about its business after sending the message, without worrying about whether it was received (or it can wait for acknowledgment if need be). There are many subtle variations on this.
One common use for messaging is to distribute load accross many servers.
At least some of the complaints are misguided. We run the anti-spam disposable email address spamgourmet.com out of HE. Some enthusiastic sg users regularly report spam that comes to unexpired disposable email addresses, and the reports often wrongly implicate the spamgourmet server (which, of course, sits in HE's IP range) as being involved.
Ironically, one of the most time-consuming tasks involved with running sg is dealing with these false reports.
but would they sign a www.bankofslushdot.com cert or a www.wollsforgo.com cert? If so, the man in the middle can put a redirect from the main website address to the typo-like address, then present a valid cert -- it would be up to the user to catch the subtle change in the displayed URL.
The man in the middle can set up an http proxy, route all IP addresses to it, and present their own certificate. Sure, it won't be from the bank, and it may be difficult to come up with one that's signed by a trusted CA, but the little lock in the browser status bar will look the same... Then, they can happily decrypt all transmissions both ways and see what's going back and forth.
Don't call it a "BitTorrent Arrest" -- some of my best friends use BitTorrent for perfectly legitimate reasons... It's really an arrest for piracy.
On a project I worked on last year, we used a traffic light kind of thing that somebody picked up from a novelty store. QA would light the green light if testing was going well, they'd light the yellow light if there were performance problems or if the defect rate was climbing, and they'd light the red light if the test system crashed.