For vim compiled with X support, "* is the register corresponding to the X clipboard. So, "*p dumps the X clipboard into your file in a much more elegant way than i and middle-click.
Some large motions (:123, gg, G, probably others) set the ' mark. So, gg to go to the top of the file (say, add a #include or an import), then '' to go back to where you were. On a somewhat related note, ^I and ^O will traverse the list of places you've jumped to -- handy for finding your way back to where you were before you jumped to five or six different places via a search.
Somebody mentioned . earlier. If you're not using . religiously, you're using vim wrong.
High on my wishlist for vim tricks is a way to put delimiters on either side of a motion that supports . to repeat it. E.g.: say I have a few words and I want to parenthesize each of them. Put the cursor at the start of the first word, [some keys](, lparen and rparen magically appear around the word, move to the next word, hit., parens appear around that word, etc.
alias find="noglob find" # Lets me say "find . -name *.pdf" (etc) without escaping the '*'. I also do this with dpkg-query.
alias topten="(sort | uniq -c | sort -rn | head)" # Print the top ten most common lines in whatever gets piped in
diff local_file.txt =(ssh remotehost cat remote_file.txt) # ssh is like a distributed file system, right? # The =(cmd) syntax is pretty useful in general, for where a command expects a filename and you want to give it input from another command.
file =some-command # =word expands to the result of a $PATH search for word. This command tells me what kind of thing I'm executing... readlink -e =some-command #... unless it says "symlink", in which case the next step is to recursively chase symlinks until I get a real file... file $(readlink -e =some-command) #... *then* I can ask what kind of file it is. $(cmd) is preferable to `cmd` because it nests properly.
Southeast Alaska is a rain forest. Not a tropical rain forest (rather, a temperate one), but a rain forest nonetheless. As it turns out, plant life grows prolifically in rain forests.
On US-491 in southeastern Utah, the mile markers go, 67, 68, 68.99, 70. I thought about stealing 68.99 just for the hell of it. The real WTF (as it were) is that 491 used to be 666, but it got changed for some reason...
If I take a relaxed, polite, and understanding attitude towards someone who's helping me, I generally get better results.
This is so true. I make a particular effort to be polite and undemanding when I'm dealing with secretaries or other administrative staff, because getting on their good side makes things much easier when I have a problem that I need their help with.
All that being said, some people are just plain dicks, and all the politeness in the world won't change them.
No kidding. One of the secretaries here is... let us say, not a people person. The first time I had to deal with her, I put on my usual "Hi (big smile), thank you so much for helping me" routine, and she just sort of looked at me like I was something unpleasant that had been spilled on her desk. I eventually figured out that she was like this all the time with everybody. Oh well. She's retiring next week, for which we're all thankful.
Wow. How is it that an "ex-hacker" who now "specialises in security from the white hat side of the fence" (from the author's bio) can have so little clue about the responsible disclosure debate and the economics of modern vulnerability research? Maybe getting lambasted on Slashdot will be a wake-up call for him to actually do his homework before he spouts off.
Calm down. It's sponsored by Usenix, so the accepted papers will be published online, freely accessible by everyone. Also, academic conferences exist to provide a forum for new research, that is, ideas and results that no one save the authors have seen before; therefore, the workshop attendees will in fact learn from their peers.
Any Nintendo classic songs you'd love to see included on the Wii version?
Korobeiniki (AKA Tetris A-Type music), as performed by Ozma. Oh man that song rocks. I'm also partial to the Zelda theme. And maybe somebody can cut a deal with the Minibosses to use some of their work...
I had a summer internship with the R&D branch of one of those other companies you mentioned. I was tasked with writing an SSL man-in-the-middle platform, so when somebody told the security group, for instance, "Yeah, we're just going to use Anonymous Diffie-Hellman mode," the security group could clearly demonstrate why that was a stupid idea. Just to be clear, this was not a hypothetical situation. I mean really, ADH! "Sweet, I have an encrypted channel to... somebody!" *sigh* You have to jump through special hoops to get OpenSSL to use ADH at all.
This whole thing is old news, though. FX told us all about the fun you can have with printers way back at Defcon 10.
Schneier, Applied Cryptography. Everything you need to know about practical cryptographic primitives and cryptosystems is in the first 5-10 chapters of that book.
Do you leave the lights and television on in your house when you aren't using them?
Firefox does not launch instantaneously. Lightbulbs do not maintain complex state while turned on. -1, Bad Analogy.
Do you complain if the car you're driving runs out of gas after running for dozens of hours?
... I really hope you meant that as a joke. Memory is not a use-once resource. -1, Really Bad Analogy.
Leaving a browser open at all times is a sufficiently common usage pattern that, your glib dismissal aside, the Mozilla developers should at least consider a significant push to find and fix the memory leaks that we're complaining about.
FWIW, Safari is in my experience a little memory-leaky, but nothing like Firefox.
(Hmm, my sig is a little hypocritical this time around...)
The world's most dangerous network: DefCON. Everybody who attends knows it. And yet *still*, people use plaintext authentication to all sorts of services. And then they appear on the Wall of Sheep.
"In the future, all I have to do is let some bacteria into your building; they float through the air conditioning system, land on your keyboard, you can't see them, you don't know they are there."
Neal Stephenson called, he wants his threat model back.
Given a quick skim of the Sandia paper (and a closer read of a couple of the important sections), it looks like these Sandia group got results roughly similar to a portion of the work that Maynor and Cache presented (sorry, PPT) at BlackHat and DefCON. From what I can tell, the techniques used by Maynor and Cache were a little more ad-hoc than the Sandia group's machine-learning approach, but unless they write up their results with more depth than a Powerpoint presentation, doing a strong comparison is sort of hard. I'm amused that the teams presented their results within days of each other, but at rather different venues. (Also, I'm sad that USENIX Security overlapped DefCON this year; I've done both in years past, and they make for a good week+ of security and drinking.)
Mea culpa. Serves me right for taking some random blogger seriously. Still, there's enough "MSM is teh suck!!!1" noise out there that you'll have to pardon me for my confusion.
Oh noes, somebody at Reuters made a mistake! Amazing as this may sound, professional news organizations do issue corrections from time to time. Why am I not defending Wikipedia in the same statement? The charged and misleading language that appeared on Wikipedia was intentionally put up by some random person.
"And journalism has sunk to a new low"? Come down off your high horse, Mr Unger.
In many ways, the whole blog concept has perhaps lowered the barrier to entry for on-line publishing a little *too* far. When anyone can publish anything you want with virtually no effort, then it no longer requires that you be inspired or motivated before your inane ramblings are out there in cyberspace. The media has adopted the trend too, with 'blog' in the context of a news site all too often meaning 'poorly researched and largely content-free "reporting" on sensationalist subject matter.'
Slashdot Mirror
The EU navies have begun convoy operations in the Gulf of Aden:
http://www.lloydslist.com/ll/news/eu-launches-naval-escorts-in-gulf-of-aden/20017577798.htm
For vim compiled with X support, "* is the register corresponding to the X clipboard. So, "*p dumps the X clipboard into your file in a much more elegant way than i and middle-click.
Some large motions (:123, gg, G, probably others) set the ' mark. So, gg to go to the top of the file (say, add a #include or an import), then '' to go back to where you were. On a somewhat related note, ^I and ^O will traverse the list of places you've jumped to -- handy for finding your way back to where you were before you jumped to five or six different places via a search.
Somebody mentioned . earlier. If you're not using . religiously, you're using vim wrong.
High on my wishlist for vim tricks is a way to put delimiters on either side of a motion that supports . to repeat it. E.g.: say I have a few words and I want to parenthesize each of them. Put the cursor at the start of the first word, [some keys](, lparen and rparen magically appear around the word, move to the next word, hit ., parens appear around that word, etc.
alias find="noglob find" # Lets me say "find . -name *.pdf" (etc) without escaping the '*'. I also do this with dpkg-query.
alias topten="(sort | uniq -c | sort -rn | head)" # Print the top ten most common lines in whatever gets piped in
diff local_file.txt =(ssh remotehost cat remote_file.txt) # ssh is like a distributed file system, right?
# The =(cmd) syntax is pretty useful in general, for where a command expects a filename and you want to give it input from another command.
file =some-command # =word expands to the result of a $PATH search for word. This command tells me what kind of thing I'm executing... ... unless it says "symlink", in which case the next step is to recursively chase symlinks until I get a real file... ... *then* I can ask what kind of file it is. $(cmd) is preferable to `cmd` because it nests properly.
readlink -e =some-command #
file $(readlink -e =some-command) #
My brothers-in-arms across the hall whipped up something which might get you started:
http://research.cens.ucla.edu/projects/2007/Systems/DTS/
They use it to manage 100 seismic sensors strung out in a 500km line across Mexico.
When I read "shotgun approach," my first thought was naturally, "I'm not really sure that a shotgun is the best way to kill roaches...."
Southeast Alaska is a rain forest. Not a tropical rain forest (rather, a temperate one), but a rain forest nonetheless. As it turns out, plant life grows prolifically in rain forests.
See:
http://en.wikipedia.org/wiki/Alaska_Panhandle
http://en.wikipedia.org/wiki/Tongass_National_Forest
On US-491 in southeastern Utah, the mile markers go, 67, 68, 68.99, 70. I thought about stealing 68.99 just for the hell of it. The real WTF (as it were) is that 491 used to be 666, but it got changed for some reason...
If I take a relaxed, polite, and understanding attitude towards someone who's helping me, I generally get better results.
This is so true. I make a particular effort to be polite and undemanding when I'm dealing with secretaries or other administrative staff, because getting on their good side makes things much easier when I have a problem that I need their help with.
All that being said, some people are just plain dicks, and all the politeness in the world won't change them.
No kidding. One of the secretaries here is... let us say, not a people person. The first time I had to deal with her, I put on my usual "Hi (big smile), thank you so much for helping me" routine, and she just sort of looked at me like I was something unpleasant that had been spilled on her desk. I eventually figured out that she was like this all the time with everybody. Oh well. She's retiring next week, for which we're all thankful.
Wow. How is it that an "ex-hacker" who now "specialises in security from the white hat side of the fence" (from the author's bio) can have so little clue about the responsible disclosure debate and the economics of modern vulnerability research? Maybe getting lambasted on Slashdot will be a wake-up call for him to actually do his homework before he spouts off.
Calm down. It's sponsored by Usenix, so the accepted papers will be published online, freely accessible by everyone. Also, academic conferences exist to provide a forum for new research, that is, ideas and results that no one save the authors have seen before; therefore, the workshop attendees will in fact learn from their peers.
Close, but no cigar. The correct answer is "Getting his Ph.D. at UCLA":
m l
http://www.wired.com/wired/archive/15.02/posts.ht
I had a summer internship with the R&D branch of one of those other companies you mentioned. I was tasked with writing an SSL man-in-the-middle platform, so when somebody told the security group, for instance, "Yeah, we're just going to use Anonymous Diffie-Hellman mode," the security group could clearly demonstrate why that was a stupid idea. Just to be clear, this was not a hypothetical situation. I mean really, ADH! "Sweet, I have an encrypted channel to... somebody!" *sigh* You have to jump through special hoops to get OpenSSL to use ADH at all.
This whole thing is old news, though. FX told us all about the fun you can have with printers way back at Defcon 10.
Yeah, but who (besides you) calls anybody on a landline anymore?
Seriously, though, everybody I know under the age of 30 uses a cell as their primary phone.
Schneier, Applied Cryptography. Everything you need to know about practical cryptographic primitives and cryptosystems is in the first 5-10 chapters of that book.
Leaving a browser open at all times is a sufficiently common usage pattern that, your glib dismissal aside, the Mozilla developers should at least consider a significant push to find and fix the memory leaks that we're complaining about.
FWIW, Safari is in my experience a little memory-leaky, but nothing like Firefox.
(Hmm, my sig is a little hypocritical this time around...)
The world's most dangerous network: DefCON. Everybody who attends knows it. And yet *still*, people use plaintext authentication to all sorts of services. And then they appear on the Wall of Sheep.
"In the future, all I have to do is let some bacteria into your building; they float through the air conditioning system, land on your keyboard, you can't see them, you don't know they are there."
Neal Stephenson called, he wants his threat model back.
(Cf. urn:isbn:0553096095)
Given a quick skim of the Sandia paper (and a closer read of a couple of the important sections), it looks like these Sandia group got results roughly similar to a portion of the work that Maynor and Cache presented (sorry, PPT) at BlackHat and DefCON. From what I can tell, the techniques used by Maynor and Cache were a little more ad-hoc than the Sandia group's machine-learning approach, but unless they write up their results with more depth than a Powerpoint presentation, doing a strong comparison is sort of hard. I'm amused that the teams presented their results within days of each other, but at rather different venues. (Also, I'm sad that USENIX Security overlapped DefCON this year; I've done both in years past, and they make for a good week+ of security and drinking.)
Mea culpa. Serves me right for taking some random blogger seriously. Still, there's enough "MSM is teh suck!!!1" noise out there that you'll have to pardon me for my confusion.
Oh noes, somebody at Reuters made a mistake! Amazing as this may sound, professional news organizations do issue corrections from time to time. Why am I not defending Wikipedia in the same statement? The charged and misleading language that appeared on Wikipedia was intentionally put up by some random person.
"And journalism has sunk to a new low"? Come down off your high horse, Mr Unger.
You are so right. Would that I had mod points....
This is old news. Googling for [networked landmines] brings up a Register story on the program from 2003.
In vi or vim, put the cursor on the opening brace and type '>%' (that's to indent; type '<%' to exdent).