And how many strippers stay in it for longer than a month without resorting to cocaine to keep doing the job?
Oh let's see how about the one I know that is doing it as a way of supporting herself through college. She's the one that was in foster care most of her life and did not have a mom's basement to crawl-out of and daddy to foot the bills. And guess what I have never seen her without her clothes on so it is not like it was a story she told me to hustle me for cash. Nope she is my wife's friend.
Or how about this one, another stripper working with the previous one. She is a single mom and her toddler has a dead beat dad. Dancing two nights a week for a basically guaranteed $4K (mostly tax free) a month is a great gig. Just drop the kid off at grandma's to work and tan.
It's about finding just how much you can embarass a girl with a dollar.
You are VERY screwed-up, completely warped:( You think you are embarrassing them? I have news for you crap for brains, they are hustling you for your money, you are the mark, they are in complete control of you.
BTW the smelly guys are a real problem. That and the bruises and scratches from bumping into stuff while dancing are the biggest complaints according to my wife's friend:)
Here is a recent anecdote. I went to a store and they had a deal where if you used a particular credit card you got 15% off of your purchase. The cashier selected cash by mistake and I had to go over to the customer service counter to straighten it all out. (I am too honest to let the poor cashier be accused of theft when the cash in the drawer did not add-up at the end of her shift.) The manager told me that she could not figure-out the 15% discount because her calculator lacked a percent key. This was the manager! I just did it on a circular in front of her, but it was terribly depressing.
Another time four of us went out to dinner and one person decided to put it all on his credit card. He was having trouble figuring-out the tip. The total on the bill was something like $100.15:(
On most unix-likes, baring any extra tweaking, you need to be root to open a raw socket and people do not run as root routinely. (Well except those using Linspire I guess...)
I just looked and there are some tools in Developer -> Applications -> Performance Tools that may be what I would like to use. In particular Shark and Sampler look promising. I wish there was a simple command line program that worked similarly.
Right I know that but I was not after tracing syscalls but rather tracing function calls. Do a man truss on solaris sometime too see what I mean about all the features that are missing from ktrace.
On Solaris there is a command truss that is the king of all truss-like commands. Unlike strace, ktrace, and BSD truss this tool can print a trace of all function calls made by an application as it runs (among many other useful things). Does anyone here know of an analogous tool for MacOS X? If not I wonder if an awk/perl script munging the output of nm to generate tracepoints for gdb where each trace point creates a new tracepoint at the instruction where the function call returns, prints out the funtion name and the contents of r2-r10 or so, then continues on or something like this would be something someone has already written.
I pulled it off. After I finished with uni and did not have a real job yet, I got by on 4 'servings' of TacoBell a day. A serving was one of bean burrito, soft taco, or hard taco. I believe that back then the prices were some combination of 49, 59, and 69 cents each for those items. I also bought really cheap three liter bottles of cola, Shasta if I remember correctly, at I think 59 cents back then. With the right amount of free water and sneaking free food and soft drinks from faculty talks and the like I was able to pull through at under $20 per week. I remember because that was my budget for food.
Now I have to admit that that was a crazy way to live and I do not recommend it to anyone, but honestly onions, green bell peppers, potatoes, rice, noodles, ramen, and frozen chicken were very inexpensive as well. When I got a real job and was still single I was able to live on that and the price was not all the worse. Definitely not $100/wk. Certainly it was worse in CA than IL but still not $100/wk...
Just so that everyone knows, on google Results 1,2, and 4 are relevant. (I do not know if any more are relevant, I did not hit page down.) I Save RX is a controversial program created by the state of IL that allows certain people to fill prescriptions from Canadian pharmacies. It is truly depressing that the relevant sites run by the state of IL do not show-up anywhere within the first 100 results from yahoo! Instead yahoo is filled with results from spammy questionable sites selling drugs online.
That was a great comparison tool Zordok, thanks for the link!
Every now and then google results annoyed me. Then once I was siting at my wife's account and I did a search for "i save rx" on yahoo and this is what I got. For those that do not know, I Save RX is the IL program that lets you fill your prescriptions from Canadian pharmacies. I went to google and got this page of results. So while the yahoo listing was full of spammy questionable sites selling drugs online, the google results had relavant results on at least lines 1, 2, and 4. I guess what I am trying to say is that after using google for a while you notice its warts, but then when you try to use some other search engine, you see just how bad the others can be!
Now I see how this was a joke and I feel really stupid and embarrassed about that. Sorry about that, I guess I do not have a keen sense of humor. Also I have a - where there should be a + in one place and I call 1 a prime in another, so that serves me right. To be honest, the whole 'Indian math guy' bit bothered me too. At least I got a nice puzzle out of it:)
Sorry maybe I was a little testy in my reply. It just bugged me that you gave such an authoritative sounding reply to someone that said they were not a maths person which was flawed. I suppose one can think of changing bases as a trivial isomorphism, in fact it is identity. If that was what would have worked here though you would not have had to change the conjecture to use all numbers smaller. That should be a tip-off that the conjecture may not be true.
In fact the logical error is that you were considering the identity transformation but really you were in fact over two different spaces. One was represented by Dn*10^n+...+D0*10^0 and the other by Hm*16^m+...+H0*16^0 and the manner in which you changed bases does not work that way as you move from one to the other. In fact simply considering that 10=2*5 and the original statement was mod 5 not mod 2 gives a gut feeling that there is more at work.
Again, cute:) Also notice that if you consider the differences between (x±y)^2 in general you have a nice little pattern:)
Incidentally my first reaction was, "they are 100 apart because 10, 24, 26 is a Pythagorean triple," but I thought there might be more too it than that. Anyway I suppose that was the 'Greek math guy' you were referring to:)
That is your longer non-trivial way to write 4pq as a sum of consecutive odds.
The trick is to realize that the sum of consecutive odds is a square. It is a simple formula that people can remember and probably every one proved via induction when they learned induction. The easy way to remember it is like this:
X
YZ XY
YYX ZZY ZZY
Maybe my lame diagrams above actually make it harder to visualize:)
But honestly now I remember why I got a degree in CS in addition to Math. It was the small bunch of antisocial pricks in the Math program that made even the worst of the CS types fun to be around in comparison. Why did you go out of your way and attack me with your reply? Look I solved this in no time in my head when I was making the bed this morning. Does that deserve to be an article on./? No it is a cute puzzle, fun to solve, but please do not present it as if it were some earth shattering discovery in mathematics. And when you make a simple mistake don't start attacking the terminology of someone else when they honestly don't see what you are talking about because of the fact that you poorly characterized the problem. Once I saw what you were talking about I had fun solving it. It was a cute puzzle as it turns-out, one that I gave my wife to work-out this morning.
Really I am not trying to be mean here. Probably no one has told you yet about how it counts to be nice. If you do not keep it in check what will happen? Is it your goal to be one of the smart-a*s profs that make their students feel like idiots? Wouldn't you rather be one of those that makes them feel intelligent - the kind that let them see that they are learning?
You do not get an article because any positive integer can be trivially represented as a sequence of odd primes by using the unary representation:
1=1 2=1+1 3=1+1+1 4=1+1+1+1
The article was about a statement that took something like 60 years to prove and then just recently a generalization was discovered and proved. Two very different things.
The grandparent post's logic is incorrect. You cannot just reduce the bases that way and hope to get anything meaningful. In fact look at the original statement and take the case of the number 4. It has the trivial partitions of:
4 1 3 2 2 1 1 2 1 1 1 1
There are 5 of them, which fits the bill of being divisible by 5. Now the grandparent poster thought he saw a pattern by reducing to base two but in order for his claim to hold he had to change it so that it was the sum of all SMALLER numbers. That should have been the tip-off that something was amiss.
Unfortunately the site is./ed so it will not do much good to actually click on any of the thumbnails. The good news is that so many of the thumbnails are of icons that in effect the cached thumbnails are essentially full-size:)
I wonder how happy all the Hymn and J-Hymn users out there are about what DVD Jon did. By releasing PyMusique, he got Apple to force everyone to use 4.7 iTunes if they want to use the iTMS. I believe that 4.7 broke Hymn and unless that has been addressed, now people will no longer be able to remove the DRM from music that they purchased from the iTMS.
What happened was fine, nothing to get your knickers into a knot about. When you buy music with DRM you are agreeing to use it according to the terms set forth. One of those terms is that you agree to how the terms may change in the future. That is why I do not buy music with DRM, the fact that what I can do with that music can change at any time.
It is too bad that the Apple DRM happens to be one of the least onerous and DVD Jon gave Apple a reason to make people move to slightly more restrictive terms with 4.7, but still just the fact that Apple can modify what you can and cannot do with the music from the iTMS is an immediate turn-off for me.
Thanks to the three people that suggested cmd+: and F5, but I already knew about those and they are not what I wanted.
The F5 is autocomplete and that what that does is it tries to finish words that you have started typing. (Which for some reason does not work on my iBook and I wish somebody had a suggestion for why not, but that is what it does on the eMac at least.)
The cmd+: brings up the Spelling dialog box. That is not exactly what I want. If you have ever tried to use that dialog box for spell checking you will know what I mean. First of all it does not reflect changes in the underlying window reliably. Secondly, when using only the keyboard it takes way too many key presses to move around and do what you want. Here it is in its gory details:
First I need to press cmd+: and then I need to press ctl+F6 to get to the dialog, and at that point I have to use the down arrow key to find the right word. At that point press return. Now press esc to make the dialog disappear lather rinse repeat... Guess what that is so much moving about that it is simply easier to cursor back and make the correction myself.
So what are the problems. Say I have some C code in an email, Oh man there are a ton of funky words (funtions names, #defines, variables, etc...) that I need to ignore. How do I ignore them, tab tab tab.... God forbid I had forgotten to press ctl+F6, now part of my email is blown away. Plus I have to skip through these 'misspelled words' each and every time, I bring-up the dialog when a word gets a red underline. I certainly do not want to add all these functions and what not to a dictionary, I have ctags and cscope for that...
What I want is a simple way to 'simulate' the behavior of a control click on a misspelled word. For example, do some key press and that simulates a control click on the previous word if the cursor is after white space. Then I would like to be able to move around the list of suggested spelling corrections (cursor keys make sense here though alphanumeric keys quickly finding words where that character first appears after removing all common prefixes would be very handy) with the keyboard and select one (return makes sense here) or dismiss the pop-up menu (esc make sense here).
There is almost something like this in Universal Access. You can turn on the Mouse Keys under the Mouse tab there. This lets me simulate a control click with ctl+5 on the numeric keypad (or fn+ctl+i on my iBook even better since those keys are near to my fingers at rest on the keyboard). The only problem with this is that the control click happens wherever the mouse cursor happens to be at the moment completely independently of where the text cursor is. Of course the current behavior is correct, but not what I would like...
Here is one thing I have noticed. We have an eMac at home and two of my three kids are preschoolers. They have a very easy time using the zero (what I mean is that the entire mouse is one big button) button mouse from Apple. It is not too big too. At the library they have these two button plus scroll wheel Microsoft ergonomic mice connected to the computers for the kids. First of all those mice are way too big and there is this big hump at the base of the mouse that makes it very difficult for my kids to use. Since they have to hold the mouse near the top, very often the mouse will turn to the side and then the motion is all wrong relative to what they expect the cursor to do on the screen. The fact that the scroll wheel is in the way and that there are two buttons also causes confusion. What happens is that they end-up just clicking repeatedly until they finally click on the left mouse button and if they click on the scroll wheel their hand rolls off.
On the other hand the size of the current Apple mouse is just perfect. It is not too small for an adult and not too big for a child. Because of the size and the fact that the whole mouse is one big button, my kids can hold the mouse near its middle, and then it does not rotate while being moved.
I have heard the argument that once you start using a computer long enough you start wanting extra mouse buttons. What I think is that those people are not sophisticated enough. Even when I was using unix primarily, I configured fvwm and vim so that I could do almost everything from the keyboard. Today there are keyboard shortcuts for almost anything on OS X plus a bunch of small apps to add even more shortcut functionality. I really do not miss a three button mouse all the much at all. In fact I use SideTrack on my iBook and think that is perfect for the times I need to copy and paste in X11.app. Maybe Apple should make a compact keyboard with a trackpad instead of a two button mouse. If that keyboard was wireless, it would be perfect for sitting on the couch too especially with two finger scrolling.
One thing about OS X that is very frustrating is that I have not figured out an easy way to use the built-in spell checker with only the keyboard. If anyone knew an easy way to pop-up that menu with suggested corrections, I would really appreciate it. Also using the accessibility features and that spelling dialog box with only the keyboard is really annoying because the things you want to do are too many key presses away, so that is not really a viable solution...
The site is slashdotted so I cannot read it, but i wonder if something akin to this could be used for compressing motion video. For example the intensity is encoded with currrent techniques, but instead of the color being encoded at a lower resolution, instead only a very small amount of colored points are encoded. Then during the decoding, the decoder uses an error function, intensity, and the time domain of previous and future frames to 'fill' the colors out.
How about this Frank Kozik picture, bonus it is still "Beastie" of a sort.
Or if you still want a daemon, maybe this by Coop will do. I am sure that won't be offensive...
What in the name!? I really really cannot understand it. How can Beastie the Daemon be at all unproffesional? For crying out loud I have a "Dirt Devil" vacuum cleaner in the house! Why would FreeBSD throw away this mascot that is so well entrenched? In a sense it has such a strong "brand identity" already. Do they really want to end-up with something so generic as the NetBSD logo?
Wow I AM a loser, I am taking this way too seriously. I never knew I was so attached to beastie.
Years ago we had a similar text based menu system on the computers (286 in a time that my friends had Pentiums at home) in the high school library. I figured out a way to load a TSR onto that system as well. I think I could rewrite the autoexec.bat or boot off a floppy, something trivial like that. Anyway once I figured-out that I too copied the menu programs and figured-out how the system worked.
The teachers and administration had more programs that they could run. (In fact the principal's daughter said a slip-up once that made it clear she had 'special treatment'.) This bothered me, just the fact that they could and I couldn't practically drove me nuts. I found out that those programs were on a different drive (network mount) than those on which the programs for students were on. Since I did not have the teacher menus or the teacher mount was password protected, I cannot remember what was the problem I ran into, I decided I needed a teacher's password to make progress.
So I wrote a TSR that was essentially a keylogger. It turned-out not to be trivial to write to a file from a TSR and I figured-out how to make my TSR so that it would be two TSRs in one. I piggy-backed onto the mouse driver and got my TSR to load with it.
I did this because I was worried that someone would notice a weird extra TSR and get suspicious. I also used some inconspicuous file for the log. I did something like XOR the keystrokes with with 0xd00d or rot13, something silly like that hoping it would not look too obvious what the file was. I installed this on every machine with a hard drive (two or three) and hoped for the best. A year or so before me another kid had gotten into a world of trouble for rigging a computer in the library so that it displayed a pornographic picture at boot, so I did not want to raise any suspicions of what I was up to.
When I started this I knew only a bunch about DOS and pascal. In fact my inspiration was the example in TurboPascal of the TSR that did keyboard clicks. I worked on this 'project' for a little more than two semesters, and I learned a considerable deal about DOS programming and also x86 assembly. It was the most I ever learned in high school:)
The funny thing is that I got bored very quickly with this after I got it working and stopped even checking the log file. I really thought that the librarian would login at some point, but her login was not there whenever I checked... She had a computer in her office, so I guess I was too optimistic. I moved onto another 'project' at that point though, I made 1.74 MB floppies put stacker on them and got a stripped down version of Win3.0 on there with sol and write:)
Those computers ran my little TSR for two years until they were upgraded to 486s and they were never discovered to my knowledge...
Slashdot Mirror
Oh let's see how about the one I know that is doing it as a way of supporting herself through college. She's the one that was in foster care most of her life and did not have a mom's basement to crawl-out of and daddy to foot the bills. And guess what I have never seen her without her clothes on so it is not like it was a story she told me to hustle me for cash. Nope she is my wife's friend.
Or how about this one, another stripper working with the previous one. She is a single mom and her toddler has a dead beat dad. Dancing two nights a week for a basically guaranteed $4K (mostly tax free) a month is a great gig. Just drop the kid off at grandma's to work and tan.
It's about finding just how much you can embarass a girl with a dollar.
You are VERY screwed-up, completely warped :( You think you are embarrassing them? I have news for you crap for brains, they are hustling you for your money, you are the mark, they are in complete control of you.
BTW the smelly guys are a real problem. That and the bruises and scratches from bumping into stuff while dancing are the biggest complaints according to my wife's friend :)
Cache Missing for Fun and Profit
Here is a recent anecdote. I went to a store and they had a deal where if you used a particular credit card you got 15% off of your purchase. The cashier selected cash by mistake and I had to go over to the customer service counter to straighten it all out. (I am too honest to let the poor cashier be accused of theft when the cash in the drawer did not add-up at the end of her shift.) The manager told me that she could not figure-out the 15% discount because her calculator lacked a percent key. This was the manager! I just did it on a circular in front of her, but it was terribly depressing.
:(
Another time four of us went out to dinner and one person decided to put it all on his credit card. He was having trouble figuring-out the tip. The total on the bill was something like $100.15
On most unix-likes, baring any extra tweaking, you need to be root to open a raw socket and people do not run as root routinely. (Well except those using Linspire I guess...)
I just looked and there are some tools in Developer -> Applications -> Performance Tools that may be what I would like to use. In particular Shark and Sampler look promising. I wish there was a simple command line program that worked similarly.
Right I know that but I was not after tracing syscalls but rather tracing function calls. Do a man truss on solaris sometime too see what I mean about all the features that are missing from ktrace.
On Solaris there is a command truss that is the king of all truss-like commands. Unlike strace, ktrace, and BSD truss this tool can print a trace of all function calls made by an application as it runs (among many other useful things). Does anyone here know of an analogous tool for MacOS X? If not I wonder if an awk/perl script munging the output of nm to generate tracepoints for gdb where each trace point creates a new tracepoint at the instruction where the function call returns, prints out the funtion name and the contents of r2-r10 or so, then continues on or something like this would be something someone has already written.
I pulled it off. After I finished with uni and did not have a real job yet, I got by on 4 'servings' of TacoBell a day. A serving was one of bean burrito, soft taco, or hard taco. I believe that back then the prices were some combination of 49, 59, and 69 cents each for those items. I also bought really cheap three liter bottles of cola, Shasta if I remember correctly, at I think 59 cents back then. With the right amount of free water and sneaking free food and soft drinks from faculty talks and the like I was able to pull through at under $20 per week. I remember because that was my budget for food.
Now I have to admit that that was a crazy way to live and I do not recommend it to anyone, but honestly onions, green bell peppers, potatoes, rice, noodles, ramen, and frozen chicken were very inexpensive as well. When I got a real job and was still single I was able to live on that and the price was not all the worse. Definitely not $100/wk. Certainly it was worse in CA than IL but still not $100/wk...
Just so that everyone knows, on google Results 1,2, and 4 are relevant. (I do not know if any more are relevant, I did not hit page down.) I Save RX is a controversial program created by the state of IL that allows certain people to fill prescriptions from Canadian pharmacies. It is truly depressing that the relevant sites run by the state of IL do not show-up anywhere within the first 100 results from yahoo! Instead yahoo is filled with results from spammy questionable sites selling drugs online.
That was a great comparison tool Zordok, thanks for the link!
Every now and then google results annoyed me. Then once I was siting at my wife's account and I did a search for "i save rx" on yahoo and this is what I got. For those that do not know, I Save RX is the IL program that lets you fill your prescriptions from Canadian pharmacies. I went to google and got this page of results. So while the yahoo listing was full of spammy questionable sites selling drugs online, the google results had relavant results on at least lines 1, 2, and 4. I guess what I am trying to say is that after using google for a while you notice its warts, but then when you try to use some other search engine, you see just how bad the others can be!
Now I see how this was a joke and I feel really stupid and embarrassed about that. Sorry about that, I guess I do not have a keen sense of humor. Also I have a - where there should be a + in one place and I call 1 a prime in another, so that serves me right. To be honest, the whole 'Indian math guy' bit bothered me too. At least I got a nice puzzle out of it :)
Sorry maybe I was a little testy in my reply. It just bugged me that you gave such an authoritative sounding reply to someone that said they were not a maths person which was flawed. I suppose one can think of changing bases as a trivial isomorphism, in fact it is identity. If that was what would have worked here though you would not have had to change the conjecture to use all numbers smaller. That should be a tip-off that the conjecture may not be true.
In fact the logical error is that you were considering the identity transformation but really you were in fact over two different spaces. One was represented by Dn*10^n+...+D0*10^0 and the other by Hm*16^m+...+H0*16^0 and the manner in which you changed bases does not work that way as you move from one to the other. In fact simply considering that 10=2*5 and the original statement was mod 5 not mod 2 gives a gut feeling that there is more at work.
Cute puzzle, thanks :)
:) Also notice that if you consider the differences between (x±y)^2 in general you have a nice little pattern :)
:)
26^2 = (25+1)^2 = 25^2 + 2*25 + 1
24^2 = (25-1)^2 = 25^2 - 2*25 + 1
Subtract them:
26^2 - 24^2 = 4*25 = 100
Again, cute
Incidentally my first reaction was, "they are 100 apart because 10, 24, 26 is a Pythagorean triple," but I thought there might be more too it than that. Anyway I suppose that was the 'Greek math guy' you were referring to
Now that I know what you are talking about it is a fun puzzle to solve it turns-out :)
... + 2(p+q) - 1 = (p+q)^2 = p^2 + 2pq + q^2 ... + 2(p-q) - 1 = (p-q)^2 = p^2 - 2pq + q^2
... + 2(p+q) - 1 = 4pq
:)
./? No it is a cute puzzle, fun to solve, but please do not present it as if it were some earth shattering discovery in mathematics. And when you make a simple mistake don't start attacking the terminology of someone else when they honestly don't see what you are talking about because of the fact that you poorly characterized the problem. Once I saw what you were talking about I had fun solving it. It was a cute puzzle as it turns-out, one that I gave my wife to work-out this morning.
Take two integers p > 1 and q > 1.
1 + 3 +
1 + 3 +
Subtract the bottom from the top:
2(p-q) - 1 +
That is your longer non-trivial way to write 4pq as a sum of consecutive odds.
The trick is to realize that the sum of consecutive odds is a square. It is a simple formula that people can remember and probably every one proved via induction when they learned induction. The easy way to remember it is like this:
X
YZ
XY
YYX
ZZY
ZZY
Maybe my lame diagrams above actually make it harder to visualize
But honestly now I remember why I got a degree in CS in addition to Math. It was the small bunch of antisocial pricks in the Math program that made even the worst of the CS types fun to be around in comparison. Why did you go out of your way and attack me with your reply? Look I solved this in no time in my head when I was making the bed this morning. Does that deserve to be an article on
Really I am not trying to be mean here. Probably no one has told you yet about how it counts to be nice. If you do not keep it in check what will happen? Is it your goal to be one of the smart-a*s profs that make their students feel like idiots? Wouldn't you rather be one of those that makes them feel intelligent - the kind that let them see that they are learning?
You do not get an article because any positive integer can be trivially represented as a sequence of odd primes by using the unary representation:
1=1
2=1+1
3=1+1+1
4=1+1+1+1
The article was about a statement that took something like 60 years to prove and then just recently a generalization was discovered and proved. Two very different things.
The grandparent post's logic is incorrect. You cannot just reduce the bases that way and hope to get anything meaningful. In fact look at the original statement and take the case of the number 4. It has the trivial partitions of:
4
1 3
2 2
1 1 2
1 1 1 1
There are 5 of them, which fits the bill of being divisible by 5. Now the grandparent poster thought he saw a pattern by reducing to base two but in order for his claim to hold he had to change it so that it was the sum of all SMALLER numbers. That should have been the tip-off that something was amiss.
cached thumbnails
Unfortunately the site is ./ed so it will not do much good to actually click on any of the thumbnails. The good news is that so many of the thumbnails are of icons that in effect the cached thumbnails are essentially full-size :)
Enjoy...
I wonder how happy all the Hymn and J-Hymn users out there are about what DVD Jon did. By releasing PyMusique, he got Apple to force everyone to use 4.7 iTunes if they want to use the iTMS. I believe that 4.7 broke Hymn and unless that has been addressed, now people will no longer be able to remove the DRM from music that they purchased from the iTMS.
What happened was fine, nothing to get your knickers into a knot about. When you buy music with DRM you are agreeing to use it according to the terms set forth. One of those terms is that you agree to how the terms may change in the future. That is why I do not buy music with DRM, the fact that what I can do with that music can change at any time.
It is too bad that the Apple DRM happens to be one of the least onerous and DVD Jon gave Apple a reason to make people move to slightly more restrictive terms with 4.7, but still just the fact that Apple can modify what you can and cannot do with the music from the iTMS is an immediate turn-off for me.
Thanks to the three people that suggested cmd+: and F5, but I already knew about those and they are not what I wanted.
The F5 is autocomplete and that what that does is it tries to finish words that you have started typing. (Which for some reason does not work on my iBook and I wish somebody had a suggestion for why not, but that is what it does on the eMac at least.)
The cmd+: brings up the Spelling dialog box. That is not exactly what I want. If you have ever tried to use that dialog box for spell checking you will know what I mean. First of all it does not reflect changes in the underlying window reliably. Secondly, when using only the keyboard it takes way too many key presses to move around and do what you want. Here it is in its gory details:
First I need to press cmd+: and then I need to press ctl+F6 to get to the dialog, and at that point I have to use the down arrow key to find the right word. At that point press return. Now press esc to make the dialog disappear lather rinse repeat... Guess what that is so much moving about that it is simply easier to cursor back and make the correction myself.
So what are the problems. Say I have some C code in an email, Oh man there are a ton of funky words (funtions names, #defines, variables, etc...) that I need to ignore. How do I ignore them, tab tab tab.... God forbid I had forgotten to press ctl+F6, now part of my email is blown away. Plus I have to skip through these 'misspelled words' each and every time, I bring-up the dialog when a word gets a red underline. I certainly do not want to add all these functions and what not to a dictionary, I have ctags and cscope for that...
What I want is a simple way to 'simulate' the behavior of a control click on a misspelled word. For example, do some key press and that simulates a control click on the previous word if the cursor is after white space. Then I would like to be able to move around the list of suggested spelling corrections (cursor keys make sense here though alphanumeric keys quickly finding words where that character first appears after removing all common prefixes would be very handy) with the keyboard and select one (return makes sense here) or dismiss the pop-up menu (esc make sense here).
There is almost something like this in Universal Access. You can turn on the Mouse Keys under the Mouse tab there. This lets me simulate a control click with ctl+5 on the numeric keypad (or fn+ctl+i on my iBook even better since those keys are near to my fingers at rest on the keyboard). The only problem with this is that the control click happens wherever the mouse cursor happens to be at the moment completely independently of where the text cursor is. Of course the current behavior is correct, but not what I would like...
Here is one thing I have noticed. We have an eMac at home and two of my three kids are preschoolers. They have a very easy time using the zero (what I mean is that the entire mouse is one big button) button mouse from Apple. It is not too big too. At the library they have these two button plus scroll wheel Microsoft ergonomic mice connected to the computers for the kids. First of all those mice are way too big and there is this big hump at the base of the mouse that makes it very difficult for my kids to use. Since they have to hold the mouse near the top, very often the mouse will turn to the side and then the motion is all wrong relative to what they expect the cursor to do on the screen. The fact that the scroll wheel is in the way and that there are two buttons also causes confusion. What happens is that they end-up just clicking repeatedly until they finally click on the left mouse button and if they click on the scroll wheel their hand rolls off.
On the other hand the size of the current Apple mouse is just perfect. It is not too small for an adult and not too big for a child. Because of the size and the fact that the whole mouse is one big button, my kids can hold the mouse near its middle, and then it does not rotate while being moved.
I have heard the argument that once you start using a computer long enough you start wanting extra mouse buttons. What I think is that those people are not sophisticated enough. Even when I was using unix primarily, I configured fvwm and vim so that I could do almost everything from the keyboard. Today there are keyboard shortcuts for almost anything on OS X plus a bunch of small apps to add even more shortcut functionality. I really do not miss a three button mouse all the much at all. In fact I use SideTrack on my iBook and think that is perfect for the times I need to copy and paste in X11.app. Maybe Apple should make a compact keyboard with a trackpad instead of a two button mouse. If that keyboard was wireless, it would be perfect for sitting on the couch too especially with two finger scrolling.
One thing about OS X that is very frustrating is that I have not figured out an easy way to use the built-in spell checker with only the keyboard. If anyone knew an easy way to pop-up that menu with suggested corrections, I would really appreciate it. Also using the accessibility features and that spelling dialog box with only the keyboard is really annoying because the things you want to do are too many key presses away, so that is not really a viable solution...
The site is slashdotted so I cannot read it, but i wonder if something akin to this could be used for compressing motion video. For example the intensity is encoded with currrent techniques, but instead of the color being encoded at a lower resolution, instead only a very small amount of colored points are encoded. Then during the decoding, the decoder uses an error function, intensity, and the time domain of previous and future frames to 'fill' the colors out.
Or if you still want a daemon, maybe this by Coop will do. I am sure that won't be offensive...
What in the name!? I really really cannot understand it. How can Beastie the Daemon be at all unproffesional? For crying out loud I have a "Dirt Devil" vacuum cleaner in the house! Why would FreeBSD throw away this mascot that is so well entrenched? In a sense it has such a strong "brand identity" already. Do they really want to end-up with something so generic as the NetBSD logo?
Wow I AM a loser, I am taking this way too seriously. I never knew I was so attached to beastie.
Years ago we had a similar text based menu system on the computers (286 in a time that my friends had Pentiums at home) in the high school library. I figured out a way to load a TSR onto that system as well. I think I could rewrite the autoexec.bat or boot off a floppy, something trivial like that. Anyway once I figured-out that I too copied the menu programs and figured-out how the system worked.
:)
:)
The teachers and administration had more programs that they could run. (In fact the principal's daughter said a slip-up once that made it clear she had 'special treatment'.) This bothered me, just the fact that they could and I couldn't practically drove me nuts. I found out that those programs were on a different drive (network mount) than those on which the programs for students were on. Since I did not have the teacher menus or the teacher mount was password protected, I cannot remember what was the problem I ran into, I decided I needed a teacher's password to make progress.
So I wrote a TSR that was essentially a keylogger. It turned-out not to be trivial to write to a file from a TSR and I figured-out how to make my TSR so that it would be two TSRs in one. I piggy-backed onto the mouse driver and got my TSR to load with it.
I did this because I was worried that someone would notice a weird extra TSR and get suspicious. I also used some inconspicuous file for the log. I did something like XOR the keystrokes with with 0xd00d or rot13, something silly like that hoping it would not look too obvious what the file was. I installed this on every machine with a hard drive (two or three) and hoped for the best. A year or so before me another kid had gotten into a world of trouble for rigging a computer in the library so that it displayed a pornographic picture at boot, so I did not want to raise any suspicions of what I was up to.
When I started this I knew only a bunch about DOS and pascal. In fact my inspiration was the example in TurboPascal of the TSR that did keyboard clicks. I worked on this 'project' for a little more than two semesters, and I learned a considerable deal about DOS programming and also x86 assembly. It was the most I ever learned in high school
The funny thing is that I got bored very quickly with this after I got it working and stopped even checking the log file. I really thought that the librarian would login at some point, but her login was not there whenever I checked... She had a computer in her office, so I guess I was too optimistic. I moved onto another 'project' at that point though, I made 1.74 MB floppies put stacker on them and got a stripped down version of Win3.0 on there with sol and write
Those computers ran my little TSR for two years until they were upgraded to 486s and they were never discovered to my knowledge...
That was an informative blog entry.