Anathem is Literature
on
Anathem
·
· Score: 2, Interesting
Stephensen has been stepping incrementally closer to being a literature author with each book he's written. Snow crash is fluff, Cryptonomicon is pretty deep, and the Baroque Cycle is a master work (in the original sense). Anathem is his first post master work book.
Many posters have made the claim: "It would be better if you removed X", for various values of X. What is instructive is that not everyone agrees on X. Stephensen had a lot to say in this book, on many topics.
I'll address a few things here, but this list isn't exhaustive: * Unresolved plot elements are not bad. Only in very bad fiction does absolutely everything happen in service of the ultimate confrontation. Some things just happen, and we learn about the characters in how they deal with them. * Characters exist for themselves, not the plot. If every character was there 'for something', this would be a (bad) video game, but it isn't, its a book. * The ultimate conclusion of the book is that intellectuals have a duty to the world to remain engaged. The first half (roughly) of the book exists to convince you that being segregated would be lovely, while the second half drives towards the negative consequences of that approach. The character development and the plot both work to develop this theme over time.
Your waranty was issued by a corporate entity, and as such, is a contract with that entity. It doesn't matter how many times the ownership of the company changes, as long as the company exists, the company has a stable timeline.
Short answer, what they are doing is not legal, nail them for breach of contract.
The HTTP protocol defines header comments, which are only valid on a few header fields. There are some increadible problems with them.
1. Comments are recursive. 2. Comments break the header continuation model used elsewhere for continued values.
This means that HTTP headers must make semantic decisions about the header type they are working with in order to properly perform their lexical parsing. It might not seem like much, but it's a sublte stone bitch.
This game has a) the stupidest premise, and b) the most awesome gameplay of any game that I've seen or played which has come out in the past 5 years. Play it, if you can get a copy!
You use linux because it works better for you, right? Why do you want to sheild management from this reality? Let them see your frustration, and keep making the case "this used to work, why is this so hard?".
Remember, managment doesn't like windows, they like money.
Why people are so willing to take the hit for other people, I'll never understand.
Okay, I use a USB drive enclosure on Linux to do my backups, and it works pretty well.
However, it took me a very long time to learn to set the 'sync' option in the mount options. USB writes much more slowly than a normal harddrive, and if sync isn't set, it is possible for the system to buffer all writes to the drive up to the point where it consumes most system memory, and the machine becomes unresponsive.
Perhaps this is fixed in 2.6; but it doesn't really matter. You are doing backups, the backup isn't done until it's all on the disk, so setting the sync option just means that your writes "seem" to take longer, and your unmount at the end seems faster. Without sync, you pay for the buffering in the unmount, which will hang while it finishes syncing the disk.
I was in the same boat as you. My hands hurt, and I was a professional programmer. I _knew_ I needed to learn to touch type, but I couldn't stop looking at the keyboard.
So I painted my keyboard black. The first week _sucked_, but by week 3, I was at 80% of where I had been before, and that 80% was touch, no looking at all.
Over the next month, I crept up to a bit faster than I had been, and that was good. But the real benifit to my speed was that, with my hands always in the correct location, all the control-Key and alt-Key keyboard shortcuts for my editor, my shell, and my web browser became available, and even second nature to me.
It is worth it.
There are a number of businesses which sell keyboards with blank keys, for use in typing classes; you can google for them. I reference this only as a means of showing that this has worked for others.
Try it, it _will_ work. Just take the plunge, accept the reduced short term productivity, and paint the keys black.
This is a bit heavy for someone 'wanting to learn some discrete math'.
The book isn't about graphs or proof or number theory. The book is about recurances, and methods of solving them. This brings in such subjects as discrete calculus, and floor and ceiling functions, but its NOT really a discrete math book.
Its a good book to read AFTER you get the basics of discrete though.
the following are very good books on proof and discrete math. Some of the titles are whimsical, but they are not toy books, they are very valuable.
"How To Prove It", "How To Solve It", "Induction and Analogy in Mathematics", and "Patterns of Plausible Inference".
However, it seems you are looking for a book to cram for a test in discrete math. Good luck, not going to find one. More so than any of the lower mathematics, discrete is the beginnings of higher logical analyisys, and you can not really 'cram' it. You have to really read the work, and really work the problems. It has to become part of you.
There seems to be this trend to blame difficulty in learning a subject on the books or the teachers. There are many, many things in the world that you are not smart enough to do; you need to accept this, and figure out what problems you can deal with.
I am not batman, I am not Johan Sebastian Bach, and I am not Richard Feynman, I have accepted this; perhaps you are not capable of Discrete Mathematics. If not, you need to leave CS, and go get in MIS or something, you will be happier.
The whole reason I love the device is that I need so fewer minutes of talk time. I'm doing the school thing, and anything which decouples communication is good. This thing has integrated wonderfully with the way I work, and I use many fewer minutes than before.
Current Apps: Email, AIM, Calendar, Asteroids Clone, ToDo, Notes, A NICE Address Book, Web Browser, SMMS Messages, Phone, Tetris Clone.
And the "Desktop Interface" is unbelievable. You go to a website, login, and you have access to all of the application features of the device. It just keeps everything synced. And these aren't pared down interfaces on the web. It's a good email client, a good calendar, etc.
I agree, you don't really want a WYSIWYG editor for docbook, as that violates the Information/Representation sepperation.
But what would be _really_ useful would be a structured editor, which provided a good mapping at several levels (ala Mozilla's composer, where I can turn on and off tabs). The point of such an editor's representation would not be for final production, but to unambigously display the information's _structure_ to the user, and to facilitate manipulating that structure.
It would be like a very ugly word processor, where the tables would always have borders, etc.
If that editor was then linked with a set of generation tools, to make it EASY to genearte and view PostScript, HTML, XHTML, etc. productions, then you'd really have something. I'd use it, for sure, and I don't mind using docbook tags now.
In fact, would mozilla's composer be a good place to start with a docbook editor? Mozilla has good DOM tools, like the inspector, which loads XML DocBook files just fine.
Okay, many people in this thread are talking about the problems with optimizing for non-true pentiums, which are real problems. I'm not.
Red Hat DOES optimize several packages for specific cpus, namely the kernel and glibc. The real wins to be gained with code optimizations are in memcopy and the place where most of this actually happens are in the c library and the kernel. So if you look, there are i686 and athalon versions of the kernel, and i686 versions of glibc (not sure about athalon glibc).
Two layer SACDs are what I was thinking of, and I confused them with HDCDs.
More informaiton is available: http://www.sonymusic.com/sacd/ http:/ /www.hdcd.com/about/index.html
There is no 'encoding', there are two layers
on
Software HDCD Decoding?
·
· Score: -1, Flamebait
HDCD disks do not 'encode' 20 bits in 16 bits. This is impossible.
They work by having two media layers. The HD layer is translucent, and standard CD players will shine right through it, not see it, and read the old-style CD layer behind it, which is reflective. Newer, HD capapble CD players will read the translucent layer.
Your songs are thus recorded twice on the disk, once on each layer. Without hardware (read 'lasers') which can read the HD layer, it is impossible to access.
The coder's already at a major imparement, so there is no effective cost associated with 'switching' to dvorak left or right.
The current options for this person, until they are healed: 1) Learn to type one handed on a keymap designed for typing one handed. 2) Learn to type one handed on a keymap not designed for typing one handed.
Either way, _learning_ is involved. The full training time for a new keymap is shown to consistantly in research (which I should cite here, but won't) to be about a month to get to full speed.
Full speed is dependant on the collision rate and inherent transit distance of the keyboard, so and Dvorak's keymaps were designed to minimize both; thus 'full speed' is faster on the dvorak mappings.
And, having learned the new mapping, the coder now has an additional option when faced with reduced mobility, or an application which requires heavy mouse use, in the future.
There are two alternate Dvorak keymaps, in addition to the one that most people refer to (and I'm typing with now) when they say 'dvorak'. Dvorak developed a right and a left keymaping, which were designed for the physically handicapped.
These keymaps are set up to be minimal for english text with one hand. Give it a look-see, you might have them on your platform.
You cannot bring a common language dictionary into a discusion based upon technical jargon. Authentication and Authorization have very distinct, and _different_ meanings, in the dialect of english used for discussions in the world of software security. The meanings do not vary, but are constant in just about all the literature in the security field.
Splash provides a distributed database using Spread. Apache-SSL has a plugable SSL conection cache layer, and there is a Splash implementation available. So, when a client connects to an Apache-SSL server in the pool, and the server checks the cache for SSL cache information, it is checking a DB it shares with all the other machines in the pool. This is exactly what is being requested for this Ask Slashdot.
I was developing a library which used glib for some of its module code, and while I was at it, I used it for the string functions as well.
Well, my library was crashing, hard, in the middle of a complex io cascade (read some file, and decided to load a module to read some other file based upon istructions in the first file, etc.) and I had checked and rechecked and rechecked the code. The code was correct, but the damn thing was crashing.
And it was crashing in some of my code, which was _perfect_ (I spent like 3 days on this, that code was checked, bracketed on both sides fully by debug statements, etc.).
Oh, and there had been perfect output from the program up to that point. So, getting another hacker to take a look at it, someone much more familiar with glib (One of the maintainers, actually) it was determined that using the %a format in my output was hosing not the print function, because glibc understood the C99 format character for printing floating and double point numbers with out loss, but in the glib code which estimated memory bounds on string operations before calling through to the underlying sprintf implementation. So the heap was getting corrupted, but it was printing out the correct value.
And then it died later, in my code, when it triped over that heap problem. Much profanity was involved.
* Ability to bring up my $EDITOR when typing in a textarea
Hell yes. I've been dying for this for a while. Not yet desperate enough to code it up, but I've looked at some of the problems that need to be solved to do it.
Once you spawn an $EDITOR, you have to block that textarea somehow, and you have to have a way to drop that block (what if the editor crashes?). This will have to be a visual clue on the textare.
Another problem is that the widget interface is very async, and this block introduces wierd statefulness into the textarea widget.
Slashdot Mirror
Stephensen has been stepping incrementally closer to being a literature author with each book he's written. Snow crash is fluff, Cryptonomicon is pretty deep, and the Baroque Cycle is a master work (in the original sense). Anathem is his first post master work book.
Many posters have made the claim: "It would be better if you removed X", for various values of X. What is instructive is that not everyone agrees on X. Stephensen had a lot to say in this book, on many topics.
I'll address a few things here, but this list isn't exhaustive:
* Unresolved plot elements are not bad. Only in very bad fiction does absolutely everything happen in service of the ultimate confrontation. Some things just happen, and we learn about the characters in how they deal with them.
* Characters exist for themselves, not the plot. If every character was there 'for something', this would be a (bad) video game, but it isn't, its a book.
* The ultimate conclusion of the book is that intellectuals have a duty to the world to remain engaged. The first half (roughly) of the book exists to convince you that being segregated would be lovely, while the second half drives towards the negative consequences of that approach. The character development and the plot both work to develop this theme over time.
Well hell, guess everybody can go home. Nothing more to search for here, it is all figured out.
Your waranty was issued by a corporate entity, and as such, is a contract with that entity. It doesn't matter how many times the ownership of the company changes, as long as the company exists, the company has a stable timeline.
Short answer, what they are doing is not legal, nail them for breach of contract.
This was one of the worst games ever. Please, please tell me this isn't true!
The HTTP protocol defines header comments, which are only valid on a few header fields. There are some increadible problems with them.
1. Comments are recursive.
2. Comments break the header continuation model used elsewhere for continued values.
This means that HTTP headers must make semantic decisions about the header type they are working with in order to properly perform their lexical parsing. It might not seem like much, but it's a sublte stone bitch.
This game has a) the stupidest premise, and b) the most awesome gameplay of any game that I've seen or played which has come out in the past 5 years. Play it, if you can get a copy!
You use linux because it works better for you, right? Why do you want to sheild management from this reality? Let them see your frustration, and keep making the case "this used to work, why is this so hard?".
Remember, managment doesn't like windows, they like money.
Why people are so willing to take the hit for other people, I'll never understand.
Okay, I use a USB drive enclosure on Linux to do my backups, and it works pretty well.
However, it took me a very long time to learn to set the 'sync' option in the mount options. USB writes much more slowly than a normal harddrive, and if sync isn't set, it is possible for the system to buffer all writes to the drive up to the point where it consumes most system memory, and the machine becomes unresponsive.
Perhaps this is fixed in 2.6; but it doesn't really matter. You are doing backups, the backup isn't done until it's all on the disk, so setting the sync option just means that your writes "seem" to take longer, and your unmount at the end seems faster. Without sync, you pay for the buffering in the unmount, which will hang while it finishes syncing the disk.
I was in the same boat as you. My hands hurt, and I was a professional programmer. I _knew_ I needed to learn to touch type, but I couldn't stop looking at the keyboard.
So I painted my keyboard black. The first week _sucked_, but by week 3, I was at 80% of where I had been before, and that 80% was touch, no looking at all.
Over the next month, I crept up to a bit faster than I had been, and that was good. But the real benifit to my speed was that, with my hands always in the correct location, all the control-Key and alt-Key keyboard shortcuts for my editor, my shell, and my web browser became available, and even second nature to me.
It is worth it.
There are a number of businesses which sell keyboards with blank keys, for use in typing classes; you can google for them. I reference this only as a means of showing that this has worked for others.
Try it, it _will_ work. Just take the plunge, accept the reduced short term productivity, and paint the keys black.
This is a bit heavy for someone 'wanting to learn some discrete math'.
The book isn't about graphs or proof or number theory. The book is about recurances, and methods of solving them. This brings in such subjects as discrete calculus, and floor and ceiling functions, but its NOT really a discrete math book.
Its a good book to read AFTER you get the basics of discrete though.
the following are very good books on proof and discrete math. Some of the titles are whimsical, but they are not toy books, they are very valuable.
"How To Prove It", "How To Solve It", "Induction and Analogy in Mathematics", and "Patterns of Plausible Inference".
However, it seems you are looking for a book to cram for a test in discrete math. Good luck, not going to find one. More so than any of the lower mathematics, discrete is the beginnings of higher logical analyisys, and you can not really 'cram' it. You have to really read the work, and really work the problems. It has to become part of you.
There seems to be this trend to blame difficulty in learning a subject on the books or the teachers. There are many, many things in the world that you are not smart enough to do; you need to accept this, and figure out what problems you can deal with.
I am not batman, I am not Johan Sebastian Bach, and I am not Richard Feynman, I have accepted this; perhaps you are not capable of Discrete Mathematics. If not, you need to leave CS, and go get in MIS or something, you will be happier.
The whole reason I love the device is that I need so fewer minutes of talk time. I'm doing the school thing, and anything which decouples communication is good. This thing has integrated wonderfully with the way I work, and I use many fewer minutes than before.
I've got a Sidekick. Its a new platform, there developer program isn't even open yet. The service is through T-Mobile.
But, the hardware itself is awesome. Its a JVM based system, and the hardware has:
Qwerty Keyboard, spin wheel, D-pad, USB, IrDA, a Phone plugin, 16M ram.
Current Apps:
Email, AIM, Calendar, Asteroids Clone, ToDo, Notes, A NICE Address Book, Web Browser, SMMS Messages, Phone, Tetris Clone.
And the "Desktop Interface" is unbelievable. You go to a website, login, and you have access to all of the application features of the device. It just keeps everything synced. And these aren't pared down interfaces on the web. It's a good email client, a good calendar, etc.
I agree, you don't really want a WYSIWYG editor for docbook, as that violates the Information/Representation sepperation.
But what would be _really_ useful would be a structured editor, which provided a good mapping at several levels (ala Mozilla's composer, where I can turn on and off tabs). The point of such an editor's representation would not be for final production, but to unambigously display the information's _structure_ to the user, and to facilitate manipulating that structure.
It would be like a very ugly word processor, where the tables would always have borders, etc.
If that editor was then linked with a set of generation tools, to make it EASY to genearte and view PostScript, HTML, XHTML, etc. productions, then you'd really have something. I'd use it, for sure, and I don't mind using docbook tags now.
In fact, would mozilla's composer be a good place to start with a docbook editor? Mozilla has good DOM tools, like the inspector, which loads XML DocBook files just fine.
Okay, many people in this thread are talking about the problems with optimizing for non-true pentiums, which are real problems. I'm not.
Red Hat DOES optimize several packages for specific cpus, namely the kernel and glibc. The real wins to be gained with code optimizations are in memcopy and the place where most of this actually happens are in the c library and the kernel. So if you look, there are i686 and athalon versions of the kernel, and i686 versions of glibc (not sure about athalon glibc).
Two layer SACDs are what I was thinking of, and I confused them with HDCDs.
/ /www.hdcd.com/about/index.html
More informaiton is available:
http://www.sonymusic.com/sacd/
http:
HDCD disks do not 'encode' 20 bits in 16 bits. This is impossible.
They work by having two media layers. The HD layer is translucent, and standard CD players will shine right through it, not see it, and read the old-style CD layer behind it, which is reflective. Newer, HD capapble CD players will read the translucent layer.
Your songs are thus recorded twice on the disk, once on each layer. Without hardware (read 'lasers') which can read the HD layer, it is impossible to access.
The coder's already at a major imparement, so there is no effective cost associated with 'switching' to dvorak left or right.
The current options for this person, until they are healed:
1) Learn to type one handed on a keymap designed for typing one handed.
2) Learn to type one handed on a keymap not designed for typing one handed.
Either way, _learning_ is involved. The full training time for a new keymap is shown to consistantly in research (which I should cite here, but won't) to be about a month to get to full speed.
Full speed is dependant on the collision rate and inherent transit distance of the keyboard, so and Dvorak's keymaps were designed to minimize both; thus 'full speed' is faster on the dvorak mappings.
And, having learned the new mapping, the coder now has an additional option when faced with reduced mobility, or an application which requires heavy mouse use, in the future.
There are two alternate Dvorak keymaps, in addition to the one that most people refer to (and I'm typing with now) when they say 'dvorak'. Dvorak developed a right and a left keymaping, which were designed for the physically handicapped.
These keymaps are set up to be minimal for english text with one hand. Give it a look-see, you might have them on your platform.
I've always been fond of using buckets instead of probing, and using Red Black trees for the buckets.
You cannot bring a common language dictionary into a discusion based upon technical jargon. Authentication and Authorization have very distinct, and _different_ meanings, in the dialect of english used for discussions in the world of software security. The meanings do not vary, but are constant in just about all the literature in the security field.
I suggest you read some of it.
Splash provides a distributed database using Spread. Apache-SSL has a plugable SSL conection cache layer, and there is a Splash implementation available. So, when a client connects to an Apache-SSL server in the pool, and the server checks the cache for SSL cache information, it is checking a DB it shares with all the other machines in the pool. This is exactly what is being requested for this Ask Slashdot.
This is a solved problem, check out the splash project http://anoncvs.aldigital.co.uk/splash/ for details.
Basically, it is an Apache module which uses the spread http://spread.org secure mesaging server to synchronizes ssl conection information.
I was developing a library which used glib for some of its module code, and while I was at it, I used it for the string functions as well.
Well, my library was crashing, hard, in the middle of a complex io cascade (read some file, and decided to load a module to read some other file based upon istructions in the first file, etc.) and I had checked and rechecked and rechecked the code. The code was correct, but the damn thing was crashing.
And it was crashing in some of my code, which was _perfect_ (I spent like 3 days on this, that code was checked, bracketed on both sides fully by debug statements, etc.).
Oh, and there had been perfect output from the program up to that point. So, getting another hacker to take a look at it, someone much more familiar with glib (One of the maintainers, actually) it was determined that using the %a format in my output was hosing not the print function, because glibc understood the C99 format character for printing floating and double point numbers with out loss, but in the glib code which estimated memory bounds on string operations before calling through to the underlying sprintf implementation. So the heap was getting corrupted, but it was printing out the correct value.
And then it died later, in my code, when it triped over that heap problem. Much profanity was involved.
* Ability to bring up my $EDITOR when typing in a textarea
Hell yes. I've been dying for this for a while. Not yet desperate enough to code it up, but I've looked at some of the problems that need to be solved to do it.
Once you spawn an $EDITOR, you have to block that textarea somehow, and you have to have a way to drop that block (what if the editor crashes?). This will have to be a visual clue on the textare.
Another problem is that the widget interface is very async, and this block introduces wierd statefulness into the textarea widget.
But I want this feature!