The truth is, there has never been a scientifically important coffee-table book. And neither Woflram nor Kruzweil will break this pattern.
Mainly because Wolfram's book would break the coffee table.
Back in the day we got in the Win 3 SDK. In a C example, to get out of 3 or 4 nested loops if there was an error condition there was a "goto wearefucked;" with the matching ":wearefucked;" at the end of the function.
Of course by the next version Microsoft had started working with IBM more and the goto had been changed.
Smalltalk is a great language. It will even mow the lawn, wash the dishes and feed the kids. Unfortunately it will also wash the lawn, feed the dishes and mow the kids.
I've read some places that one of the proposed remedies is to force Microsoft to release Office for Linux. It would be a *very* bad idea to force a company to develop and release software that they don't want to succeed. When Office for Linux comes out and it's a steaming pile of camel dung that makes Linux look bad, how can you prove that Microsoft did it on purpose instead of it being just buggy software, basically just version 1.0 problems?
I have a couple of questions for the people rewriting the Lisp in C++:
In his article Paul Graham stated that 20-25% of the code took advantage of Lisp's macros, and that it wasn't in there for fun, it was necessary. What are the people rewriting this in C++ using to duplicate the functionality that the macro facility provides?
If zero people understand Lisp, how the heck do you understand the Lisp code enough to rewrite it? What's going on over there? I'm imagining the scene in 2001 where all the homonids are standing around trying to understand the monolith.
"If PacMan had effected us as kids we'd be running around in dark rooms, munching pills and listening to electronic music"
Ever been to a rave?
I sat through a dog & pony show on this 2 yrs ago-
on
A New Web Image Format
·
· Score: 1
Put on by AT&T. Got a "compressed" T-shirt for my trouble (shrink-wrapped, about the size of 3 CD jewel cases stacked).
Anyway, things not mentioned elsewhere. The main "cool" things that the exhibitors mentioned was that DjVu would take an image that had graphics and text on it, distinguish between the two and on the same image compress the graphics part with wavelets and the text part with a fax-like compression whose name I don't remember.
The text compression worked by taking all the characters and deciding which ones were similar, store a small image for the character and just remember the placement of that character. For example "hi there" would store h, i, t, e, and r once then slap the images down in the correct places when decompressing. The nice part about this is that it works just as well on Chinese characters or other pictograph languages as it does on Western languages. Those of you wanting to scan and store Asian languages should give this a look.
As for PDF, if you're generating the PDF file yourself you're not going to do any better with DjVu since it just treats the PDF as a big image. I got the 1040 PDF off of the IRS site and ran a DjVu compressor on it and it came out about the same size since PDF's just have strings of characters for text.
I think this is a reaction to C# and.NET but not the reaction people may think about.
Assume for a moment that Sun isn't evil and greedy with Java but really has wanted to take it open source. Up until now they've been afraid that Microsoft would come in to whatever process that Sun no longer has veto power over and try to "corrupt" Java.
Now that Microsoft has C# and.NET to play with and has renounced Java then Sun can now safely open up Java. Microsoft can't change direction now and try to get back in the Java game without looking incredibly foolish.
I've read Alan Cooper's book "The Inmates are Running the Asylum", and a part of it that really irked me was Cooper's opinion of how much developers should be involved during the useability planning, i.e. none. He made developers out to be clueless philistines that should be left in the dark until the UI is done, so that they don't pollute the UI design with worries about implementation. ThoughtMill, on the other hand (from their website http://www.thoughtmill.com) seems to take more of a approach of working with the developers. Which method of collaboration do you think works better when designing applications?
A college near me (Georgia Tech) is offering a master's degree in Human/Computer Interaction. Do you think formal programs in HCI (and this one in particular) are worth the money and effort, or do people get at least almost as much benefit from reading and doing on their own?
Slashdot Mirror
The truth is, there has never been a scientifically important coffee-table book. And neither Woflram nor Kruzweil will break this pattern.
Mainly because Wolfram's book would break the coffee table.
Back in the day we got in the Win 3 SDK. In a C example, to get out of 3 or 4 nested loops if there was an error condition there was a "goto wearefucked;" with the matching ":wearefucked;" at the end of the function.
Of course by the next version Microsoft had started working with IBM more and the goto had been changed.
Bill Gates reads Slashdot, then it's a sentiment that bears repeating, yes.
Smalltalk is a great language. It will even mow the lawn, wash the dishes and feed the kids. Unfortunately it will also wash the lawn, feed the dishes and mow the kids.
I've read some places that one of the proposed remedies is to force Microsoft to release Office for Linux. It would be a *very* bad idea to force a company to develop and release software that they don't want to succeed. When Office for Linux comes out and it's a steaming pile of camel dung that makes Linux look bad, how can you prove that Microsoft did it on purpose instead of it being just buggy software, basically just version 1.0 problems?
If this was a two-guys-programming startup what makes you think there's a spec?
...and stop calling me Shirley.
I have a couple of questions for the people rewriting the Lisp in C++:
In his article Paul Graham stated that 20-25% of the code took advantage of Lisp's macros, and that it wasn't in there for fun, it was necessary. What are the people rewriting this in C++ using to duplicate the functionality that the macro facility provides?
If zero people understand Lisp, how the heck do you understand the Lisp code enough to rewrite it? What's going on over there? I'm imagining the scene in 2001 where all the homonids are standing around trying to understand the monolith.
"If PacMan had effected us as kids we'd be running around in dark rooms, munching pills and listening to electronic music"
Ever been to a rave?
Put on by AT&T. Got a "compressed" T-shirt for my trouble (shrink-wrapped, about the size of 3 CD jewel cases stacked).
Anyway, things not mentioned elsewhere. The main "cool" things that the exhibitors mentioned was that DjVu would take an image that had graphics and text on it, distinguish between the two and on the same image compress the graphics part with wavelets and the text part with a fax-like compression whose name I don't remember.
The text compression worked by taking all the characters and deciding which ones were similar, store a small image for the character and just remember the placement of that character. For example "hi there" would store h, i, t, e, and r once then slap the images down in the correct places when decompressing. The nice part about this is that it works just as well on Chinese characters or other pictograph languages as it does on Western languages. Those of you wanting to scan and store Asian languages should give this a look.
As for PDF, if you're generating the PDF file yourself you're not going to do any better with DjVu since it just treats the PDF as a big image. I got the 1040 PDF off of the IRS site and ran a DjVu compressor on it and it came out about the same size since PDF's just have strings of characters for text.
I think this is a reaction to C# and .NET but not the reaction people may think about.
Assume for a moment that Sun isn't evil and greedy with Java but really has wanted to take it open source. Up until now they've been afraid that Microsoft would come in to whatever process that Sun no longer has veto power over and try to "corrupt" Java.
Now that Microsoft has C# and .NET to play with and has renounced Java then Sun can now safely open up Java. Microsoft can't change direction now and try to get back in the Java game without looking incredibly foolish.
I've read Alan Cooper's book "The Inmates are Running the Asylum", and a part of it that really irked me was Cooper's opinion of how much developers should be involved during the useability planning, i.e. none. He made developers out to be clueless philistines that should be left in the dark until the UI is done, so that they don't pollute the UI design with worries about implementation. ThoughtMill, on the other hand (from their website http://www.thoughtmill.com) seems to take more of a approach of working with the developers. Which method of collaboration do you think works better when designing applications?
A college near me (Georgia Tech) is offering a master's degree in Human/Computer Interaction. Do you think formal programs in HCI (and this one in particular) are worth the money and effort, or do people get at least almost as much benefit from reading and doing on their own?