Actually, the DVDs do indeed have 60fps interlaced video. Progressive scan DVD players have to reverse the process to get back to the 24fps. That's why some progressive scan players produce better results than others....
Check out http://www.hometheaterhifi.com/volume_7_4/dvd-benc hmark-part-5-progressive-10-2000.html for one explanation.
Yep, the intro module was a section of the one in the new D&D basic game (which I picked up, being a sucker for anything which purports to teach D&D to newbies... one of these days, _one_ of the three or four basic sets I have around will catch my wife's interest).
I think that was a good thing... we had a few new players at our game who seemed to enjoy themselves and will probably go further in the hobby.
Bah. 2e was a feeble attempt to fix AD&D with some of the stuff from Mentzer D&D. D&D3e was just Rolemaster with different dice, finally kowtowing to the skill-based set. Bah! Give me races as classes or give me death!
Note that the Book of Erotic Fantasy _is_ still an Open Gaming License product. So while they could revoke d20 license, one can still create content under a lesser license.
Actually, D&D, AD&D, and D&D3e have about the same amount in common, rules-wise. D&D3e is closer to AD&D in feel, but there are still a number of changes.
However, they're all close enough that you'd probably have felt at home no matter what. The adventures run were fairly introductory, designed to be playable with people without much experience in the game.
Actually, I had a laptop stolen from my locked room, before school began when I was the only one there with a key besides the cleaning crew and RAs. So I'd recommend a locked cabinet as well.
Can you provide some examples? Most recently when the Mozilla/Firefox maintainers suddenly removed Direct Postscript printing without warning. (Since reverted.)
That said, most companies don't want great hackers. They want coders. That is interchangable people who can write code to produce a product.
Every company I've worked for wanted hackers. Unfortunately, they didn't realize it and couldn't try to hire for it, but when they got one in their clutches they hung on hard. So I'd say more companies than one would think actually appreciate hackers.
I do make a habit of avoiding truly bad management when I choose jobs, so maybe it's only places with smart managers (immediate managers, not the higher ups) who appreciate hackers, but I still believe there's a lot more out there who would enjoy employing one than know it.
As typically observed, I'm seeing the benchmarks take serious advantage of java's GC mechanism, whereby, they never pay the piper. With C++, the piper is constantly being paid.
That's not a bug in the tests, it's a feature.
The theory behind garbage collection isn't just that it allows the programmer to avoid the effort of watching when to delete things. It's that garbage collection can actually improve performance on certain workloads.
Forcing a garbage collection for every delete is completely unfair, since it does a full scan of memory, as opposed to just twiddling bits to free a single data value.
There's no memory leak for these benchmarks... both C++ and Java free all memory used when the process exits. Perhaps you'd prefer a longer-running test with lot of garbage generation (forcing gc to run at some point).
The best languages for text adventures are probably Inform and TADS, two languages specifically designed for that purpose.
Their advantages are a good runtime system including the parser, large development communities (well, as large as interactive fiction gets), lots of sample code to help learn, good documentation. The runtime is the key. Each has interpreters on lots and lots of platforms, and they take care of things so you don't have to. Things like undo, parsing commands, formatting, etc.
I think the only complaint people have with the X behavior is it "clobbering your copy buffer". I'd suggest the problem isn't that it shoves data there automatically, I'd suggest the problem is that it's a buffer, not a ring.
For example, emacs stores the last several things you've copied in its kill ring, and when pasting makes it very easy to paste earlier items from the ring. So I never have to worry about losing something I've pasted before. Emacs also gives named clipboard-like slots called registers, so if I have a few different things I don't want any copying to mess up, I can put them there and easily access.
There are clipboard managers for X which do somewhat the same thing, but the problem they have is that they try to exist outside of the application, and thus you have to move your mouse or invoke some strange key sequence to pull data out of them. Basically, making you work harder. Maybe it would be nice there were a standard interface such that applications could integrate it into their keybindings. Starting with "Paste" and "No, try an earlier one". MS Office also has something like this, but has the same usability problems as the X ones (with the additional caveat of being rather Office specific).
So that's my suggestion. Kill the clobbering behavior, and you kill the complaints.
The feature is that it stores its drawings in SVG, not any random XML format. That's a bit harder, and far more useful, than just using an XML format instead of a binary one.
Come on, you only needed to read just a sentence or two more of the article to get the explanation.
I have a 2GHz P4 with a gig of RAM on WinXP, and eclipse is no speed demon. I suppose if I had another gig of memory, it would probably be OK (despite only taking about 400M of memory).
It's acceptable for simple things like popping up menus, but way too much of my life is spent waiting for it to process CVS results, scan the objects to bring up the Open Type dialog, switch between perspectives, settle down after hitting a breakpoint.... It doesn't fall behind when I type, but that isn't saying much.
The MySQL people have their own view of the GPL which may not reflect reality. Notably, you can do practically anything you want with GPL'd software if you don't distribute it. Note that "Activities other than copying, distribution and modification are not covered by this License; they are outside its scope." Also note that 2b only specifies that the results need be under the GPL (thus necessitating GPL-compatible licenses) if you distribute it. Their view of whether making and using copies within an organization consists distribution conflicts with the FSF's view on that matter.
MySQL's revenue model is predicated on convincing as many people as possible that they need licenses, but you should really talk to your own lawyers. Or just pay, because you get support and help its development. But be aware that MySQL will try to convince you to buy licenses even when you may strictly not need them.
It bothers me that a show as bad as Enterprise is given money for season after season, given chance after chance so that it finally gets "a couple good episodes this season" while shows like Firefly are shown out of order, moved all over the schedule, preempted, and then left to rot despite being far better than Enterprise has managed after all this time.
I like to hope that if Paramount and UPN weren't sinking cash into Enterprise, that money might go to other, better genre shows. That maybe, we'd even get some original plots, rather than rehash after rehash of old Trek plots. Is it in the Trek bible that writers may only try halfway original plots twice a season? (And by which, I mean original to Trek. I don't mean to demand completely original ideas.)
Do mills really work that way, or do they cut it as needed. I'd think that repeated cuttings to get down to A4 would result in rather inaccurate paper.
Education is just like software development
on
The Flickering Mind
·
· Score: 1
Why do I say that? Because like software, nobody knows really knows how to make education work. But everyone has their pet theory. Education is intensely faddish. Look at open plan schools, phonics, and what to do with computers.
For studies which really look into this, as in software the key predictors aren't teaching methods, amount of funding, technology or anything like that, but the people involved. The biggest predictor is how parents feel about education, and right after that is the teacher. But influencing parents is hard, and finding good teachers is hard (especially in the numbers we need to staff schools), so schools go through serial flings with methods and ideas they hope will be silver bullets and address all their ills. Unsurprisingly, these silver bullets fail, and the school goes on to try another.
I'd suggest that realistically, there's very little a given school can do to improve. It's really a societal problem. Society needs to give up on quick fixes, convince people that education matters, and realize that it's the people that matter.
The problem with the "programmer analyst" view you describe is that it implies that software development is like a cow: you feed the cow the parameters of the problem, and the system falls to the ground at the other end. And the system you get is generally of the same quality.
After you're done with your top-down work of art, all those details you elided as "the time required to build the system" can mean the difference between solving the problem or watching the company go bankrupt due to poor choice of tools. Some languages will be better able to express the higher level design than others. Sure, you could describe the Mona Lisa to someone using Swahili over morse code, but showing them a picture of it will be a lot quicker. Different languages don't just shift around the essential complexity of programming, they also add their own complexities which have to be managed. And you have to face those complexities each time you modify the system. Remember that maintenance of systems is usually the highest cost in software development... and when it isn't, it's usually because the system was scrapped.
So, no, language choice will rarely mean that you can't solve a problem. But poor language choice can mean you're eventually unable to afford continuing to solve the ever-changing problem. Of course, by that time, your "programmer analyst" has long since moved on to other projects, leaving the people in the trenches to gripe about legacy systems.
I'd have to disagree. In general, you can't trust a 2-year programmer to write maintainable code, no matter how good the design.
In addition, unless you're implementing a very boring, commonplace system, the architect needs to have experience implementing it so that s/he learns the picky little unexpected details that end up changing the overall design. Otherwise, you end up with a pretty set of UML docs, and a codebase which is kludge upon kludge dealing with the real world.
Knuth has some good comments on this in, I believe, his Digital Typography book.
Slashdot Mirror
Actually, the DVDs do indeed have 60fps interlaced video. Progressive scan DVD players have to reverse the process to get back to the 24fps. That's why some progressive scan players produce better results than others....
c hmark-part-5-progressive-10-2000.html for one explanation.
Check out http://www.hometheaterhifi.com/volume_7_4/dvd-ben
Yep, the intro module was a section of the one in the new D&D basic game (which I picked up, being a sucker for anything which purports to teach D&D to newbies... one of these days, _one_ of the three or four basic sets I have around will catch my wife's interest).
I think that was a good thing... we had a few new players at our game who seemed to enjoy themselves and will probably go further in the hobby.
Bah. 2e was a feeble attempt to fix AD&D with some of the stuff from Mentzer D&D. D&D3e was just Rolemaster with different dice, finally kowtowing to the skill-based set. Bah! Give me races as classes or give me death!
Note that the Book of Erotic Fantasy _is_ still an Open Gaming License product. So while they could revoke d20 license, one can still create content under a lesser license.
Actually, D&D, AD&D, and D&D3e have about the same amount in common, rules-wise. D&D3e is closer to AD&D in feel, but there are still a number of changes.
However, they're all close enough that you'd probably have felt at home no matter what. The adventures run were fairly introductory, designed to be playable with people without much experience in the game.
Only if you're in park, in my experience.
Not me, but my wife is in the Atkinson lab at the med school....
Actually, I had a laptop stolen from my locked room, before school began when I was the only one there with a key besides the cleaning crew and RAs. So I'd recommend a locked cabinet as well.
Can you provide some examples?
Most recently when the Mozilla/Firefox maintainers suddenly removed Direct Postscript printing without warning. (Since reverted.)
That said, most companies don't want great hackers. They want coders. That is interchangable people who can write code to produce a product.
Every company I've worked for wanted hackers. Unfortunately, they didn't realize it and couldn't try to hire for it, but when they got one in their clutches they hung on hard. So I'd say more companies than one would think actually appreciate hackers.
I do make a habit of avoiding truly bad management when I choose jobs, so maybe it's only places with smart managers (immediate managers, not the higher ups) who appreciate hackers, but I still believe there's a lot more out there who would enjoy employing one than know it.
As typically observed, I'm seeing the benchmarks take serious advantage of java's GC mechanism, whereby, they never pay the piper. With C++, the piper is constantly being paid.
That's not a bug in the tests, it's a feature.
The theory behind garbage collection isn't just that it allows the programmer to avoid the effort of watching when to delete things. It's that garbage collection can actually improve performance on certain workloads.
Forcing a garbage collection for every delete is completely unfair, since it does a full scan of memory, as opposed to just twiddling bits to free a single data value.
There's no memory leak for these benchmarks... both C++ and Java free all memory used when the process exits. Perhaps you'd prefer a longer-running test with lot of garbage generation (forcing gc to run at some point).
The best languages for text adventures are probably Inform and TADS, two languages specifically designed for that purpose.
Their advantages are a good runtime system including the parser, large development communities (well, as large as interactive fiction gets), lots of sample code to help learn, good documentation. The runtime is the key. Each has interpreters on lots and lots of platforms, and they take care of things so you don't have to. Things like undo, parsing commands, formatting, etc.
Why don't your purchase orders just use a font where those letters can't be confused?
I think the only complaint people have with the X behavior is it "clobbering your copy buffer". I'd suggest the problem isn't that it shoves data there automatically, I'd suggest the problem is that it's a buffer, not a ring.
For example, emacs stores the last several things you've copied in its kill ring, and when pasting makes it very easy to paste earlier items from the ring. So I never have to worry about losing something I've pasted before. Emacs also gives named clipboard-like slots called registers, so if I have a few different things I don't want any copying to mess up, I can put them there and easily access.
There are clipboard managers for X which do somewhat the same thing, but the problem they have is that they try to exist outside of the application, and thus you have to move your mouse or invoke some strange key sequence to pull data out of them. Basically, making you work harder. Maybe it would be nice there were a standard interface such that applications could integrate it into their keybindings. Starting with "Paste" and "No, try an earlier one". MS Office also has something like this, but has the same usability problems as the X ones (with the additional caveat of being rather Office specific).
So that's my suggestion. Kill the clobbering behavior, and you kill the complaints.
The feature is that it stores its drawings in SVG, not any random XML format. That's a bit harder, and far more useful, than just using an XML format instead of a binary one.
Come on, you only needed to read just a sentence or two more of the article to get the explanation.
I have a 2GHz P4 with a gig of RAM on WinXP, and eclipse is no speed demon. I suppose if I had another gig of memory, it would probably be OK (despite only taking about 400M of memory).
It's acceptable for simple things like popping up menus, but way too much of my life is spent waiting for it to process CVS results, scan the objects to bring up the Open Type dialog, switch between perspectives, settle down after hitting a breakpoint.... It doesn't fall behind when I type, but that isn't saying much.
The MySQL people have their own view of the GPL which may not reflect reality. Notably, you can do practically anything you want with GPL'd software if you don't distribute it. Note that "Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope." Also note that 2b only specifies that the results need be under the GPL (thus necessitating GPL-compatible licenses) if you distribute it. Their view of whether making and using copies within an organization consists distribution conflicts with the FSF's view on that matter.
MySQL's revenue model is predicated on convincing as many people as possible that they need licenses, but you should really talk to your own lawyers. Or just pay, because you get support and help its development. But be aware that MySQL will try to convince you to buy licenses even when you may strictly not need them.
It bothers me that a show as bad as Enterprise is given money for season after season, given chance after chance so that it finally gets "a couple good episodes this season" while shows like Firefly are shown out of order, moved all over the schedule, preempted, and then left to rot despite being far better than Enterprise has managed after all this time.
I like to hope that if Paramount and UPN weren't sinking cash into Enterprise, that money might go to other, better genre shows. That maybe, we'd even get some original plots, rather than rehash after rehash of old Trek plots. Is it in the Trek bible that writers may only try halfway original plots twice a season? (And by which, I mean original to Trek. I don't mean to demand completely original ideas.)
You missed that he's running Windoes 2000 Professional, not XP. Service pack 2 for that was released May, 2001.
Do mills really work that way, or do they cut it as needed. I'd think that repeated cuttings to get down to A4 would result in rather inaccurate paper.
Why do I say that? Because like software, nobody knows really knows how to make education work. But everyone has their pet theory. Education is intensely faddish. Look at open plan schools, phonics, and what to do with computers.
For studies which really look into this, as in software the key predictors aren't teaching methods, amount of funding, technology or anything like that, but the people involved. The biggest predictor is how parents feel about education, and right after that is the teacher. But influencing parents is hard, and finding good teachers is hard (especially in the numbers we need to staff schools), so schools go through serial flings with methods and ideas they hope will be silver bullets and address all their ills. Unsurprisingly, these silver bullets fail, and the school goes on to try another.
I'd suggest that realistically, there's very little a given school can do to improve. It's really a societal problem. Society needs to give up on quick fixes, convince people that education matters, and realize that it's the people that matter.
The problem with the "programmer analyst" view you describe is that it implies that software development is like a cow: you feed the cow the parameters of the problem, and the system falls to the ground at the other end. And the system you get is generally of the same quality.
After you're done with your top-down work of art, all those details you elided as "the time required to build the system" can mean the difference between solving the problem or watching the company go bankrupt due to poor choice of tools. Some languages will be better able to express the higher level design than others. Sure, you could describe the Mona Lisa to someone using Swahili over morse code, but showing them a picture of it will be a lot quicker. Different languages don't just shift around the essential complexity of programming, they also add their own complexities which have to be managed. And you have to face those complexities each time you modify the system. Remember that maintenance of systems is usually the highest cost in software development... and when it isn't, it's usually because the system was scrapped.
So, no, language choice will rarely mean that you can't solve a problem. But poor language choice can mean you're eventually unable to afford continuing to solve the ever-changing problem. Of course, by that time, your "programmer analyst" has long since moved on to other projects, leaving the people in the trenches to gripe about legacy systems.
The question is whether the person with 10 years had 10 years of experience, or 1 year repeated ten times.
The latter programmer is common and not worth hiring.
I'd have to disagree. In general, you can't trust a 2-year programmer to write maintainable code, no matter how good the design.
In addition, unless you're implementing a very boring, commonplace system, the architect needs to have experience implementing it so that s/he learns the picky little unexpected details that end up changing the overall design. Otherwise, you end up with a pretty set of UML docs, and a codebase which is kludge upon kludge dealing with the real world.
Knuth has some good comments on this in, I believe, his Digital Typography book.
Oh, I think I still have WP6 lying around somewhere. That one was so buggy that WPO2k had TeX-level stability in comparison.