Seems to me like quantum chemists are just about the last people we'd need after the fall of civilization. Hell, at least computer scientists can get to work on the next Stonehenge. Of course if recovery from the fall involves massive population rebuilding, the pretty package can be her contribution to the cause.
If you're hardcore into leading-edge language stuff (and based on the functional programming references I figure you might be), Lambda the Ultimate'll give you a run for your money.
As someone who uses unit testing for application development, I'd have to wonder whether the cost of setting up such a system would be worth the benefits? One of the big challenges in automated testing is measuring behaviour to check correctness.
How do you check that a kernel driver is using hardware correctly? It's more or less difficult to measure the beavhiour externally depending on the system. Effectively you need to use mock/simulated interfaces -- in this case probably virtual machines -- but then what kind of code coverage would you get?
Personally, for the kernel, I'd guess the bang-for-buck of adding static checking would be higher than dynamic checking.
I lost track of the SCM status for the kernel, but my crude understanding is that the kernel developers rolled their own, git. Is git a fully-featured SCM? And if not, could using git be causing any additional workload that would be alleviated by using darcs or whatever?
And just for the record, what's the strategic plan for kernel SCM?
Actually, the theory of Puncutated Equilibrium would happily say that our rate of evolution is currently low. Perhaps it's not a coincidence that the signs of selection found in the article occured in a period of socioenvironmental change?
I kind of doubt that you'll have time to give even the most basic instruction in something like ISIS II, so any "teaching" you do will be taken advantage of by only the most determined and brightest*. If you've got such kids around, why waste the resource on playing with stuff you don't "know what to do with"? Instead, put them to work!
I homeschooled part of highschool and therefore had the time to get way in over my head in a part-time technical position. I learned tonnes while working for peanuts.
See if you can get students to get credit working for you (in BC students are required to do h hours of work experience for graduation and some special programs are designed to give them lots more). If not, then at least set it up as an official extracurricular. There's a small chance they might actually make your job easier and they'll learn more than doing anything else you could think of.
I'd say in a school more than almost any other organization you need to be concerned with intranet security. At least a couple of students will try and crack your systems and, as any/.er will tell you, you should not underestimate their competence. They have not signed employement agreements and will not pay attention to "rules". Sure, you might be able to bring disciplinary action to bear, but wouldn't it be better for all parties if you were proactive?
Therefore, I'd strongly recommend you Tear Down the Firewall if you can. Of course if all your workstations are Windows, that might not be practical, but you should at least consider it. (And just because all the boxes have their own firewalls doesn't mean you can't have a redundant one for the subnet.)
I was only half-serious, the McDonalds Rule of War has been disproved. That being said, your counterexamples can be responded to:
Wilhelminian Germany was in opposition to the Allies (particularly Britain) as a result of its attempts to catch up in the Empire game after its late formation as a nation.
Weimar Germany was suffering from economic failure when Hitler became Chancellor, it was the conversion to a war economy which fuelled the recovery.
The US is not as rich as it would like to be, particularly in oil.* Gulf War 2 could also be seen as a distraction for the increasingly inevitable non-rich status.
* Natural-resource-rich is probably more important than being rich in general. I assert that The Middle East is so screwed up because they sell all their resources rather than using them internally.
The generalization of the McDonald's Rule of War is that people who live in rich places don't fight wars. So the problem is that the Middle East is already too inhospitable, making it more so will only increase the violence (until everyone is dead, which I guess your plan would speed up). Terraforming the Middle East would be much more effective.
In Distraction, the Dutch declare war on the US so that the Americans will take the Netherlands over and deal with the flooding crisis by relocating the entire population. Incidentally, much of the book takes place in Louisiana, but without any flooding.
Personally, I think the Dutch would be better off in space.
Does anybody have any data on relative efficiency of these? I don't like bicycling so I'm interested in skateboards, scooters, and rollerblades as alternatives, but I'm not going to pick one just because it's cool...
Another solution would be to contract with a third party. There are companies that do nothing but provide you with email solutions. They can do this based on very strong commercial products. These companies themselves will host and run the hardware for you. They will do all of the configuration and deployment and maintenance and administration for you. I'm not familiar with their prices, though - but do look into it. The upgrades and crashes and migrations are their responsibility. Meeting QOS is their responsibility. They will deal with the commercial mailserver vendor(s) for you. They already have support contracts with them. All you do is tell them how big of a deployment you want and you're set.
As someone who used to support an email server for a living, I'd have to strongly second the recommendation for outsourcing. Email is not your main business and you don't have enough customers to afford learning enough to give your customers the service they'll demand. (Although maybe it'd be cheaper to make those mistakes now than when you have so many customers you have no choice but to do it yourself?)
95% of your customers would probably be happiest with addresses forwarded to GMail accounts that are configured to put the forwarded address in the From: field. Think about it.
My alma matter's capstone projects were supposed to have an external client, so what most of the groups did was find a client and then ask what they want. The lucky ones got to write a Postgresql frontend in PHP, the unlucky ones got to write an Access frontend in VB. I'm sure they all have wonderful jobs right now writing SQL Server frontends in C#.
I had plenty of stuff on my resume by fourth year and aspirations of grad school, so I figured I could afford to do a project that wouldn't include soul-sucking functionality. I found two partners who wanted to do something interesting and didn't feel like putting in the effort of finding a client. We came up with the general project idea first, and then shopped it around until we found someone who would sign off on it.
It was overly ambitious and we were lazy, so we only got to a working proof of concept. But the "clients" didn't actually need it, so they were willing to sign off on that. All the Profs were impressed with our design skills and vision, so I got an A and something interesting to talk about in interviews.
Yeah, you're right, because Ruby's Perl lineage brings the noise. The problem I always have when programming Python is that I can never guess how to do something in advance: sometimes the library provides a function, sometimes a method; sometimes the identifier is abbreviated, sometimes it's not. In Ruby I'm guessing the up-front learning curve is higher but there is more regularity in methodology.
What I really want is a language with the pureness of Ruby and the clarity of Python. It appears that the more evolved a language is, the more popular it will become, so we'll always be stuck with crud.:(
Why are you talking about Java? No one in this thread suggested Java was any good: it's not.
The problem with Python is that even if the primitives are objects, the language is designed as if they were made objects as an afterthought; eg: why do you write abs(-7) instead of (-7).abs?! (If Python were elegant, you wouldn't have global identifiers like abs().)
I would recommend Python, because I'm more familiar with it than Ruby. It has a clear, elegant syntax, and many concepts in it exist in other languages as well.
I'm also more familiar with Python than Ruby, but I've been meaning to get into Ruby because I perceive it as being more elegant than Ruby. It seems to me that amongst scripting languages there's a clear spectrum from evolved to designed.
PHP is the most evolved: OO is an extension, and the library tends to get refactored occasionally to work in all the features that were hacked in.
Python is in the middle: it's not pure OO and the library doesn't feel as designed from the top-down as it could be.
I'm hoping that Ruby, as it is more recent and purely OO is the most designed.
So Python is certainly better than PHP, but I think students might have an easier time learning something that's more elegant and uniform. Obviously Assembly and Scheme are great, but I think they're an awfully hard sell for anyone who isn't planning on getting a CompSci degree.
I'm not familiar with this area of research, but I am a CompSci grad student.
The experiment is basically three players controlling scouting, defense, and resource collection in a simple real-time strategy game. Waves of incoming units must be identified, and if from a hostile force, destroyed by the defender or avoided by the resource collector. The objective of the mission is to maximise resource collection while not using defensive forces against unidentified units. The players communicate using highly constrained communication (I didn't see a mention of any communication beyond one bit broadcast by the scout per incoming unit).
The control team is three humans. The experimental team is better described as a human-supported agent team than an agent-supported human team, as there is only one human advising the scout agent on incoming unit trajectories. Obviously, the agents outperformed the humans as the speed of the scenario increased.
I really hope someone versed in the literature can explain why these are results beyond the well-known lay use of bots in various games. The only interesting thing I can think of is that it reminds me of speculation I read somewhere that a massive AI like SkyNet (nice dept, eds!) might use organic, analog components like whales or obsessive-compulsive humans to do calculations that are not efficient in silico. The article is also valuable in that I know what video game undergrads will do for grad school.
Congratulations, you Get It:-)...Atom gives you more-or-less the same thing
Um, wow, thanks.:) But why support Atom if it isn't RDF? It's a shame the community got fractured in the first place, but it seems like the powers behind Atom + the powers behind RSS 1.0 are enough to tell Dave to stop whining.
Does RSS 1.0 have any problems that aren't inherint to RDF? And does RDF have any problems that aren't caused by expressing it in XML?
Sure, RSS 1.0 takes more work to understand up front, but once you get RDF isn't it just another schema? And these days, now that blog software has automatic feeds and there are aggregators available on every platform, how many humans actually need to read it?
That's because RSS 9.x (which should include "2.0") has a very similar design philosophy to Perl. It's not supposed to be elegant, it's supposed to do whatever needs to be done to work ASAP. It's also writer-friendly over reader-friendly.
Slashdot Mirror
So the connection'll be obvious if I watch the movie, then?
Seems to me like quantum chemists are just about the last people we'd need after the fall of civilization. Hell, at least computer scientists can get to work on the next Stonehenge. Of course if recovery from the fall involves massive population rebuilding, the pretty package can be her contribution to the cause.
If you're hardcore into leading-edge language stuff (and based on the functional programming references I figure you might be), Lambda the Ultimate'll give you a run for your money.
Can you spell it out?
As someone who uses unit testing for application development, I'd have to wonder whether the cost of setting up such a system would be worth the benefits? One of the big challenges in automated testing is measuring behaviour to check correctness.
How do you check that a kernel driver is using hardware correctly? It's more or less difficult to measure the beavhiour externally depending on the system. Effectively you need to use mock/simulated interfaces -- in this case probably virtual machines -- but then what kind of code coverage would you get?
Personally, for the kernel, I'd guess the bang-for-buck of adding static checking would be higher than dynamic checking.
I lost track of the SCM status for the kernel, but my crude understanding is that the kernel developers rolled their own, git. Is git a fully-featured SCM? And if not, could using git be causing any additional workload that would be alleviated by using darcs or whatever?
And just for the record, what's the strategic plan for kernel SCM?
Actually, the theory of Puncutated Equilibrium would happily say that our rate of evolution is currently low. Perhaps it's not a coincidence that the signs of selection found in the article occured in a period of socioenvironmental change?
I kind of doubt that you'll have time to give even the most basic instruction in something like ISIS II, so any "teaching" you do will be taken advantage of by only the most determined and brightest*. If you've got such kids around, why waste the resource on playing with stuff you don't "know what to do with"? Instead, put them to work!
I homeschooled part of highschool and therefore had the time to get way in over my head in a part-time technical position. I learned tonnes while working for peanuts.
See if you can get students to get credit working for you (in BC students are required to do h hours of work experience for graduation and some special programs are designed to give them lots more). If not, then at least set it up as an official extracurricular. There's a small chance they might actually make your job easier and they'll learn more than doing anything else you could think of.
I'd say in a school more than almost any other organization you need to be concerned with intranet security. At least a couple of students will try and crack your systems and, as any /.er will tell you, you should not underestimate their competence. They have not signed employement agreements and will not pay attention to "rules". Sure, you might be able to bring disciplinary action to bear, but wouldn't it be better for all parties if you were proactive?
Therefore, I'd strongly recommend you Tear Down the Firewall if you can. Of course if all your workstations are Windows, that might not be practical, but you should at least consider it. (And just because all the boxes have their own firewalls doesn't mean you can't have a redundant one for the subnet.)
Please, think of the children.
* Natural-resource-rich is probably more important than being rich in general. I assert that The Middle East is so screwed up because they sell all their resources rather than using them internally.
The generalization of the McDonald's Rule of War is that people who live in rich places don't fight wars. So the problem is that the Middle East is already too inhospitable, making it more so will only increase the violence (until everyone is dead, which I guess your plan would speed up). Terraforming the Middle East would be much more effective.
Maybe Haiti can take in the refugees? Provided Canada doesn't annex it first, of course...
In Distraction, the Dutch declare war on the US so that the Americans will take the Netherlands over and deal with the flooding crisis by relocating the entire population. Incidentally, much of the book takes place in Louisiana, but without any flooding.
Personally, I think the Dutch would be better off in space.
Especially if it's the son of Senator Roark! Or maybe they'd just let Kevin deal with it...
Does anybody have any data on relative efficiency of these? I don't like bicycling so I'm interested in skateboards, scooters, and rollerblades as alternatives, but I'm not going to pick one just because it's cool...
95% of your customers would probably be happiest with addresses forwarded to GMail accounts that are configured to put the forwarded address in the From: field. Think about it.
My alma matter's capstone projects were supposed to have an external client, so what most of the groups did was find a client and then ask what they want. The lucky ones got to write a Postgresql frontend in PHP, the unlucky ones got to write an Access frontend in VB. I'm sure they all have wonderful jobs right now writing SQL Server frontends in C#.
I had plenty of stuff on my resume by fourth year and aspirations of grad school, so I figured I could afford to do a project that wouldn't include soul-sucking functionality. I found two partners who wanted to do something interesting and didn't feel like putting in the effort of finding a client. We came up with the general project idea first, and then shopped it around until we found someone who would sign off on it.
It was overly ambitious and we were lazy, so we only got to a working proof of concept. But the "clients" didn't actually need it, so they were willing to sign off on that. All the Profs were impressed with our design skills and vision, so I got an A and something interesting to talk about in interviews.
Yeah, you're right, because Ruby's Perl lineage brings the noise. The problem I always have when programming Python is that I can never guess how to do something in advance: sometimes the library provides a function, sometimes a method; sometimes the identifier is abbreviated, sometimes it's not. In Ruby I'm guessing the up-front learning curve is higher but there is more regularity in methodology.
:(
What I really want is a language with the pureness of Ruby and the clarity of Python. It appears that the more evolved a language is, the more popular it will become, so we'll always be stuck with crud.
Why are you talking about Java? No one in this thread suggested Java was any good: it's not.
The problem with Python is that even if the primitives are objects, the language is designed as if they were made objects as an afterthought; eg: why do you write abs(-7) instead of (-7).abs?! (If Python were elegant, you wouldn't have global identifiers like abs().)
I'm also more familiar with Python than Ruby, but I've been meaning to get into Ruby because I perceive it as being more elegant than Ruby. It seems to me that amongst scripting languages there's a clear spectrum from evolved to designed.
So Python is certainly better than PHP, but I think students might have an easier time learning something that's more elegant and uniform. Obviously Assembly and Scheme are great, but I think they're an awfully hard sell for anyone who isn't planning on getting a CompSci degree.
I'm not familiar with this area of research, but I am a CompSci grad student.
The experiment is basically three players controlling scouting, defense, and resource collection in a simple real-time strategy game. Waves of incoming units must be identified, and if from a hostile force, destroyed by the defender or avoided by the resource collector. The objective of the mission is to maximise resource collection while not using defensive forces against unidentified units. The players communicate using highly constrained communication (I didn't see a mention of any communication beyond one bit broadcast by the scout per incoming unit).
The control team is three humans. The experimental team is better described as a human-supported agent team than an agent-supported human team, as there is only one human advising the scout agent on incoming unit trajectories. Obviously, the agents outperformed the humans as the speed of the scenario increased.
I really hope someone versed in the literature can explain why these are results beyond the well-known lay use of bots in various games. The only interesting thing I can think of is that it reminds me of speculation I read somewhere that a massive AI like SkyNet (nice dept, eds!) might use organic, analog components like whales or obsessive-compulsive humans to do calculations that are not efficient in silico. The article is also valuable in that I know what video game undergrads will do for grad school.
Does RSS 1.0 have any problems that aren't inherint to RDF? And does RDF have any problems that aren't caused by expressing it in XML?
Sure, RSS 1.0 takes more work to understand up front, but once you get RDF isn't it just another schema? And these days, now that blog software has automatic feeds and there are aggregators available on every platform, how many humans actually need to read it?
That's because RSS 9.x (which should include "2.0") has a very similar design philosophy to Perl. It's not supposed to be elegant, it's supposed to do whatever needs to be done to work ASAP. It's also writer-friendly over reader-friendly.
RSS 1.0 is slightly more complex but a gajillion times more elegant. It has actual standards for metadata.