I change a name of a member variable in C# via the property view and what does it do? it changes the name only where declared. how utterly useless. if i do the same thing in eclipse it changes the name everywhere it is used. and this is probably the most basic example of refactoring(eclipse makes refactoring a joy). after having the pleasure of using some ides with proper refactoring i simply feel hamstrung when using one of MSs environments. this is yet another example of them refusing to innovate - they have a browser that is all but useless due to no tabs or popup blocking and their IDEs are becoming increasingly useless when compared to those out there with some incredible innovation...
just like luxury car owners don't care whether or not everyone owns the same car they do - mercedes (or bmw, etc) owners don't secretly wish everyone drove a mercedes even if they believe deep down inside that mercedes are the best cars on earth...not that i would know anything about luxury cars or car/computer analogies:)
i have a $500 envision 17" LCD and have played a lot of tribes2, bf1942, aa, and most recently cod - i have never had any ghosting or bluring, and i enjoy very nice framerates with my 3ghz pentium4(ht) and various nvidia cards. also, i have a $25 ms optical mouse and prefer it over the other things i have tried:regular mice, touchpads, and even a $100 kennsington trackball. maybe i'm just not very sensitive to sensory input or something, but i do not have the same problems others appear to have with these things. further, i would also rate myself as a better than average player - usually finishing a level in the top 3-4 (even on servers overrun by clan players)...but everything is relative and you can probably kick my butt with your crt and wheel mouse;)
Of course oss is created all over the world, so take the following with a grain of salt since i can only speak to the american perspective.
i think your question is terribly american (or any prosperous nation). on the "ladder of needs", americans are at the top: shelter - check, food - check, other basic needs - check. at the top of the ladder are the real worm fuzzies like self actualization. basically, we've got everything we need and most things we want, so we have a lot of time to sit around thinking about the "meaning of life", etc. your questions is aimed at a country where many are in severe poverty. ask again in a couple or three generations (assuming their economy survives - hopefully it will). in your spare time you use computers - in large portions of india spare time is spent trying to stay alive.
The problem isn't specific to India - we had it here during the "boom". Tech jobs were plentiful so the universities started churning out tech grads. Some were good, but many were not - they were just there because it was easy money (or whatever). There was a lot a "chaff" in the boom (idiots if you will) - the bust has helped to remove these people (in my experience). In India I see the same problem multiplied by an order of magnitude: there are a billion people, and many are in poverty, so getting into tech has an incredible draw - much, much more than in the states during the boom ($10000 is upper, upper middle class). Indians are smart people, but there is a lot of chaff...
Actually, I'm pretty sure you sue where you're more likely to win...given things like the judges, that state's laws against such things, the courts track record, etc...Colorado is a very conservative and religious state, maybe they think that will be an advantage over a liberal state like california- maybe they think the CO judge will cringe at the very word "penis" - of course i'm just speculating here;)
Re:A function of the human brain?
on
The Golden Ratio
·
· Score: 1
growing up i could remember a whole lot more phone numbers than now. the reason was i was familiar with the area codes and exchanges. mentally i would group people of the same area code together and then do the same with the exchanges. luckily "area" codes are geographic so i didn't "remember" the three digits of the area code, just that they lived in the north west side of town. similarly most neighborhoods had similar exchanges, so when i could, i would just remember what subdivision they lived in. so, in order to remember a ten digit number, on average i'd only have to remember 5 things - their neighborhood plus the four digits on the end of their number (i could figure out their area code usually by what side of town their neighborhood was on). i don't do this much anymore because i'm less fimiliar with this town and my mobile phone remembers it all for me.
i'm not trying to prove you wrong (i don't care one way or the other what the limit of human memory is), but i don't think 10-digit phone numbers disproves the previous guy. i do think the previous guy was over-simplifying. as i've already explained, my mind tends to group things hierarchically, so maybe this "six" limit is the maximum depth to a leaf node...but who knows...
it is for this reason that i stopped using linux a few years ago....you see, i am more elite than you...google doesn't even know the os i currently use.
i hate patents as much as the next guy, but i do believe the interface on my ipod (touchpad/no moving parts) is patentable. i say this not knowing (or caring for that matter) which parts are actually patented. there are two things i think are innovative. first, the touch wheel - its great. like you said earlier models had a wheel. second, the software that works with the wheel - its great because it doesn't just track position/movement, but acceleration as well. maybe its obvious that that is the way it should work, but frankly it surprised me when i realized how it worked. i started scrolling through the list of several thousand songs i'd just copied to it and started getting mad because it was starting to look like it was going to take all day to get to the bottom of the list (which is where the song i was looking for was). i was ready to take it back to the store. in frustration i started moving my finger faster and faster and then i noticed i was at the bottom of the list. needless to say i was delighted. so, i don't know if a scroll wheel is patentable or if acceleration tracking is patentable, but the combination of them should be (imo). also, remember the old adage "all ideas are obvious once seen." indeed, after i realized how the ipod worked i thought to myself "of course that's how it should work" - in fact i might have come up with that same design/implementation myself (after some r&d), but as they say "necessity is the mother of invention"...one final thought. when did apple patent the wheel? i know that there have been quite a few players (even pdas) on the market for a long time. if the ipods design is so obvious, then why didn't anyone else patent it first. because it is *so* obvious? then why didn't someone release a player or pda that made use of it? if apple hadn't patented it what would the timeline look like? i suggest it looks like 1) everyone is competing with a bunch of different mp3 player designs (the market is searching for an innovative solution) 2) apple release the ipod with an unpatented intface, 3) it starts becoming the most popular player (as it has today, even being more expensive and very proprietary) 4) all/most new players clone the ipods interface (as they're trying to do today) 5) apple looks back and thinks "damn we should have patented the interface". like i said, i hate patents (usually software/algorithm ones), but i am not sure about this one...
Everything can be spun as a conspiracy. Dan Geer who is quoted several times throughout the article was the guy fired for the report he co-wrote that slammed microsoft. Most stuff I've read by security pundits is, at its source, based on the idea that software makers should be held accountable for what they sell. Accepting the liablity is a bitter pill, but if MS chooses to swallow it in order to make their product look more attractive than open source, well i think its perfect example of adding value to your product.
Why not send prisoners that already have a life sentence? Find four that might not kill eachother (probably the biggest problem - they're more likely to kill eachother in transit), and give them some science and medical training. It doesn't have to be a lot because you'll always have the real "brains" here on earth guiding them through anything tricky. They'd basically be robots, but better.
Oh, come on - we were already being pretty far-fetched. Send twelve and we can call them the dirty dozen.
"select" is actually very useful for the case where you might want more than *one* thread for a connection. in the case where a client is properly multithreaded, you can continue to receive requests even though previous requests haven't yet completed. the ideal solution is to have a thread pool, where a thread is assigned to incoming requests, not just an incoming connection. in the latter case you can get the situation where you may have 10 threads (one per connection), but only a single connection is getting hammered(and all client threads are being forced to wait on a single server thread - which is probably io bound). in the former case, you'll have 10 threads in a pool with no ties to any particular thread. when a single connection starts getting hammered, more threads can be assigned to serivce that single connection. Java's nio (and select) are perfect for implementing such a system. so, while select can be used as you suggest in order to decrease throughput, it can also be used as i suggest to increase it dramatically.
I am no expert on economics, so i have a couple of questions:
Everyone talks about the leveling of the playing field, but they seem to focus on india's side going up, and almost never on the us's going down. i am talking about standard of living. what happens as the american standard of living declines (as the rest of the world rises)? they keep talking about new jobs being created "higher up the foodchain", but we can't all be CEOs, doctors, lawyers, etc. The American middle class is going to start having less money, because it will be going overseas. We'll have less, we'll spend less. Luckily everythings is now being made much more cheaply overseas, so things will cost less(right?). It seems that even real estate would suffer as people had less to pay for homes. Also, why go to college if it isn't going to help me get a job? I know a lot of this is a slippery slope, but I'd like to get the opinions of some of the economists out there...
I agree with most of what you say. I do, however, see a couple good things about PEs:
1. EITs have to prove themselves, its not enough to know the theory
2. EITs must endure a period of on-the-job-training where a mentor oversees their work
Where i started after college it was assumed that college was enough to completely equip me with everything i needed to be a successful software developer. Not only can i say from experience that this was not true for myself, but through observation I can say that it was not enough for many others. Further, I can say that most veteran developers I have come in contact with in the industry do not know how to develope robust/flexible/extensible/secure/internationalize d solutions (or not in a manner that would represent a "best practice"). These veteran developers are very brilliant people (wrt. CS theory and algorithms), but they do not know how to develope software, they know how to hack...
I guess all I am saying is that there is a significant amount of learning even after all the theory is understood and largely we assume this will be learned on our own sort of trial-by-fire. The best case scenario is that these new developers will only hack out crappy software until they learn some industry best practices (supposedly on their own). The worst case is that they'll hack out crappy software for many, many years to come...i've experience the former and seen the latter in many, many "veterans"...
I think required mentorship/probationary period for newbies and some sort of re-certification for "veterans" would go a long to improving software quality...
as far as i can tell this solution requires the recipient to also perform the same computation. if this assumption is incorrect please let me know. if it is correct, then this solution is as stupid as other have suggested. What is to stop a spammer from just attaching fake headers to their email? if it is just some sort of hash, then a simple random number generator could be used to generate fake headers that you'll have to waste time just figuring out that they're fake... (this is a common attack wrt to other security systems). the spam gets through to those not using the new system, and frustrates those that are until they quit.
someone above pointed out the real problem imho. its my machine having to check for this hash. this process will take the same amount of processing power as it does for the sender. the difference is there is still thousands of spammers with thousands of trojaned boxes or server farms. in the short term, i would imagine the spammers would just continue to spam as much as they possibly can assuming that we'll eventually get tired of wasting cpu time calculating a hash.
Other engineers are licensed professionals, why not software engineers? I think as the public continues to experience the adverse affects of crappy software, we actually have a shot at selling ourselves as professionals. This might mean more education or dicipline and it may raise the bar to a level where many of the existing developers can't attain licensing, but so what - they shouldn't be writing code anyway. I think licensing would be a great solution to the continued outsourcing of developers, and yes I'll continue to say so as long as slashdot keeps posting stories about it:)
it was decent scifi and so a lot of slashdot readers were instant fans, myself included. the universe was not the sanitary one of star trek nor the mystical one of starwars. it was a universe where technology hadn't been humanities savior, and there was enough of it to see that it probably never would. the universe had limited resources: fuel, food, etc. the universe required the characters to be people who did what it took to survive. this allowed for them to be interesting characters without requiring they be jedi or android. the firefly's crew was made up of prositutes, priests, and even the captain killed at least one "bad guy" in cold blood. i didn't watch every episode, but the few i caught i really enjoyed.
when i think about how unsuccessful we've been at coping with software development in the internet age i wonder if the skills are truly commodiities. i think the current market conditions treat them as if they were, but markets change. for example, most people have the perception that sofware is generally very low quality, but in the past this low quality has not had the expensive consequences that are now being realized due to the many security issues in software. as consumers begin realizing that software *must* be done better they'll begin pettitioning the industry and politicians for protection. One option is professional licensing of software developers. Several states have attempted passing legislation that requires software developer licensing, but there wasn't any industry support. Today, the multi-national corporations may still abhor the idea, but the consumers might now find an ally in the software developers who are scared for their jobs as they continue to move oversees. i for one for would look forward to having to be licensed.
One interesting and perhaps obvious aspect of this whole argument is: we don't need to upgrade as often as MS finances would require us to. This is why free software is successful. Yes "it's ready when it's ready", and that soon enough in almost all cases. MS of course realizes this and that's why last week Gates said longhorn will be done when it is done. They can't compete with open source and still meet rigid ship dates - something has to give.
Slashdot Mirror
i have always heard this trend referred to as outsourcing/offshoring. the above is an example of offshoring.
I change a name of a member variable in C# via the property view and what does it do? it changes the name only where declared. how utterly useless. if i do the same thing in eclipse it changes the name everywhere it is used. and this is probably the most basic example of refactoring(eclipse makes refactoring a joy). after having the pleasure of using some ides with proper refactoring i simply feel hamstrung when using one of MSs environments. this is yet another example of them refusing to innovate - they have a browser that is all but useless due to no tabs or popup blocking and their IDEs are becoming increasingly useless when compared to those out there with some incredible innovation...
just like luxury car owners don't care whether or not everyone owns the same car they do - mercedes (or bmw, etc) owners don't secretly wish everyone drove a mercedes even if they believe deep down inside that mercedes are the best cars on earth...not that i would know anything about luxury cars or car/computer analogies :)
f9, f11? use corners....expose is awesome
i have a $500 envision 17" LCD and have played a lot of tribes2, bf1942, aa, and most recently cod - i have never had any ghosting or bluring, and i enjoy very nice framerates with my 3ghz pentium4(ht) and various nvidia cards. also, i have a $25 ms optical mouse and prefer it over the other things i have tried:regular mice, touchpads, and even a $100 kennsington trackball. maybe i'm just not very sensitive to sensory input or something, but i do not have the same problems others appear to have with these things. further, i would also rate myself as a better than average player - usually finishing a level in the top 3-4 (even on servers overrun by clan players)...but everything is relative and you can probably kick my butt with your crt and wheel mouse ;)
Of course oss is created all over the world, so take the following with a grain of salt since i can only speak to the american perspective.
i think your question is terribly american (or any prosperous nation). on the "ladder of needs", americans are at the top: shelter - check, food - check, other basic needs - check. at the top of the ladder are the real worm fuzzies like self actualization. basically, we've got everything we need and most things we want, so we have a lot of time to sit around thinking about the "meaning of life", etc. your questions is aimed at a country where many are in severe poverty. ask again in a couple or three generations (assuming their economy survives - hopefully it will). in your spare time you use computers - in large portions of india spare time is spent trying to stay alive.
The problem isn't specific to India - we had it here during the "boom". Tech jobs were plentiful so the universities started churning out tech grads. Some were good, but many were not - they were just there because it was easy money (or whatever). There was a lot a "chaff" in the boom (idiots if you will) - the bust has helped to remove these people (in my experience). In India I see the same problem multiplied by an order of magnitude: there are a billion people, and many are in poverty, so getting into tech has an incredible draw - much, much more than in the states during the boom ($10000 is upper, upper middle class). Indians are smart people, but there is a lot of chaff...
Actually, I'm pretty sure you sue where you're more likely to win...given things like the judges, that state's laws against such things, the courts track record, etc...Colorado is a very conservative and religious state, maybe they think that will be an advantage over a liberal state like california- maybe they think the CO judge will cringe at the very word "penis" - of course i'm just speculating here ;)
growing up i could remember a whole lot more phone numbers than now. the reason was i was familiar with the area codes and exchanges. mentally i would group people of the same area code together and then do the same with the exchanges. luckily "area" codes are geographic so i didn't "remember" the three digits of the area code, just that they lived in the north west side of town. similarly most neighborhoods had similar exchanges, so when i could, i would just remember what subdivision they lived in. so, in order to remember a ten digit number, on average i'd only have to remember 5 things - their neighborhood plus the four digits on the end of their number (i could figure out their area code usually by what side of town their neighborhood was on). i don't do this much anymore because i'm less fimiliar with this town and my mobile phone remembers it all for me.
i'm not trying to prove you wrong (i don't care one way or the other what the limit of human memory is), but i don't think 10-digit phone numbers disproves the previous guy. i do think the previous guy was over-simplifying. as i've already explained, my mind tends to group things hierarchically, so maybe this "six" limit is the maximum depth to a leaf node...but who knows...
it is for this reason that i stopped using linux a few years ago....you see, i am more elite than you...google doesn't even know the os i currently use.
i hate patents as much as the next guy, but i do believe the interface on my ipod (touchpad/no moving parts) is patentable. i say this not knowing (or caring for that matter) which parts are actually patented. there are two things i think are innovative. first, the touch wheel - its great. like you said earlier models had a wheel. second, the software that works with the wheel - its great because it doesn't just track position/movement, but acceleration as well. maybe its obvious that that is the way it should work, but frankly it surprised me when i realized how it worked. i started scrolling through the list of several thousand songs i'd just copied to it and started getting mad because it was starting to look like it was going to take all day to get to the bottom of the list (which is where the song i was looking for was). i was ready to take it back to the store. in frustration i started moving my finger faster and faster and then i noticed i was at the bottom of the list. needless to say i was delighted. so, i don't know if a scroll wheel is patentable or if acceleration tracking is patentable, but the combination of them should be (imo). also, remember the old adage "all ideas are obvious once seen." indeed, after i realized how the ipod worked i thought to myself "of course that's how it should work" - in fact i might have come up with that same design/implementation myself (after some r&d), but as they say "necessity is the mother of invention"...one final thought. when did apple patent the wheel? i know that there have been quite a few players (even pdas) on the market for a long time. if the ipods design is so obvious, then why didn't anyone else patent it first. because it is *so* obvious? then why didn't someone release a player or pda that made use of it? if apple hadn't patented it what would the timeline look like? i suggest it looks like 1) everyone is competing with a bunch of different mp3 player designs (the market is searching for an innovative solution) 2) apple release the ipod with an unpatented intface, 3) it starts becoming the most popular player (as it has today, even being more expensive and very proprietary) 4) all/most new players clone the ipods interface (as they're trying to do today) 5) apple looks back and thinks "damn we should have patented the interface". like i said, i hate patents (usually software/algorithm ones), but i am not sure about this one...
Everything can be spun as a conspiracy. Dan Geer who is quoted several times throughout the article was the guy fired for the report he co-wrote that slammed microsoft. Most stuff I've read by security pundits is, at its source, based on the idea that software makers should be held accountable for what they sell. Accepting the liablity is a bitter pill, but if MS chooses to swallow it in order to make their product look more attractive than open source, well i think its perfect example of adding value to your product.
Why not send prisoners that already have a life sentence? Find four that might not kill eachother (probably the biggest problem - they're more likely to kill eachother in transit), and give them some science and medical training. It doesn't have to be a lot because you'll always have the real "brains" here on earth guiding them through anything tricky. They'd basically be robots, but better. Oh, come on - we were already being pretty far-fetched. Send twelve and we can call them the dirty dozen.
"hyperactive boosterism"...yea, i hate those kinds of websites...
"select" is actually very useful for the case where you might want more than *one* thread for a connection. in the case where a client is properly multithreaded, you can continue to receive requests even though previous requests haven't yet completed. the ideal solution is to have a thread pool, where a thread is assigned to incoming requests, not just an incoming connection. in the latter case you can get the situation where you may have 10 threads (one per connection), but only a single connection is getting hammered(and all client threads are being forced to wait on a single server thread - which is probably io bound). in the former case, you'll have 10 threads in a pool with no ties to any particular thread. when a single connection starts getting hammered, more threads can be assigned to serivce that single connection. Java's nio (and select) are perfect for implementing such a system. so, while select can be used as you suggest in order to decrease throughput, it can also be used as i suggest to increase it dramatically.
I am no expert on economics, so i have a couple of questions:
Everyone talks about the leveling of the playing field, but they seem to focus on india's side going up, and almost never on the us's going down. i am talking about standard of living. what happens as the american standard of living declines (as the rest of the world rises)? they keep talking about new jobs being created "higher up the foodchain", but we can't all be CEOs, doctors, lawyers, etc. The American middle class is going to start having less money, because it will be going overseas. We'll have less, we'll spend less. Luckily everythings is now being made much more cheaply overseas, so things will cost less(right?). It seems that even real estate would suffer as people had less to pay for homes. Also, why go to college if it isn't going to help me get a job? I know a lot of this is a slippery slope, but I'd like to get the opinions of some of the economists out there...
I agree with most of what you say. I do, however, see a couple good things about PEs:e d solutions (or not in a manner that would represent a "best practice"). These veteran developers are very brilliant people (wrt. CS theory and algorithms), but they do not know how to develope software, they know how to hack...
1. EITs have to prove themselves, its not enough to know the theory
2. EITs must endure a period of on-the-job-training where a mentor oversees their work
Where i started after college it was assumed that college was enough to completely equip me with everything i needed to be a successful software developer. Not only can i say from experience that this was not true for myself, but through observation I can say that it was not enough for many others. Further, I can say that most veteran developers I have come in contact with in the industry do not know how to develope robust/flexible/extensible/secure/internationaliz
I guess all I am saying is that there is a significant amount of learning even after all the theory is understood and largely we assume this will be learned on our own sort of trial-by-fire. The best case scenario is that these new developers will only hack out crappy software until they learn some industry best practices (supposedly on their own). The worst case is that they'll hack out crappy software for many, many years to come...i've experience the former and seen the latter in many, many "veterans"...
I think required mentorship/probationary period for newbies and some sort of re-certification for "veterans" would go a long to improving software quality...
as far as i can tell this solution requires the recipient to also perform the same computation. if this assumption is incorrect please let me know. if it is correct, then this solution is as stupid as other have suggested. What is to stop a spammer from just attaching fake headers to their email? if it is just some sort of hash, then a simple random number generator could be used to generate fake headers that you'll have to waste time just figuring out that they're fake... (this is a common attack wrt to other security systems). the spam gets through to those not using the new system, and frustrates those that are until they quit.
someone above pointed out the real problem imho. its my machine having to check for this hash. this process will take the same amount of processing power as it does for the sender. the difference is there is still thousands of spammers with thousands of trojaned boxes or server farms. in the short term, i would imagine the spammers would just continue to spam as much as they possibly can assuming that we'll eventually get tired of wasting cpu time calculating a hash.
Other engineers are licensed professionals, why not software engineers? I think as the public continues to experience the adverse affects of crappy software, we actually have a shot at selling ourselves as professionals. This might mean more education or dicipline and it may raise the bar to a level where many of the existing developers can't attain licensing, but so what - they shouldn't be writing code anyway. I think licensing would be a great solution to the continued outsourcing of developers, and yes I'll continue to say so as long as slashdot keeps posting stories about it :)
it was decent scifi and so a lot of slashdot readers were instant fans, myself included. the universe was not the sanitary one of star trek nor the mystical one of starwars. it was a universe where technology hadn't been humanities savior, and there was enough of it to see that it probably never would. the universe had limited resources: fuel, food, etc. the universe required the characters to be people who did what it took to survive. this allowed for them to be interesting characters without requiring they be jedi or android. the firefly's crew was made up of prositutes, priests, and even the captain killed at least one "bad guy" in cold blood. i didn't watch every episode, but the few i caught i really enjoyed.
when i think about how unsuccessful we've been at coping with software development in the internet age i wonder if the skills are truly commodiities. i think the current market conditions treat them as if they were, but markets change. for example, most people have the perception that sofware is generally very low quality, but in the past this low quality has not had the expensive consequences that are now being realized due to the many security issues in software. as consumers begin realizing that software *must* be done better they'll begin pettitioning the industry and politicians for protection. One option is professional licensing of software developers. Several states have attempted passing legislation that requires software developer licensing, but there wasn't any industry support. Today, the multi-national corporations may still abhor the idea, but the consumers might now find an ally in the software developers who are scared for their jobs as they continue to move oversees. i for one for would look forward to having to be licensed.
no we haven't, we've just become more specific - want we want is to further the advancement of American society.
thanks to globalization, even if you move to the other side of the world, you'll still be buying food cloths, etc from the same companies...
One interesting and perhaps obvious aspect of this whole argument is: we don't need to upgrade as often as MS finances would require us to. This is why free software is successful. Yes "it's ready when it's ready", and that soon enough in almost all cases. MS of course realizes this and that's why last week Gates said longhorn will be done when it is done. They can't compete with open source and still meet rigid ship dates - something has to give.