they are just echoing the times (speaking of, when did the nytimes start doing this, its very annoying..puff peice on street hot dog vendors includes three quotes at the end)
the invisible hand is the new messiah.
your quote from the article is extremely relevant. it shows that god has sided with apple, so they have the moral imperative. how else is a poor reader supossed to be able to judge such a complicated issue?
i disagree. when really old and stable code is too crufty to extend into a new domain, its a great time to refactor it. take it piece by piece and do fairly extensive regressions. if something breaks, you should have an excellent idea where it might be.
oh yeah, i forgot to mention (i had to look to see if it had already been published)...the other cool part is the global architecture. that is there is a large switching fabric connecting all the pims together. aside from the normal reads and writes, it also supports parcels, which is actually a whole migratory thread state. it just gets put in the run queue at the target.
so if there is any spacial locality to be exploited, you can move the thread rather than the data. because this is MTA style you would expect the overall throughput gains to be modest, but you have just reduced your dependency on network throughput correspondingly.
yeah, they're still working on it. i dont know alot of the earlier work, so i dont know how much of this is novel.
they are really fixated on the physical aspects of the memory arrays and building an effective cpu architecture around the context of dram rows (i.e., a thread context is a row, including registers, etc)
so its a little more than just the pin count and interface electronics argument.
his work these days centers around efficiencies of access gained by putting the dram and processing elements on the same die. partially removing the serialization associated with the standard synchronous memory interface. The architecture also plans on using MTA-style threads to hide latency and increase concurrency.
oh yes. definately. while the article is grossly sensationalistic i suspect that if you were the us dod you'd have some reason to be concerned. not just because of the bad press.
while all actually secret activities aren't on the net, there is alot you can infer from the other random office crap lying around. contracts, billing, travel, social contacts, supply ordering, etc. enough that you've probably saved yourself some time in deciding what to go after with your real intelligence resources
you wank. 'gcc compatible', yes great. you extend a language and then force all sort of people to deal with the resultant garbage. there is a standard, use it. the gcc extentions can be conveient, but they don't really change the scope of the language.
and the intel compiler can really scream over gcc in some applications, and its almost never worse. its a good compiler for a crappy instruction set. why would you not want to use it? because you're a religous fanatic?
its not illegal. its just that the agencies responsible for handing out research money, part of the executive branch, have made this their policy.
think about how many state laws have been passed under the threat of witholding highway fuding.
there are lots of ways the three branches can push something in and of themselves. they can choose to block something (veto, fillibuster, amendment), but it costs.
you're holding sid karin up as an example of how to fight waste and cronyism in funding for scientific research? and then the doe supercomputing efforts?
i'd never heard the name of this shell before. what a terrible shame. one could actually make a lovely shell out of monads. in fact that sort of what they are...but instead they apparently just used the name, because it was...what, cool? i've always just hated them from a distance, but this is actually offensive.
yes, i never used an LMI...i learned to program for real on a symbolics and was extremely disappointed when i had to start using a sun.
there is no question that today the barriers to entry make it infeasible, but i still disagree that there is anything fundamental that makes it so. lisp is actually easier to understand in the limit than something like perl.
i'll give you the commodity pc, the value was in the environment, not hardware tagging.
the only real complaint i have about your statements is that you make it seem inevitable, rather than historical accident and social hysteresis.
here you are wrong. did you ever use one? the symbolics was by far the most usable and productive machine at the time.
its primary strength was exactly that you could toss together something pretty nice in basically no time at all.
dont you think it might have something to do with marketing, barriers to entry, and culture? there is nothing fundamentally economic about the failure of symbolics (except the barrier to entry bit, the volume was so low that they were very expensive machines)
i dont use gentoo...i dont even use linux very much...but what i dont understand is that you "open source" people dont seem to understand the intrinsic value of *having the source*. do you realize how utterly ridiculous it is that if i want to look at ifconfig, i have to dig around and find that its in some package called netutils, then find out which version is installed, then get the release patches and apply them.
how did you people get so fixated on binaries. its all about the source.
the 13 colonies had the highest standard of living in the world at the time? from my understanding it was less than that afforded to the dominated states of the roman empire. what could you possibly base such an assertion upon?
forget funding, how could they possibly come up with that much content on internet security. 'in other late breaking news, the internet still lacks a decent pki, and script kiddies run rampant'
as a 'senior' person who has done hiring before there are jobs like this. as other people have said, dont look at the 'senior' jobs. they really are looking with people who have experience in the area, can set direction, who know how to deal with large projects, etc.
every place i have been we hired people out of school. while its hard to interview these people, the primary thing i've always been interested in is projects they have done, either outside of school or research projects at school. emphasize your contribution rather than the project itself. you will almost certainly be doing coding questions. the tools you learned in school (algorithms, complexity, analysis) aren't useless in the real world. trot them out.
you will be expected to contribute, but will be giving you smaller projects that are less critical path, tracking your progress more closely, and hopefully be giving you tactical advice. the worst places for entry level are places where no one has time to deal with you, they stick you in a cube, give you some vauge description of your job and ignore you, then lay you off because you aren't performing well. if that happens, demand attention. ask them often what you should be doing and tell them where you are stuck.
look for and apply for things you have some experience or interest in. its easier to get a job doing something you've done, so without some effort once you become a (sysytems, ui, tools, whatever) it will stick with you. in that vein dont beleive the 'we really need qa right now, help us out and we'll put you in development later' line. they will always need qa, and you will be a qa person forever (unless you want to be qa, in which case you can just get a qa job at the place of your choosing and stop bothering slashdot).
while everyone wants the best assurance someone is going to be productive by asking for 'senior' people, the smart groups mix in some junior people as well. the good ones are far more energentic, they are more ignorant and eager to prove themselves, you can get alot out of them. they aren't as closed-minded, they are cheaper, and in a year or two worth just as much. its an investment of management time with a certain associated risk and a good payoff. of course the company needs to be healthy enough to vent the accumulated deadwood on occasion.
Slashdot Mirror
actually anyone we dont like. you understand how it is, times are harsh.
they are just echoing the times (speaking of, when did the nytimes start
doing this, its very annoying..puff peice on street hot dog vendors includes
three quotes at the end)
the invisible hand is the new messiah.
your quote from the article is extremely relevant. it shows that god
has sided with apple, so they have the moral imperative. how else
is a poor reader supossed to be able to judge such a complicated
issue?
i disagree. when really old and stable code is too crufty to extend
into a new domain, its a great time to refactor it. take it piece
by piece and do fairly extensive regressions. if something breaks,
you should have an excellent idea where it might be.
substitute 'adjacency updates' for 'pheromones' and you have a generic dynamic routing protocol...
is there anyone else here who remembers this man's
earlier tenure?
oh yeah, i forgot to mention (i had to look to see if it had already been published)...the other cool part is the global architecture. that is there is a large switching fabric connecting all the pims together. aside from the normal reads and writes, it also supports parcels, which is actually a whole migratory thread state. it just gets put in the run queue at the target.
so if there is any spacial locality to be exploited, you can move the thread rather than the data. because this is MTA style you would expect the overall throughput gains to be modest, but you have just reduced your dependency on network throughput correspondingly.
yeah, they're still working on it. i dont know alot of the earlier work, so i dont know how much of this is novel.
they are really fixated on the physical aspects of the memory arrays and building an effective cpu architecture around the context of dram rows (i.e., a thread context is a row, including registers, etc)
so its a little more than just the pin count and interface electronics argument.
i know its hopeless..but,
his work these days centers around efficiencies of access gained by putting the dram and processing elements on the same die. partially removing the serialization associated with the standard synchronous memory interface. The architecture also plans on using MTA-style threads to hide latency and increase concurrency.
citeseer
oh yes. definately. while the article is grossly sensationalistic i suspect that if you were the us dod you'd have some reason to be concerned. not just because of the bad press.
while all actually secret activities aren't on the net, there is alot you can infer from the other random office crap lying around. contracts, billing, travel, social contacts, supply ordering, etc. enough that you've probably saved yourself some time in deciding what to go after with your real intelligence resources
you wank. 'gcc compatible', yes great. you extend a language and then force all sort of people to deal with the resultant garbage. there is a standard, use it. the gcc extentions can be conveient, but they don't really change the scope of the language.
and the intel compiler can really scream over gcc in some applications, and its almost never worse. its a good compiler for a crappy instruction set. why would you not want to use it? because you're a religous fanatic?
its not illegal. its just that the agencies responsible for handing out research money, part of the executive branch, have made this their policy.
think about how many state laws have been passed under the threat of witholding highway fuding.
there are lots of ways the three branches can push something in and of themselves. they can choose to block something (veto, fillibuster, amendment), but it costs.
at this point video doesn't need its own standard. use 10 gig-e, infiniband, pci-e over cable, or any general bit pipe of sufficient breadth.
my genetically engineered chauffeur-lemur
duh
you're holding sid karin up as an example of how to fight waste and cronyism in funding for scientific research? and then the doe supercomputing efforts?
give me some of what you're smoking *right now*
i'm sorry, i got lost. which specific countries were you referring to?
i'd never heard the name of this shell before. what a terrible shame. one could actually make a lovely shell out of monads. in fact that sort of what they are...but instead they apparently just used the name, because it was...what, cool? i've always just hated them from a distance, but this is actually offensive.
yes, i never used an LMI...i learned to program for real on a symbolics and was extremely disappointed when i had to start using a sun.
there is no question that today the barriers to entry make it infeasible, but i still disagree that there is anything fundamental that makes it so. lisp is actually easier to understand in the limit than something like perl.
i'll give you the commodity pc, the value was in the environment, not hardware tagging.
the only real complaint i have about your statements is that you make it seem inevitable, rather than historical accident and social hysteresis.
here you are wrong. did you ever use one? the symbolics was by far the most usable and productive machine at the time.
its primary strength was exactly that you could toss together something pretty nice in basically no time at all.
dont you think it might have something to do with marketing, barriers to entry, and culture? there is nothing fundamentally economic about the failure of symbolics (except the barrier to entry bit, the volume was so low that they were very expensive machines)
i dont use gentoo...i dont even use linux very much...but what i dont understand is that you "open source" people dont seem to understand the intrinsic value of *having the source*. do you realize how utterly ridiculous it is that if i want to look at ifconfig, i have to dig around and find that its in some package called netutils, then find out which version is installed, then get the release patches and apply them.
how did you people get so fixated on binaries. its all about the source.
great. have fun in your playground. the rest of us will try to get on with building and using systems without you. i know it will be hard.
try direct-on-die evaporative cooling
the 13 colonies had the highest standard of living in the world at the time? from my understanding it was less than that afforded to the dominated states of the roman empire. what could you possibly base such an assertion upon?
you delusional idiot. how much information gathering goes on in the us, without warrant, precisely for the purposes of fishing. get a clue.
forget funding, how could they possibly come up with that much content on internet security. 'in other late breaking news, the internet still lacks a decent pki, and script kiddies run rampant'
as a 'senior' person who has done hiring before there are jobs like this. as other people have said, dont look at the 'senior' jobs. they really are looking with people who have experience in the area, can set direction, who know how to deal with large projects, etc.
every place i have been we hired people out of school. while its hard to interview these people, the primary thing i've always been interested in is projects they have done, either outside of school or research projects at school. emphasize your contribution rather than the project itself. you will almost certainly be doing coding questions. the tools you learned in school (algorithms, complexity, analysis) aren't useless in the real world. trot them out.
you will be expected to contribute, but will be giving you smaller projects that are less critical path, tracking your progress more closely, and hopefully be giving you tactical advice. the worst places for entry level are places where no one has time to deal with you, they stick you in a cube, give you some vauge description of your job and ignore you, then lay you off because you aren't performing well. if that happens, demand attention. ask them often what you should be doing and tell them where you are stuck.
look for and apply for things you have some experience or interest in. its easier to get a job doing something you've done, so without some effort once you become a (sysytems, ui, tools, whatever) it will stick with you. in that vein dont beleive the 'we really need qa right now, help us out and we'll put you in development later' line. they will always need qa, and you will be a qa person forever (unless you want to be qa, in which case you can just get a qa job at the place of your choosing and stop bothering slashdot).
while everyone wants the best assurance someone is going to be productive by asking for 'senior' people, the smart groups mix in some junior people as well. the good ones are far more energentic, they are more ignorant and eager to prove themselves, you can get alot out of them. they aren't as closed-minded, they are cheaper, and in a year or two worth just as much. its an investment of management time with a certain associated risk and a good payoff. of course the company needs to be healthy enough to vent the accumulated deadwood on occasion.