Better amend that again, as I'm under 40 as well. I remember DECUS, TU-78 autoloading tape drives (nasty at times, but such cool loading and unloading sounds), and the 11/750 (programmed in VAX Pascal and Fortran).
Too bad someone couldn't shrink one of those onto a chip, and sell a Mac-Mini-sized VAX w/ VMS 6.0 and compilers.
So, we have a critical DEC cutoff-age-big of 37; anybody younger want to bid it down further?
Don't forget some of his non-fiction (or at least non-fictionalized) works, such as, " The Three Most Important Things in Life: Sex, Violence, and Labor Relations" which can be found (apparently officially) at http://harlanellison.com/iwrite/mostimp.htm/ . Two suitably bizarre accounts, and then his Half-day employment by Disney.
Having read some of his personal essays, as well as the standards such as, "I Have No Mouth, but I Must Scream", I would say the operative adjective isn't "prickly" but rather "obstreperous".
Yes, marauding bands of senior citizens spraying grafitti for lack of any other diversion, and gangs of young thugs calling their congressman demanding the return of free TV (and some free drugs, in an unmarked envelope, while they're at it) would just be a bad scene.
How has the takeover by Novell affected your perception of the stability of the system? I stopped running SuSE somewhere around 8.2 when a binary driver from Adaptec, certified on both SuSE and RedHat, ran stabiliy on RH 9.0 and locked up under load on SuSE. I was sorry to let it go, as RH's desktop at that point seemed to unfinished in comparison.
I had always liked the overall polish of the system, and stuck with it even after they abandoned rc.config for the standard (sigh) sys-v init.d scripts. It was, at the time, similar to RH in being cutting edge, without taking that last step over the cliff.
Woodsmith runs articles of that type. You'll learn how to make nicely inlaid, beveled, quality, furniture with tools that will only take up twice the space of your existing computer + gadget collection. Of course, they have articles on how to build cabinets to store the tools and wood as well.
Depressingly, I saw one of them, and then decided not to let a bad movie adaptation ruin another short story for me.
Personally, I'm looking forward to the plug-in supplementary memory from the second, although the idea of vacation memories without the hassle of flying is appealing.
Within the limitation that some mechanism for simulating the blood/brain barrier will have to be devised, this should lead to a new generation of drug screens. Now you can test the effect of new drugs at physiological concentrations on real brain cells. This potentially means no more guess-work based on rat models, and less endangering of real patients during the phase three trials.
Of course, people with more vision than I have will undoubtedly be using this as a way of testing their Borg prototypes, but that's progress of a sort as well. Seriously enough, this will allow you to do the necessary tests to make sure that human cells interface correctly with cybernetic implants, thereby speeding development of bionic eyes, neuro-muscular interfaces, etc.
So, how long until, "we can remember it for you wholesale", or "johnny mnemonic"?
The same reason that certain majoritarian religious groups think of themselves as under siege and about to be exterminated. There is a feeling of community formed by shared threat, that you just can't get any other way. In the beginning, it might even have been correct, but there comes a point where it's simply a stimulus response (look: an advance in computing. AAAAAgh! It's going to wipe out Linux)
Tomorrrow morning an asteroid could hit Redmond, Steve Jobs abmit that he really wants to make movies and sell music, and major computer vendors admit that Stallman was right all along, leaving nothing but Linux on the computing landscape, and these people respond one of two ways. They would either wander off and lose interest, because it's now mainstream and not cool any more, or they'd still claim that at any moment their movement is going to be snuffed out by rapacious commercial interests and indifferent, bovine, consumers, and reman the barricades.
Of course, some times you should be worried. x86 pricepoints for Desktop/Notebook MacOS-X and Cheap, Stable, Scalable, Solaris-boxen in the backroom sounds like the closest to IT heaven that I can imagine. (this is from the standpoint of someone who's administered enough medium-big iron for about 20 years now)
Last year or so I was having fits with RHEL on AMD-64, and thought that I'd at least try installing Debian on a box in order to get a second opinion. I was surprised to find no AMD-64 (despite their balooning popularity), but plenty of odd architectures that probably had only two or three users.
Personally, this seems to me a good time to guarantee AMD64 releases on a timely schedule, and (gently) tell the MIPS and VAX users that their release schedule is going to be somewhat slowed.
I applaud what Debian does (really; I ran it rather happily on a laptop for a couple of years), but am often baffled how they go about doing it.
My more mature, informed, response, is that I'd like to see where the electric motor bushing idea ends up. It's also interesting to see their high surface area to volume being put to use as filter, though it remains to be seen whether they can be made more cost effective than Zeolitic materials
The painting idea is interesting, but from the size of the picture, it doesn't seem that we're looking at single-molecule type resolution in which case conventional lithography is probably superior. On the other hand, I saw a talk by Tom Mallouk of PSU where he demonstrated nanopropulsion with peroxide as fuel, driven by difference in peroxidation rates of different metals on the device. This could be a way of making finely detailed structures of that nature.
Just once, though, I'd like to see a bit less cheekiness in science journalists; nanodustpans, indeed.
Every few years it seems that some variant of using the GPU comes back for scientific computing. I seem to remember in the early 90s a group using the graphics card for the additional memory it could provide. I run quantum-chemistry simulations for a living (basically large quantities of matrix algebra), so anything that could speed up calculations currently taking weeks would be appreciated.
Personally, I'd like to see someone port BLAS (or the ATLAS variant) to a set of standard gpus, so that we could speed up matrix ops. I've been hoping for a more general-purpose solution making it to market, such as the old Celerity strap-on vector unit except for modern IA32/AMD64/PPC, but this may be the better solution.
For those of us who don't have a budget for a Power5 or Cray system, maybe a pair of PCI-e cards running the matrix algebra and FFT routines would be the way to go.
Fair enough, and I'll admit that my field (chemistry), even with the massive downsizings is still doing better than the national average as far as unemployment/underemployment within the field.
Maybe the problem is that for us, at least, while you get the graduate degree because it's needed for a job (industrial or otherwise), nobody prepares you for the reality of much of industrial work.
Personally, I'd like to see the criteria for tenure be based a little more on scholarly merit and less on money, as it is in the humanities, but I'll admit that we're more expensive to keep around than philosophers.
At least I'm not at an institution where my first thought on hearing the group's latest results is, "time to up my proton blockers again". (true quote from a professor at a nearby top 10 institution.
It's like that in most fields. My professors talked about being able to go straight from grad school to tenure-track. By the 70s and 80s, it was common to do one two-year post-doc. Now two postdocs are more and more common.
One of the major problems is that we have a giant apparatus set up in the universities which presumes that a scientist will bring in N dollars in order to get tenure. To do science on that scale, you need assistants, and the cheapest assistants you can get are PhD students. At this point, return to step one.
There used to be more outlets in industry and government labs, as well as smaller institutions with smaller research groups, but the tendency now is to adopt the Big10 football approach to research. Spend like mad to attract the top talent, and expect everyone to produce as much money as possible to support this system.
I don't know what the answer is either, but I've taken the hired-gun approach in that if someone will pay me to study it (and isn't a company that has pre-determined result they want for PR purposes, i.e. no "cigarettes are good for you" money), then there are a wide variety of topics nominally under the rubric of my field that I could become interested in. I've already taken that approach in putting certain ideas in the notebook for later, and gearing my current research onto topics that NIH is likely to be interested in. I could see urge to fudge a little in high-pressure cases, but also understand the greater desire of rapidly dropping projects that won't result in publications and that aren't going anywhere.
Maybe the problem is that we need a predator introduced into the system earlier than Assistant professor tenure reviews. Something that has a digestion strong enough to eat grad students. Alternatively, since there seems to be an infinite amount of money sloshing around for inept actors/actresses, sports, and churches, maybe we just need to institute a science tax and expand the national labs. Personally, my money is on the predator.
The definition when I was in college was, "a sport involves a ball". Therefore, Fencing, Cross-Country, Ultimate Frisbee(*) and Equestrian Team are not sports, while Basketball, baseball, golf, and... umm... let's not go there, are sports.
*frisbee might be a sport, only in the sense that a frisbee is topologically equivalent to a squashed hand-ball, but we leave this proof as an exercise for the reader.
Strangely enough, sometimes theoretical underpinnings are crucial to end-user applications. Yes, maybe this should have been funded by a grant from NIH (new bacterial simulator), NSF (ibid), FDA/DoAg (food contamination), or DOE (model potential industrial organisms), rather than DoHS, but e. Coli is both (a) well-understood and (b) a real problem in our food supply. A good e. coli simulator could, as they've already demonstrated, teach us quite a bit about potential pathogens, or be generalized into an organismal simulator for other, more potentially useful, bacteria.
In principle, presume that you model various strategies of optimizing for methanogens, leading to a shorter development time for biofuels, or even optimize e. coli for better use in industrial synthesis (i.e. an e. coli that will produce 20% more HGH for the same feed input). As long as the model is directly related to experiments, this is a great idea, and will lead us towards a future where eukaryotes (such as your cells, or more plausibly food crops), can be digitally screened and optimized for their response to various promotors and inhibitors.
What's the worst that happens; someone has figured out how to get DoHS to pay for an improved version of Life, and fabulous new screensavers are developed! In all seriousness, the outcome should be much more promising than that. This is one more step to placing usable molecular biology tools into everyone's hands. You could look at individual molecules, and now you can simulate the workhorse of microbiology, without having to have petri dishes or autoclaves around until the last stages of the experiments. I, for one (remembering the smell of agar in the morning), welcome this development.
When Windows NT (4.0?) came out, it was discovered that you only needed to tweak to registry keys to turn Workstation into Server.
You can argue for the consumer (why shouldn't I have access to the full thing for the lower price), or for Microsoft (why shouldn't we charge people who will generate more support calls more for the full version). In the end, the Unixish systems do it right, though. SuSE (Novell) and Redhat do this as well, although in those cases you can get the source, adn compile the missing pieces yourself. Even in OS-X you have that option, though you'll lose the nifty graphical interfaces.
While I disagree with the pricing differentials, I am generally in favor of a starter version of OS's that make it hard to hurt yourself. AIX-lite, anyone?
Since these parts start on earth, you're not going to catch anything that the flight control team doesn't already have. Frankly, Skylab fell on the aussies, and nobody there caught any extraterrestrial bug that anybody outside Oz noticed.
Since the article is slashdotted, I hope they at least take care to plot where the villages are, before letting parts fall everywhere.
It's a teleconferencing on caffeine solution from NCSA. It's used for large, distributed, meetings, and interactive collaboration. We had an access point at my last job, and it works, if you've got the hardware/bandwidth to throw at it. Nice for distributed learning and lectures, plus can be extended.
On a more personal level, i.e. if you can run a skunkworks for a while while you build the environment, would be for you and a couple of like-minded developers to start playing with OpenCroquet. This gives you a persistent virtual environment, with the ability to run programs from the remote participants as screens within the CroquetSpace. http://www.opencroquet.org/
Of course, these are academic solutions, for people with user communities used to space-cadet solutions. However, a Croquetspace with your architects/engineers meeting in a Cave, with people displaying and interactively working on everything from blueprints through solid-models would be majorly cool.
I call 2003 Server. I ran that to support a moderate-sized group plus some computational (Matlab/Mathematica) packages, and it ran stabily and securely. Despite the alleged common code base, it's really quite a different experience from desktop XP. During the various outbreaks I never had a problem with the 2003 box, though we had enough supposedly patched XP desktops catch an infection, or otherwise act in a wonky manner.
Of course, we ran 2003 Server on a machine with redundant power, ECC memory, and hardware RAID-5, so maybe that helped as well.
The only problem with it as a home OS is that it would add another layer of complexity to the life of the already baffled user, their head would explode, and you'd have to clean up after them.
Slashdot Mirror
Better amend that again, as I'm under 40 as well. I remember DECUS, TU-78 autoloading tape drives (nasty at times, but such cool loading and unloading sounds), and the 11/750 (programmed in VAX Pascal and Fortran).
Too bad someone couldn't shrink one of those onto a chip, and sell a Mac-Mini-sized VAX w/ VMS 6.0 and compilers.
So, we have a critical DEC cutoff-age-big of 37; anybody younger want to bid it down further?
As a general rule, I think any method of data storage that could encourage someone to "Get Medieval" on you is just generally a Bad Thing (tm).
Don't forget some of his non-fiction (or at least non-fictionalized) works, such as, " The Three Most Important Things in Life: Sex, Violence, and Labor Relations" which can be found (apparently officially) at http://harlanellison.com/iwrite/mostimp.htm/ . Two suitably bizarre accounts, and then his Half-day employment by Disney.
Having read some of his personal essays, as well as the standards such as, "I Have No Mouth, but I Must Scream", I would say the operative adjective isn't "prickly" but rather "obstreperous".
Yes, marauding bands of senior citizens spraying grafitti for lack of any other diversion, and gangs of young thugs calling their congressman demanding the return of free TV (and some free drugs, in an unmarked envelope, while they're at it) would just be a bad scene.
Nah. The revival process would throw his rhythm off.
How has the takeover by Novell affected your perception of the stability of the system? I stopped running SuSE somewhere around 8.2 when a binary driver from Adaptec, certified on both SuSE and RedHat, ran stabiliy on RH 9.0 and locked up under load on SuSE. I was sorry to let it go, as RH's desktop at that point seemed to unfinished in comparison.
I had always liked the overall polish of the system, and stuck with it even after they abandoned rc.config for the standard (sigh) sys-v init.d scripts. It was, at the time, similar to RH in being cutting edge, without taking that last step over the cliff.
For posting that link to Slashdot I hope NCSA throws a yellow flag and puts you in the penalty box.
On the previous cluster I managed, I always wanted a Penalty Box for certain users. I figured a PDP-11/70 running G77 would suffice.
Woodsmith runs articles of that type. You'll learn how to make nicely inlaid, beveled, quality, furniture with tools that will only take up twice the space of your existing computer + gadget collection. Of course, they have articles on how to build cabinets to store the tools and wood as well.
Depressingly, I saw one of them, and then decided not to let a bad movie adaptation ruin another short story for me.
Personally, I'm looking forward to the plug-in supplementary memory from the second, although the idea of vacation memories without the hassle of flying is appealing.
Within the limitation that some mechanism for simulating the blood/brain barrier will have to be devised, this should lead to a new generation of drug screens. Now you can test the effect of new drugs at physiological concentrations on real brain cells. This potentially means no more guess-work based on rat models, and less endangering of real patients during the phase three trials.
Of course, people with more vision than I have will undoubtedly be using this as a way of testing their Borg prototypes, but that's progress of a sort as well. Seriously enough, this will allow you to do the necessary tests to make sure that human cells interface correctly with cybernetic implants, thereby speeding development of bionic eyes, neuro-muscular interfaces, etc.
So, how long until, "we can remember it for you wholesale", or "johnny mnemonic"?
The same reason that certain majoritarian religious groups think of themselves as under siege and about to be exterminated. There is a feeling of community formed by shared threat, that you just can't get any other way. In the beginning, it might even have been correct, but there comes a point where it's simply a stimulus response (look: an advance in computing. AAAAAgh! It's going to wipe out Linux)
Tomorrrow morning an asteroid could hit Redmond, Steve Jobs abmit that he really wants to make movies and sell music, and major computer vendors admit that Stallman was right all along, leaving nothing but Linux on the computing landscape, and these people respond one of two ways. They would either wander off and lose interest, because it's now mainstream and not cool any more, or they'd still claim that at any moment their movement is going to be snuffed out by rapacious commercial interests and indifferent, bovine, consumers, and reman the barricades.
Of course, some times you should be worried. x86 pricepoints for Desktop/Notebook MacOS-X and Cheap, Stable, Scalable, Solaris-boxen in the backroom sounds like the closest to IT heaven that I can imagine. (this is from the standpoint of someone who's administered enough medium-big iron for about 20 years now)
Last year or so I was having fits with RHEL on AMD-64, and thought that I'd at least try installing Debian on a box in order to get a second opinion. I was surprised to find no AMD-64 (despite their balooning popularity), but plenty of odd architectures that probably had only two or three users. Personally, this seems to me a good time to guarantee AMD64 releases on a timely schedule, and (gently) tell the MIPS and VAX users that their release schedule is going to be somewhat slowed. I applaud what Debian does (really; I ran it rather happily on a laptop for a couple of years), but am often baffled how they go about doing it.
My immediate response is, "nifty".
My more mature, informed, response, is that I'd like to see where the electric motor bushing idea ends up. It's also interesting to see their high surface area to volume being put to use as filter, though it remains to be seen whether they can be made more cost effective than Zeolitic materials
The painting idea is interesting, but from the size of the picture, it doesn't seem that we're looking at single-molecule type resolution in which case conventional lithography is probably superior. On the other hand, I saw a talk by Tom Mallouk of PSU where he demonstrated nanopropulsion with peroxide as fuel, driven by difference in peroxidation rates of different metals on the device. This could be a way of making finely detailed structures of that nature.
Just once, though, I'd like to see a bit less cheekiness in science journalists; nanodustpans, indeed.
Every few years it seems that some variant of using the GPU comes back for scientific computing. I seem to remember in the early 90s a group using the graphics card for the additional memory it could provide. I run quantum-chemistry simulations for a living (basically large quantities of matrix algebra), so anything that could speed up calculations currently taking weeks would be appreciated.
Personally, I'd like to see someone port BLAS (or the ATLAS variant) to a set of standard gpus, so that we could speed up matrix ops. I've been hoping for a more general-purpose solution making it to market, such as the old Celerity strap-on vector unit except for modern IA32/AMD64/PPC, but this may be the better solution.
For those of us who don't have a budget for a Power5 or Cray system, maybe a pair of PCI-e cards running the matrix algebra and FFT routines would be the way to go.
Fair enough, and I'll admit that my field (chemistry), even with the massive downsizings is still doing better than the national average as far as unemployment/underemployment within the field.
Maybe the problem is that for us, at least, while you get the graduate degree because it's needed for a job (industrial or otherwise), nobody prepares you for the reality of much of industrial work.
Personally, I'd like to see the criteria for tenure be based a little more on scholarly merit and less on money, as it is in the humanities, but I'll admit that we're more expensive to keep around than philosophers.
At least I'm not at an institution where my first thought on hearing the group's latest results is, "time to up my proton blockers again". (true quote from a professor at a nearby top 10 institution.
It's like that in most fields. My professors talked about being able to go straight from grad school to tenure-track. By the 70s and 80s, it was common to do one two-year post-doc. Now two postdocs are more and more common.
One of the major problems is that we have a giant apparatus set up in the universities which presumes that a scientist will bring in N dollars in order to get tenure. To do science on that scale, you need assistants, and the cheapest assistants you can get are PhD students. At this point, return to step one.
There used to be more outlets in industry and government labs, as well as smaller institutions with smaller research groups, but the tendency now is to adopt the Big10 football approach to research. Spend like mad to attract the top talent, and expect everyone to produce as much money as possible to support this system.
I don't know what the answer is either, but I've taken the hired-gun approach in that if someone will pay me to study it (and isn't a company that has pre-determined result they want for PR purposes, i.e. no "cigarettes are good for you" money), then there are a wide variety of topics nominally under the rubric of my field that I could become interested in. I've already taken that approach in putting certain ideas in the notebook for later, and gearing my current research onto topics that NIH is likely to be interested in. I could see urge to fudge a little in high-pressure cases, but also understand the greater desire of rapidly dropping projects that won't result in publications and that aren't going anywhere.
Maybe the problem is that we need a predator introduced into the system earlier than Assistant professor tenure reviews. Something that has a digestion strong enough to eat grad students. Alternatively, since there seems to be an infinite amount of money sloshing around for inept actors/actresses, sports, and churches, maybe we just need to institute a science tax and expand the national labs. Personally, my money is on the predator.
Golf is a sport.
The definition when I was in college was, "a sport involves a ball". Therefore, Fencing, Cross-Country, Ultimate Frisbee(*) and Equestrian Team are not sports, while Basketball, baseball, golf, and... umm... let's not go there, are sports.
*frisbee might be a sport, only in the sense that a frisbee is topologically equivalent to a squashed hand-ball, but we leave this proof as an exercise for the reader.
Strangely enough, sometimes theoretical underpinnings are crucial to end-user applications. Yes, maybe this should have been funded by a grant from NIH (new bacterial simulator), NSF (ibid), FDA/DoAg (food contamination), or DOE (model potential industrial organisms), rather than DoHS, but e. Coli is both (a) well-understood and (b) a real problem in our food supply. A good e. coli simulator could, as they've already demonstrated, teach us quite a bit about potential pathogens, or be generalized into an organismal simulator for other, more potentially useful, bacteria.
In principle, presume that you model various strategies of optimizing for methanogens, leading to a shorter development time for biofuels, or even optimize e. coli for better use in industrial synthesis (i.e. an e. coli that will produce 20% more HGH for the same feed input). As long as the model is directly related to experiments, this is a great idea, and will lead us towards a future where eukaryotes (such as your cells, or more plausibly food crops), can be digitally screened and optimized for their response to various promotors and inhibitors.
What's the worst that happens; someone has figured out how to get DoHS to pay for an improved version of Life, and fabulous new screensavers are developed! In all seriousness, the outcome should be much more promising than that. This is one more step to placing usable molecular biology tools into everyone's hands. You could look at individual molecules, and now you can simulate the workhorse of microbiology, without having to have petri dishes or autoclaves around until the last stages of the experiments. I, for one (remembering the smell of agar in the morning), welcome this development.
Deja vu all over again.
When Windows NT (4.0?) came out, it was discovered that you only needed to tweak to registry keys to turn Workstation into Server.
You can argue for the consumer (why shouldn't I have access to the full thing for the lower price), or for Microsoft (why shouldn't we charge people who will generate more support calls more for the full version). In the end, the Unixish systems do it right, though. SuSE (Novell) and Redhat do this as well, although in those cases you can get the source, adn compile the missing pieces yourself. Even in OS-X you have that option, though you'll lose the nifty graphical interfaces.
While I disagree with the pricing differentials, I am generally in favor of a starter version of OS's that make it hard to hurt yourself. AIX-lite, anyone?
The intelligent life was there Nov, 2003 (Supercomputing), but has since gone home. Nothing left but the golfers, I'm afraid.
Since these parts start on earth, you're not going to catch anything that the flight control team doesn't already have. Frankly, Skylab fell on the aussies, and nobody there caught any extraterrestrial bug that anybody outside Oz noticed.
Since the article is slashdotted, I hope they at least take care to plot where the villages are, before letting parts fall everywhere.
I wonder what Borat has to say about this?
Personally, i always loved the campus motto: "Where Fun Comes to Die".
I miss U. of C.
Wonder what the Chicago Weekly News' (the less disciplined, more anti-authoritarian, campus paper) take on this incident will be?
Weirdly enough, a recent article indicated that you actually want the alcohol; probably as a solvent, though possibly as a vascodialator.
However, you want only the minimal amount, as above that level you increase your chance of death from clumsiness.
http://www.accessgrid.org/
It's a teleconferencing on caffeine solution from NCSA. It's used for large, distributed, meetings, and interactive collaboration. We had an access point at my last job, and it works, if you've got the hardware/bandwidth to throw at it. Nice for distributed learning and lectures, plus can be extended.
On a more personal level, i.e. if you can run a skunkworks for a while while you build the environment, would be for you and a couple of like-minded developers to start playing with OpenCroquet. This gives you a persistent virtual environment, with the ability to run programs from the remote participants as screens within the CroquetSpace. http://www.opencroquet.org/
Of course, these are academic solutions, for people with user communities used to space-cadet solutions. However, a Croquetspace with your architects/engineers meeting in a Cave, with people displaying and interactively working on everything from blueprints through solid-models would be majorly cool.
I call 2003 Server. I ran that to support a moderate-sized group plus some computational (Matlab/Mathematica) packages, and it ran stabily and securely. Despite the alleged common code base, it's really quite a different experience from desktop XP. During the various outbreaks I never had a problem with the 2003 box, though we had enough supposedly patched XP desktops catch an infection, or otherwise act in a wonky manner.
Of course, we ran 2003 Server on a machine with redundant power, ECC memory, and hardware RAID-5, so maybe that helped as well.
The only problem with it as a home OS is that it would add another layer of complexity to the life of the already baffled user, their head would explode, and you'd have to clean up after them.