Well, you won't be getting a Halon system any more, as they have been outlawed due to the CFC laws. I'm sure there's some non-CFC-based replacement tho.
I'm assuming you are building this small room in a rented-office type situation, probably in a building that normally houses more people than computers.
I've been where you are, and let me say that your #1 problem will probably be cooling. If it's really a closet/old office/whatever, it is unlikely your building management will be able to get enough cooling to you. I'd recommend planning for one of those free-standing moving cooling units that can vent into the drop ceiling. By planning, I mean both power and space.
Power: Again, better get in touch with your bldg management. Most office circuits are in the range of 20 amps, which sounds like a lot, but isn't -- you shouldn't plan on using all 20 amps, and remember that if you coldstart everything at once, they will pull a lot more than 20. I would plan on an amp per server at least, and go no more than 16 boxes per circuit. You may need a 220 circuit for a movable AC or some other weird piece of equipment.
Do NOT do the latter yourself. Hire a professional electrician. One mistake in this area can not only ruin your business, but potentially take your life.
Right on, brutha. Very well written. And one conclusion you can easilly draw is that the industry is trying like hell to keep those "independent content producers" you talk about in your last paragraph OUT OF THE PICTURE.
This is more obvious (and more evilly so) in the record industry.
[on another note: Valenti should stop bitching about his 2 out of 10 -- for a wakeup slap, he should talk to some venture capitalists -- for whom 2 out of 10 might be a GOOD success rate!]
People in this thread need to enable their satire bit! The point I was trying to get across was the largely-self-inflated sense of importance that the movie industry has in the US. They need to learn to work and play better with others -- because they have competition that is nipping on their heels, and they can't always push people around.
In fact, this is just a case of history repeating itself. The MPAA ought to remember how they (the film industry) wrested control from Thomas Edison who held lots of patents on film production and film stock -- and used them to weild monopoly control, until he was dope-slapped by the Supreme Court.
Cheaper only relative to the tarriffed product. In fact, tarriffs make products more EXPENSIVE by artificially inflating the price of the local product and supporting inefficient industries.
See the US steel industry for a fine example of this.
Although people in this thread have already pointed out that I only included first-run revenues in my figures, you can look it up yourself: in the last year for which there is figures (2000) the movie industry took in something like 7bn$, and the video game industry topped 8bn$. And movie attendence has been in decline for several years, so they've largely received this revenue through increases in admissions price.
http://www.mpaa.org has some nice charts. I can't find a similar site for the video game industry, but you can find revenue figures in tons of articles.
Fair enough -- and when the Bollywood version of Jack Valenti starts making such incendiary remarks and stomping his feet in the Washington Post, screaming that the sky is falling, then we can take him/her on as well:-)
Won't work *without* legislation???!? Won't work *WITH* legislation!
I really really wish that the Movie and Record industry would lose their image of self-importance in our society that's largely propped up by the Hollywood star machine. It's already well-known that the VIDEO GAME INDUSTRY grosses more money than the film industry (and hey, probably nets more too -- put that in your "2 out of 10" pipe and smoke it, Mr. Valenti!).
And more to the point, IBM alone grosses more than the film, TV, and music industry put together! If I were Valenti, I'm not sure I'd be making such a ruckus. What if IBM, Sony, Dell, Microsoft, you-name-it got together and said "these movie people are a pain in the ass -- rather than build copy protection into our hardware/software for THEM, we'll just BUY THEM OUT and give away loads of free movies to our customers!"???
True, yes, I agree that "completely self-contained" is a wonderful feature (have you seen tinyapps.org???). My basic tenet would be "if you need to use it (registry, whatever), you MUST provide a way to clean it up".
First, Amazon has a set list of discounts, which the hardcover version probably falls into, and the electronic version does not.
Second, Amazon's gross margin on a hardcover version may be higher, so they have more room to discount. For instance, if the suggested list price of the HC is $30, and the electronic is $24, and they both are "sold" to Amazon by the publisher for $20, Amazon has more room to come down on the HC version.
Third, Amazon may just not care all that much!
Re:This CAN be trivially done on any un*x i know..
on
UNIX Process Cryogenics?
·
· Score: 3, Informative
You can't. The previous poster was making it sound too easy. Real checkpointing needs to save Kernel state as well -- file handles, device driver state, you name it. It isn't as simple as saving the in-memory image of the process.
You might think it's BS, but I used to work for a company that had the local exchange "234". The CTO of the company had the extension "5678". You know the rest. He changed his extension after about two days. Apparently the telco couldn't GIVE away that exchange prefix until we took it.
In a way, it does -- but what is the deal with GSM, anyway?
Yes, all of Europe uses it -- but invariably, whenever Motorola et al introduce some snazzy new phone, it's GSM-only first. What, they don't care about any American early adopters??? Going with GSM with the Treo means throwing away the three biggest cellular providers in the US.
And more to the point, I would think that the ACM would take umbrage at CACM being classified as a journal of "software development and research"...! There is a LOT more in all those magazines than that.
I seriously doubt the GOVERNMENT was thinking anthrax when this rule was written! I'm sure there's some much more obscure reason than that. Besides, I don't think Anthrax propagates particularly well in water...
It's not the fact that the government is paying researchers to develop some wonderful thing that they then go off and own completely. The government always retains rights FOR GOVERNMENT PURPOSES -- usually to everything, including source code, design drawings, what have you.
Now, they can't just turn around, take the source code, and give it to the public, but they can re-use it for any government purpose. I believe they can even give it to another government contractor to do government things. It isn't as bleak as is made out.
An argument can also be made that exclusive commercial rights are used as an incentive to get more people to step up to the plate and bid on various research proposals. I work for a R&D shop that gets funds from the government, and while I suspect we would probably be fine if we had to give away code, we certainly try to make commercial use of what we build (only sometimes successfully:-). Oooh, I'm evil:-)
Finally, a lot of people don't realize it, but the government LIKES it when people have an incentive to commercialize things. $300 toilet seats aside, the government would MUCH RATHER just go out and buy some COTS (Commercial Off The Shelf) technology than pay to have it developed. If the company they give the R&D funds to can commercialize a product and get it on the GSA schedule, the government frequently considers that a job well done. This may happen just as quickly with open source, but I don't think that's been proven yet.
Finally, there are a whole class of things for which (at least initially) the government might not see a commercial use for, AND the contractor might not either [some strange intelligence application, or weird battlefield piece of hardware, for instance]. If the government can say "and if you find a commercial use for this, you can have it", more people may be interested in scratching their heads over the problem enough to respond to an RFP or bid on a proposal.
Wow, #9 sure stands out as a sore thumb in that list, don't it???:-) I think I can fathom the reason (no pun intended), but it seems afully specific compared to the others (and there are a lot of similar things that you think they WOULD have included!).
Did they really have that many people asking about wells?
Yes, I own the house. But it is a decidedly non-traditional house.
The house is an open, three-story loft-style abode. This means that there are 3.5 floors, all open to a central core, with no internal walls to speak of! (other than bathroom, half-bath, a few closets).
And, as there is no basement or attic, I can't run under or over to get from one side of the house to another, which is where a cable run would need to go -- due to arrangement of the "office" on floor 2.5 and the entertainment center on the ground floor on the opposite wall.
I could pull the baseboard off, but I would have to drill a hole through each and every stud in one wall to get a cable to my TV/stereo cabinet -- this is not terribly practical.
I could rearrange the entire "downstairs" to accomodate this project, but I like the way things are set up!
Plus, this is the wireless era! I should be able to do this:-)
Well... maybe. I don't honestly know if this will work. Will an access point talk to another access point? If so, then your suggestion will definitely work. The setup I'm envisioning has the access point in my "computer loft", and any clients (be they laptops or whatever) having a smaller/cheaper (!) solution.
I'd like the same thing -- except with 10BT on one end. Essentially, I'd like a IEEE802.11->10BT bridge.
Why? I don't have the ability to run cable from my "lan" area to my stereo without it looking extremely unsightly. But I'd love to be able to buy an audiotron. If I had such a bridge, I'd be all set.
Does anyone out there know of such a thing, for a reasonable cost?
Re: standard databases. The performance DOES actually represent a real-world system, as long as you have reasonable expectations of the real world system. I.e. do not train your system on a broadcast news corpus and then feed it Jerry Springer:-). But if your test and training is reasonably matched, your performance matches as well. There are real-world systems that do this.
As to whether the task is worth doing -- well, that's for neither of us to answer, but the market + Virage, et al -- right?:-)
Yes, I'm watching BBC on a US TV. I'm probably behind the times on that.
I don't know of a single speech recognition system that is trained on CC. I work in the industry -- the transcripts aren't accurate enough. Every piece of training data is generated by a human annotator specifically generating the data for speech rec training. The words are aligned to millisecond intervals. You can't train a speech rec system on a paraphrase.
Also, the published results of numbers of companies are in the 90% range. These are ajudicated results -- not marketing material.
I agree with you that speech rec may be one technique -- but it's the only one that scales. People aren't going to be willing to pay to transcribe everything -- they just aren't. But people want (and should be able to) search everything -- form today's CNN to today's TVLand TV repeat. The only practical way to do that is some automatic approach.
Slashdot Mirror
Well, you won't be getting a Halon system any more, as they have been outlawed due to the CFC laws. I'm sure there's some non-CFC-based replacement tho.
GOOD ONE! Hah! I can't COUNT the number of times I've been inconvenienced by this! A phone with a LOOOOOOONNNG cord! :-)
I'm assuming you are building this small room in a rented-office type situation, probably in a building that normally houses more people than computers.
I've been where you are, and let me say that your #1 problem will probably be cooling. If it's really a closet/old office/whatever, it is unlikely your building management will be able to get enough cooling to you. I'd recommend planning for one of those free-standing moving cooling units that can vent into the drop ceiling. By planning, I mean both power and space.
Power: Again, better get in touch with your bldg management. Most office circuits are in the range of 20 amps, which sounds like a lot, but isn't -- you shouldn't plan on using all 20 amps, and remember that if you coldstart everything at once, they will pull a lot more than 20. I would plan on an amp per server at least, and go no more than 16 boxes per circuit. You may need a 220 circuit for a movable AC or some other weird piece of equipment.
Do NOT do the latter yourself. Hire a professional electrician. One mistake in this area can not only ruin your business, but potentially take your life.
Yes, and it's worth noting that this use of the term predates the software sense :-)
I was wondering if this article got posted simply because of that confusion...
Right on, brutha. Very well written. And one conclusion you can easilly draw is that the industry is trying like hell to keep those "independent content producers" you talk about in your last paragraph OUT OF THE PICTURE.
This is more obvious (and more evilly so) in the record industry.
[on another note: Valenti should stop bitching about his 2 out of 10 -- for a wakeup slap, he should talk to some venture capitalists -- for whom 2 out of 10 might be a GOOD success rate!]
People in this thread need to enable their satire bit! The point I was trying to get across was the largely-self-inflated sense of importance that the movie industry has in the US. They need to learn to work and play better with others -- because they have competition that is nipping on their heels, and they can't always push people around.
In fact, this is just a case of history repeating itself. The MPAA ought to remember how they (the film industry) wrested control from Thomas Edison who held lots of patents on film production and film stock -- and used them to weild monopoly control, until he was dope-slapped by the Supreme Court.
Cheaper only relative to the tarriffed product. In fact, tarriffs make products more EXPENSIVE by artificially inflating the price of the local product and supporting inefficient industries.
See the US steel industry for a fine example of this.
Although people in this thread have already pointed out that I only included first-run revenues in my figures, you can look it up yourself: in the last year for which there is figures (2000) the movie industry took in something like 7bn$, and the video game industry topped 8bn$. And movie attendence has been in decline for several years, so they've largely received this revenue through increases in admissions price.
http://www.mpaa.org has some nice charts. I can't find a similar site for the video game industry, but you can find revenue figures in tons of articles.
Fair enough -- and when the Bollywood version of Jack Valenti starts making such incendiary remarks and stomping his feet in the Washington Post, screaming that the sky is falling, then we can take him/her on as well :-)
Won't work *without* legislation???!? Won't work *WITH* legislation!
I really really wish that the Movie and Record industry would lose their image of self-importance in our society that's largely propped up by the Hollywood star machine. It's already well-known that the VIDEO GAME INDUSTRY grosses more money than the film industry (and hey, probably nets more too -- put that in your "2 out of 10" pipe and smoke it, Mr. Valenti!).
And more to the point, IBM alone grosses more than the film, TV, and music industry put together! If I were Valenti, I'm not sure I'd be making such a ruckus. What if IBM, Sony, Dell, Microsoft, you-name-it got together and said "these movie people are a pain in the ass -- rather than build copy protection into our hardware/software for THEM, we'll just BUY THEM OUT and give away loads of free movies to our customers!"???
True, yes, I agree that "completely self-contained" is a wonderful feature (have you seen tinyapps.org???). My basic tenet would be "if you need to use it (registry, whatever), you MUST provide a way to clean it up".
Well, strictly speaking, the program should
a) come with an uninstaller;
b) said uninstaller should wipe the registry keys
First, Amazon has a set list of discounts, which the hardcover version probably falls into, and the electronic version does not.
Second, Amazon's gross margin on a hardcover version may be higher, so they have more room to discount. For instance, if the suggested list price of the HC is $30, and the electronic is $24, and they both are "sold" to Amazon by the publisher for $20, Amazon has more room to come down on the HC version.
Third, Amazon may just not care all that much!
You can't. The previous poster was making it sound too easy. Real checkpointing needs to save Kernel state as well -- file handles, device driver state, you name it. It isn't as simple as saving the in-memory image of the process.
You might think it's BS, but I used to work for a company that had the local exchange "234". The CTO of the company had the extension "5678". You know the rest. He changed his extension after about two days. Apparently the telco couldn't GIVE away that exchange prefix until we took it.
In a way, it does -- but what is the deal with GSM, anyway?
Yes, all of Europe uses it -- but invariably, whenever Motorola et al introduce some snazzy new phone, it's GSM-only first. What, they don't care about any American early adopters??? Going with GSM with the Treo means throwing away the three biggest cellular providers in the US.
And more to the point, I would think that the ACM would take umbrage at CACM being classified as a journal of "software development and research"...! There is a LOT more in all those magazines than that.
I seriously doubt the GOVERNMENT was thinking anthrax when this rule was written! I'm sure there's some much more obscure reason than that. Besides, I don't think Anthrax propagates particularly well in water...
But maybe they meant "OIL wells"... hmmm...
An important point that may have been missed:
:-). Oooh, I'm evil :-)
It's not the fact that the government is paying researchers to develop some wonderful thing that they then go off and own completely. The government always retains rights FOR GOVERNMENT PURPOSES -- usually to everything, including source code, design drawings, what have you.
Now, they can't just turn around, take the source code, and give it to the public, but they can re-use it for any government purpose. I believe they can even give it to another government contractor to do government things. It isn't as bleak as is made out.
An argument can also be made that exclusive commercial rights are used as an incentive to get more people to step up to the plate and bid on various research proposals. I work for a R&D shop that gets funds from the government, and while I suspect we would probably be fine if we had to give away code, we certainly try to make commercial use of what we build (only sometimes successfully
Finally, a lot of people don't realize it, but the government LIKES it when people have an incentive to commercialize things. $300 toilet seats aside, the government would MUCH RATHER just go out and buy some COTS (Commercial Off The Shelf) technology than pay to have it developed. If the company they give the R&D funds to can commercialize a product and get it on the GSA schedule, the government frequently considers that a job well done. This may happen just as quickly with open source, but I don't think that's been proven yet.
Finally, there are a whole class of things for which (at least initially) the government might not see a commercial use for, AND the contractor might not either [some strange intelligence application, or weird battlefield piece of hardware, for instance]. If the government can say "and if you find a commercial use for this, you can have it", more people may be interested in scratching their heads over the problem enough to respond to an RFP or bid on a proposal.
Wow, #9 sure stands out as a sore thumb in that list, don't it??? :-) I think I can fathom the reason (no pun intended), but it seems afully specific compared to the others (and there are a lot of similar things that you think they WOULD have included!).
Did they really have that many people asking about wells?
Well, since you asked :-)
:-)
Yes, I own the house. But it is a decidedly non-traditional house.
The house is an open, three-story loft-style abode. This means that there are 3.5 floors, all open to a central core, with no internal walls to speak of! (other than bathroom, half-bath, a few closets).
And, as there is no basement or attic, I can't run under or over to get from one side of the house to another, which is where a cable run would need to go -- due to arrangement of the "office" on floor 2.5 and the entertainment center on the ground floor on the opposite wall.
I could pull the baseboard off, but I would have to drill a hole through each and every stud in one wall to get a cable to my TV/stereo cabinet -- this is not terribly practical.
I could rearrange the entire "downstairs" to accomodate this project, but I like the way things are set up!
Plus, this is the wireless era! I should be able to do this
Well... maybe. I don't honestly know if this will work. Will an access point talk to another access point? If so, then your suggestion will definitely work. The setup I'm envisioning has the access point in my "computer loft", and any clients (be they laptops or whatever) having a smaller/cheaper (!) solution.
I'd like the same thing -- except with 10BT on one end. Essentially, I'd like a IEEE802.11->10BT bridge.
Why? I don't have the ability to run cable from my "lan" area to my stereo without it looking extremely unsightly. But I'd love to be able to buy an audiotron. If I had such a bridge, I'd be all set.
Does anyone out there know of such a thing, for a reasonable cost?
Re: standard databases. The performance DOES actually represent a real-world system, as long as you have reasonable expectations of the real world system. I.e. do not train your system on a broadcast news corpus and then feed it Jerry Springer :-). But if your test and training is reasonably matched, your performance matches as well. There are real-world systems that do this.
:-)
As to whether the task is worth doing -- well, that's for neither of us to answer, but the market + Virage, et al -- right?
Yes, I'm watching BBC on a US TV. I'm probably behind the times on that.
I don't know of a single speech recognition system that is trained on CC. I work in the industry -- the transcripts aren't accurate enough. Every piece of training data is generated by a human annotator specifically generating the data for speech rec training. The words are aligned to millisecond intervals. You can't train a speech rec system on a paraphrase.
Also, the published results of numbers of companies are in the 90% range. These are ajudicated results -- not marketing material.
I agree with you that speech rec may be one technique -- but it's the only one that scales. People aren't going to be willing to pay to transcribe everything -- they just aren't. But people want (and should be able to) search everything -- form today's CNN to today's TVLand TV repeat. The only practical way to do that is some automatic approach.