It needs it. A while ago I bought a 9800 and openGL performance was so poor under linux (after the half a day and two kernel recompiles it took me to get the drivers to work at all) that it was slower than the GF Ti4600 it replaced. It was so bad that the 9800 was sold on and I put the Ti4600 back in. My last upgrade was to a GF 6800GT, I'm not going to touch ATI again until they make using their drivers as simple as NVidia's
sh./NVIDIA-Linux-x86-1.0-XXXX-pkg1.run
and I get remotely decent performance from their driver and opengl implementation. Shame really, ATI hardware is good, they just seem to hire muppets to write the software for them:/
With no more costs involved in transporting anything instantly you would be looking at cheaper merchandise across the board. That's arguable. In the UK, I've observed a very intereting trend that applies directly to this theory. When fuel prices go up, the price of goods goes up to match the increase in transport costs. When the price of fuel goes down, the goods stay the same price, the difference now forming an extra profit for the hauliers or the retailers. Price reductions across the board when fuel prices decrease is unheard of.
Add a Replicator into the mix and you're really talking about things costing nothing. Only if the ownership of replicators is universal. Consider what would happen if governments heavily regulated their production and ownership and corporations made them so expensive only a handful of places could afford them (and they would do, because they'd know that replicators would be the end of them: any company dealing in material products as opposed to IP would cease to exist overnight while the monetary basis of governments around the world would collapse - something those in charge would do everything possible to avoid)
That was the original intention, however the US threw a strop over the fact that they couldn't blackout chunks of galileo (probably thanks to US threats to destroy the satellites if they couldn't turn them off). The EU caved over the issue and agreed to "harmonise the technology of the networks" - essentially, Galileo will work in sync with GPS and the US blackout of GPS will work on Galileo
Great idea and I agree completely. But unless it is very simple, basic stuff using software written for that purpose (which would involve a lot of work) it's going to get very expensive. You could cut a lot of corners, but still...
Just look at rendering software. If you're not interested in scene quality and just want to go for construction then licenses for a game with an editor and a bit of planning would let you do machinima. But for the "real thing", if you want to teach them anything other than Anim8or or POV (and the latter would scare off a lot of CS students, let alone anyone else) you're looking at Blender (The Terrifying UIed One) or one of the Big Name renderers. Lightwave 8 set me back over £1000, 3D Studio Max is three times that, Maya is even more. And those are per user licenses, I dread to think how much it'd cost to buy a classroom full of licenses for them, certainly more than any educational institution I know could afford. I think it would be incredible to be able to teach kids these things and I'm certain they'd be interested.
Chroma keying software could be hacked up with V4L, a decent webcam and some C code (or, I think, Blender can do it as well?) but it isn't trivial and the stuff that works well is also costly. Whether the cheaper homebrew version would be enough for high-school projects I don't know, I'd guess it could be..
And building computers? I can see lots of static damage, bent pins and broken connectors down that route;) Again, if the place had the money to provide suitable anti-static setups (somehow I doubt some of the tricks/.ers will use to ground themselves will get past school health and safety) and the parts students can build the machines from then great idea. But unless it is restricted to a few students it could end up being a cash drain.
I sound a lot more negative about it than I am - I really wish it could be possible. It's just that the minute "school" appears in the equation, financial considerations take on a significant degree of importance.
Re:OT: Question for Slashdotters
on
Lego Logic Gates
·
· Score: 1
Just a note...
Maplin Electronics is getting to be a bit of a joke these days. Four or five years ago I'd have agreed with you, but the range of components they have is dropping year on year to be replaced by standard consumer electronics.
However, in the UK RS Components now supplies individuals as well as companies and their range of stuff is vast. You don't get the dead tree catalogues unless you've got a commercial account (which, I believe, still requires a minimum monthly order) but if you register online you can get at all the details of the products on the website and there's no minimum order size so it's great for hobbyists.
It nowadys seems to be run by a bunch of paranoid, MBA'd marketing droids with neither a knack for innovation, nor a clue what the customer wants. Oh, they know what the customer (what the Suits call the "consumer" or "cash cow") wants. They just decide that, because the lawyers are jumping up and down and screaming blue murder and the media sorts are having aneurysms over letting the unwashed masses actually use the stuff they by in the way they want, the customer can't have it.
Sony's engineers are still some of the best and they still kno whow to innovate. It's just that the Suits take a hatchet to the features of most of their products so you end up with a compromise between the engineers trying to give the customer all the features and the Suits trying to turn the product into a single-use, single function piece of plastic that the customer can only rent.
They didn't use them after the first season, some places say that they didn't use them after the pilot in fact. The rest was done using mainly intel machines running windows and lightwave for the render work and some alpha boxes handling the renderfarm filestore.
No, its the discussion, especially the flow and context of information, as I'm attempting to prove in this post in fact. You'll find that my comments make far more sense if they are placed within the sections of your post currently being addressed (I'm being nice and splitting it according to the pragraphs in your post). The only alternative way to provide the context that inline posting provides is to replicate parts of the original message in your response: a waste of time and effort, as well as another source of errors.
And a remotely decent emailer will let you do this by providing distinct colouration of the various levels of quoting. Some probably don't. I don't use them.
That's what archive folders and clients capable of following threads are for. There should be no need for excessive duplication of information across messages simply because you should already have the thing - if you're deleting message when you've read them then you have bigger problems on the horizon, if you're not then you should be able to just select the prvious message in the thread. And while it might not make a dent in yours, I guess you don't run a corporate network handling thousands or millions of mails per day - if 60% of those mails are just wasted quote tails, persuading staff to quote only what is needed for context could massively reduce traffic and associated costs.
When you get a spam, you put it in a special folder and the client repeatedly accesses the site
So how do you determine which is the right site programmatically?
Go off the email address? Won't work becasue the vast majority of spam uses forged From addresses (I regularly get bounces for spams some asshat has sent with my domain in the from:)
Write something that interprets the email headers and attacks the originating IP? Won't work thansk to the army of windows boxes running proxies to hide the real sender - you'll just end up attaching an innocent, if ignorant, DSL peon.
Write something that grabs URLs from the email and attacks that? Won't work either.. well, it will work, it just means that now all a spammer has to do is bung the URL of a competitor or someone they don't like in there and now you're doing a DDoS for them.
Pretty much any scheme you come up with has so many ways around it or possible abuses that it'd be more dangerous than the problem itself. Even if it isn't determined programmatically, relying on some degree of user interaction or target selection, it is likely to be open to abuse.
You do know that for technical issues, especially technical issues that have anything to do with piracy, the Beeb is little better than a press-release reprint agency, right?
Yeah, from time to time one or two of the jouralists manage to get up a more well researched argument, but for the most part if company X tells them that Y is "doing great things to fight piracy" or industry X (which they conveniently forget has just had record sales) tells them it is being "destroyed by piracy" then they'll just bung the story up without really doing any significant digging.
The advert revenue on cable allows the cable company to reduces the cost to the subscriber*, effectively the cable subscriber is paying for their subscription in two ways: money and viewing time.
With cable internet it's a different kettle of fish: the subsciber's $49 goes to the cable company, but the revenue from the advertisement doesn't go anywhere near the cable company, it is used by the site maintainer to pay for bandwidth costs. In this case the cable internet subscriber is paying their subscription and the costs of a third party.
The two cases aren't really equivalent: the former is a simple trade of one cost for another, the latter is two costs - one from the cable company and one from the website owner.
* as long as you assume that the cost of the subscription really is > $49. Which it probably isn't, but such is the way of business.
Going off my current experience (reinstalling from a retail DVD right now after a windows reinstall), downloading it will take about as long as installing, decrypting and activating the damned thing. The only difference is that with retail you at least have something physical to show for it. Still takes sodding ages to install either way...
So now I know who to blame! Damn you, damn you and all your kind for the torture you put us poor buyers through! Must you insist on that barrier of plastic between us and our purchases so that we have to spend an age struggling to find a point that will admit a knifepoint or fingernail, only to have the plastic tear before removing any substantial amount of the wrapping?
Yet, if they insist Valve has to sell via steam for the same price as Vivendi do in the stores, then they are forcing Valve to take yet more profit from sales. That makes electronic distribution even more attractive to software developers and more of a threat to conventional publishers.
Normally I hate these sort of catch-22 situations. But not when it's a publisher having to deal with it, then I just laugh.
Hmm.. if we can assume that Maths:= Dutch then, given that this is Dutch Maths, we can substitute Dutch for Maths to prove that ti is completely Double Dutch...
My guess, going off experience of university hall networks in the UK (one of the people I work with is a net admin in one) is that they are doing it so that they can keep an eye on what the students are doing and pull the access of students who are doing things they aren't allowed to do.
If the students set up their own parallel network outside the control of the university then it is possible that the university admins may get reports of activity they need to stop while being unable to actually take action against the student short of suspension or expulsion. This could create a range of problems for the university so they decide to prevent the establishment of a secondary network.
In short - I think this this isn't a technical problem, it's a legal arse-covering move using a technical issue as a cover.
Why in the world didn't the other scientists use this exact same setup when trying to reproduce the results? If you're trying to repeat a result, don't you make sure all variables are the same?
IANANP (nuclear physicist) but I'd expect that getting a 100% setup is difficult - you probably have to remove all impurities and make sure the distribution of atoms in the electrodes is correct (or, more likely, pay a significant amount to obtain materials to the required spec).
Besides which, scientists checking something that they are biased against anyway (either because they see no theoretical basis for the whole idea or because, if it was reproduced reliably, it would threaten their theories or jobs) aren't likely to spend any more than the minimum time and money necessary to say "well, that doesn't work, back to my tokamak". There's a LOT of money tied up in things that would become of questionable importance if cold fusion could be obtained reliably and scaled to useful levels.
Slashdot Mirror
It needs it. A while ago I bought a 9800 and openGL performance was so poor under linux (after the half a day and two kernel recompiles it took me to get the drivers to work at all) that it was slower than the GF Ti4600 it replaced. It was so bad that the 9800 was sold on and I put the Ti4600 back in. My last upgrade was to a GF 6800GT, I'm not going to touch ATI again until they make using their drivers as simple as NVidia's
./NVIDIA-Linux-x86-1.0-XXXX-pkg1.run
:/
sh
and I get remotely decent performance from their driver and opengl implementation. Shame really, ATI hardware is good, they just seem to hire muppets to write the software for them
With no more costs involved in transporting anything instantly you would be looking at cheaper merchandise across the board.
That's arguable. In the UK, I've observed a very intereting trend that applies directly to this theory. When fuel prices go up, the price of goods goes up to match the increase in transport costs. When the price of fuel goes down, the goods stay the same price, the difference now forming an extra profit for the hauliers or the retailers. Price reductions across the board when fuel prices decrease is unheard of.
Add a Replicator into the mix and you're really talking about things costing nothing.
Only if the ownership of replicators is universal. Consider what would happen if governments heavily regulated their production and ownership and corporations made them so expensive only a handful of places could afford them (and they would do, because they'd know that replicators would be the end of them: any company dealing in material products as opposed to IP would cease to exist overnight while the monetary basis of governments around the world would collapse - something those in charge would do everything possible to avoid)
That was the original intention, however the US threw a strop over the fact that they couldn't blackout chunks of galileo (probably thanks to US threats to destroy the satellites if they couldn't turn them off). The EU caved over the issue and agreed to "harmonise the technology of the networks" - essentially, Galileo will work in sync with GPS and the US blackout of GPS will work on Galileo
Great idea and I agree completely. But unless it is very simple, basic stuff using software written for that purpose (which would involve a lot of work) it's going to get very expensive. You could cut a lot of corners, but still...
;) Again, if the place had the money to provide suitable anti-static setups (somehow I doubt some of the tricks /.ers will use to ground themselves will get past school health and safety) and the parts students can build the machines from then great idea. But unless it is restricted to a few students it could end up being a cash drain.
Just look at rendering software. If you're not interested in scene quality and just want to go for construction then licenses for a game with an editor and a bit of planning would let you do machinima. But for the "real thing", if you want to teach them anything other than Anim8or or POV (and the latter would scare off a lot of CS students, let alone anyone else) you're looking at Blender (The Terrifying UIed One) or one of the Big Name renderers. Lightwave 8 set me back over £1000, 3D Studio Max is three times that, Maya is even more. And those are per user licenses, I dread to think how much it'd cost to buy a classroom full of licenses for them, certainly more than any educational institution I know could afford. I think it would be incredible to be able to teach kids these things and I'm certain they'd be interested.
Chroma keying software could be hacked up with V4L, a decent webcam and some C code (or, I think, Blender can do it as well?) but it isn't trivial and the stuff that works well is also costly. Whether the cheaper homebrew version would be enough for high-school projects I don't know, I'd guess it could be..
And building computers? I can see lots of static damage, bent pins and broken connectors down that route
I sound a lot more negative about it than I am - I really wish it could be possible. It's just that the minute "school" appears in the equation, financial considerations take on a significant degree of importance.
Just a note...
Maplin Electronics is getting to be a bit of a joke these days. Four or five years ago I'd have agreed with you, but the range of components they have is dropping year on year to be replaced by standard consumer electronics.
However, in the UK RS Components now supplies individuals as well as companies and their range of stuff is vast. You don't get the dead tree catalogues unless you've got a commercial account (which, I believe, still requires a minimum monthly order) but if you register online you can get at all the details of the products on the website and there's no minimum order size so it's great for hobbyists.
It nowadys seems to be run by a bunch of paranoid, MBA'd marketing droids with neither a knack for innovation, nor a clue what the customer wants.
Oh, they know what the customer (what the Suits call the "consumer" or "cash cow") wants. They just decide that, because the lawyers are jumping up and down and screaming blue murder and the media sorts are having aneurysms over letting the unwashed masses actually use the stuff they by in the way they want, the customer can't have it.
Sony's engineers are still some of the best and they still kno whow to innovate. It's just that the Suits take a hatchet to the features of most of their products so you end up with a compromise between the engineers trying to give the customer all the features and the Suits trying to turn the product into a single-use, single function piece of plastic that the customer can only rent.
Well, at least we know what Holly's /. ID is now...
Probably you're joking, but in case...
They didn't use them after the first season, some places say that they didn't use them after the pilot in fact. The rest was done using mainly intel machines running windows and lightwave for the render work and some alpha boxes handling the renderfarm filestore.
Not that DS9 has bad lighting or cheesy makeup. ;)
It didn't?!
No, its the discussion, especially the flow and context of information, as I'm attempting to prove in this post in fact. You'll find that my comments make far more sense if they are placed within the sections of your post currently being addressed (I'm being nice and splitting it according to the pragraphs in your post). The only alternative way to provide the context that inline posting provides is to replicate parts of the original message in your response: a waste of time and effort, as well as another source of errors.
And a remotely decent emailer will let you do this by providing distinct colouration of the various levels of quoting. Some probably don't. I don't use them.
That's what archive folders and clients capable of following threads are for. There should be no need for excessive duplication of information across messages simply because you should already have the thing - if you're deleting message when you've read them then you have bigger problems on the horizon, if you're not then you should be able to just select the prvious message in the thread. And while it might not make a dent in yours, I guess you don't run a corporate network handling thousands or millions of mails per day - if 60% of those mails are just wasted quote tails, persuading staff to quote only what is needed for context could massively reduce traffic and associated costs.
When you get a spam, you put it in a special folder and the client repeatedly accesses the site
So how do you determine which is the right site programmatically?
Go off the email address? Won't work becasue the vast majority of spam uses forged From addresses (I regularly get bounces for spams some asshat has sent with my domain in the from:)
Write something that interprets the email headers and attacks the originating IP? Won't work thansk to the army of windows boxes running proxies to hide the real sender - you'll just end up attaching an innocent, if ignorant, DSL peon.
Write something that grabs URLs from the email and attacks that? Won't work either.. well, it will work, it just means that now all a spammer has to do is bung the URL of a competitor or someone they don't like in there and now you're doing a DDoS for them.
Pretty much any scheme you come up with has so many ways around it or possible abuses that it'd be more dangerous than the problem itself. Even if it isn't determined programmatically, relying on some degree of user interaction or target selection, it is likely to be open to abuse.
You do know that for technical issues, especially technical issues that have anything to do with piracy, the Beeb is little better than a press-release reprint agency, right?
Yeah, from time to time one or two of the jouralists manage to get up a more well researched argument, but for the most part if company X tells them that Y is "doing great things to fight piracy" or industry X (which they conveniently forget has just had record sales) tells them it is being "destroyed by piracy" then they'll just bung the story up without really doing any significant digging.
They're stored in protected databases.
If I found a colo company writing a password down, I'd never touch them with a 100 foot cat5 lead.
It seems everybody else hated it
Not me, I think it's rather good actually.
Not really.
The advert revenue on cable allows the cable company to reduces the cost to the subscriber*, effectively the cable subscriber is paying for their subscription in two ways: money and viewing time.
With cable internet it's a different kettle of fish: the subsciber's $49 goes to the cable company, but the revenue from the advertisement doesn't go anywhere near the cable company, it is used by the site maintainer to pay for bandwidth costs. In this case the cable internet subscriber is paying their subscription and the costs of a third party.
The two cases aren't really equivalent: the former is a simple trade of one cost for another, the latter is two costs - one from the cable company and one from the website owner.
* as long as you assume that the cost of the subscription really is > $49. Which it probably isn't, but such is the way of business.
Going off my current experience (reinstalling from a retail DVD right now after a windows reinstall), downloading it will take about as long as installing, decrypting and activating the damned thing. The only difference is that with retail you at least have something physical to show for it. Still takes sodding ages to install either way...
On a slightly different note, who knows where I can find the MEPs for London?
Go to the European Parliament, UK Office - UK MEPs page, click on London.
So now I know who to blame! Damn you, damn you and all your kind for the torture you put us poor buyers through! Must you insist on that barrier of plastic between us and our purchases so that we have to spend an age struggling to find a point that will admit a knifepoint or fingernail, only to have the plastic tear before removing any substantial amount of the wrapping?
;)
Damn you all!
Yet, if they insist Valve has to sell via steam for the same price as Vivendi do in the stores, then they are forcing Valve to take yet more profit from sales. That makes electronic distribution even more attractive to software developers and more of a threat to conventional publishers.
Normally I hate these sort of catch-22 situations. But not when it's a publisher having to deal with it, then I just laugh.
Brings a whole new meaning to digesting a film...
No, looks like a metaprediction. As such it does not have to be bound by the rules of prediction.
;)
Yeah, I need to work on this sense of humour thing...
Now we have all the technology we can imagine (well almost) so we have to dream about something else. ;)
You need to work on your imagination.
Hmm.. if we can assume that Maths := Dutch then, given that this is Dutch Maths, we can substitute Dutch for Maths to prove that ti is completely Double Dutch...
If so, why is this a problem for the University?
My guess, going off experience of university hall networks in the UK (one of the people I work with is a net admin in one) is that they are doing it so that they can keep an eye on what the students are doing and pull the access of students who are doing things they aren't allowed to do.
If the students set up their own parallel network outside the control of the university then it is possible that the university admins may get reports of activity they need to stop while being unable to actually take action against the student short of suspension or expulsion. This could create a range of problems for the university so they decide to prevent the establishment of a secondary network.
In short - I think this this isn't a technical problem, it's a legal arse-covering move using a technical issue as a cover.
Why in the world didn't the other scientists use this exact same setup when trying to reproduce the results? If you're trying to repeat a result, don't you make sure all variables are the same?
IANANP (nuclear physicist) but I'd expect that getting a 100% setup is difficult - you probably have to remove all impurities and make sure the distribution of atoms in the electrodes is correct (or, more likely, pay a significant amount to obtain materials to the required spec).
Besides which, scientists checking something that they are biased against anyway (either because they see no theoretical basis for the whole idea or because, if it was reproduced reliably, it would threaten their theories or jobs) aren't likely to spend any more than the minimum time and money necessary to say "well, that doesn't work, back to my tokamak". There's a LOT of money tied up in things that would become of questionable importance if cold fusion could be obtained reliably and scaled to useful levels.