When you print you mix ink in each pixel. However the cameras don't allow more than one pixel at each point. In order to get your 300dpi you need a 21.6MP camera.
I guess you're going to have to wait a few years before the fancy professional cameras can take advantage of consumer-level pritners.
Try scaling a photo up a bit, printing it to a decent printer, or doing any editing with photoshop/gimp and you'll really appreciate those 11MP.
I find that any resolution under 300dpi looks cheap, which is rarely the effect I'm trying to produce. Given that, a 2MP camera gives around 4"x4" which is or smaller than an ordinary photo.
This camera is a significant improvement, 11MP gives more like 10"^2, big enough for almost all uses.
FWIW, I'm still holding off buying a digital camera because my $300SLR is better than a $1000 digital camera. I can understand someone with a $2000SLR medium format camera saying much the same thing about this camera.
Why _do_ people buy Ximian?
on
Inside Ximian
·
· Score: 3, Interesting
I've often wondered why people bother with ximian. Are the packages it releases any better than the ones released by gnome itself?
Sure, it has a pretty autoupdate feature, but then so does debian and mandrake, and it can be added to redhat,.... And if you install it then your installation seems to be not quite compatible with a standard gnome install.
I can see why people would install gnome2 over kde3, although I personally prefer kde, but why would you install ximian gnome over normal gnome?
Is it yet another linux company that is going to crash and burn once it runs out of VC? Just what is there to encourage people to pay them money?
Consider SIMD, very valuable in consoles. It is useful for practically every graphics operation, from shading to panning to zooming to...
How many flops can an ordinary P4 do if you define the operations as changing a piece of data rather than seperate FPU instructions? 512 gigaflops if memory serves me right. That makes these less than ten times faster than current technology, which I find quite believable given it is a) not available for tw years and b) optimised for SIMD.
A simple (and cheap solution). Install a few quickcams and set them up on different webcam sites. Bandwidth requiremnts are within what DSL will give you.
Input can be handled by a normal switch, remote controlled via ethernet.
This is the stable release of the kernel (yes, I know you knew that already). That means it contains fixes, bug patches etc. for various exotic things (less exotic things would have been fixed before.19) and it doesn't contain any new significant features.
Now you say you want a changelog? And how exactly is someone to produce a changelog that shows you which bugfixes will apply to you? i386 only? Only affecting popular chipsets? Only affecting new hardware? Old hardware?
I guess my point is: There is no reason not to upgrade to 2.4.19. Unlike a major upgrade (e.g. 2.2 to 2.4) this is very, very unlikely to break anything.
It seems a number of people are posting here without understanding the issues. I won't atempt to say what I think is the right answer below, but I will attempt to fix the errors other posters are making.
Zimbabwe has a corrupt government: It drives white farmers off farms so there is little incentive for people with money to invest in agriculture; It rigs elections for political power. In many ways it makes Microsoft look like a friendly guy. It has nothing whatsoever to do with this decision.
Zimbabwe would love to accept the food. But just as Zimbabwe must feed its population, it must also protect its income and if a single farmer anywhere plants this corn it could destroy what remains of the country. If it sacrifices future earning potential in return for food then it has no chance of getting out of the third world ever.
This has nothing (directly) to do with Monsanto's patents on GE corn. A starving man will happily ignore his fears about GE being dangerous or his ideologies about patented food in order to feed his family. Perhaps the EU could be more tolerant about accepting GE imports, but then perhaps the US could be more willing to supply consumers with what they want.
Normal corn is not genetically engineered. It is crossbred but it has only ever been crossbred with other grains, never with soya beans or frogs. It may be that crossing it with these things doesn't make it any more dangerous and the EU's policy is unnecessary caution, or it might not be. Either way there is a difference between GM food and selective breeding.
I hope that clears up some of the FUD being posted. It still leaves open a number of possible solutions:
Mill the corn first so nobody can plant it -- as I said above, starving people are quite happy eating GE food.
Change the EU laws so it is willing to risk contamination.
Give them non GE food instead of GE corn.
Probably others I haven't thought of
Oh, and if GE grain is shipped to NZ it is destroyed by customs; No sane exporter of food can dare import GE food. The market for GE food is just too small with the US on its own making ten times more than the total demand for GE food, it isn't just Zimbabwe that fears GE imports because of its export market.
the benefits of accurate timekeeping
on
Do You Have The Time?
·
· Score: 4, Informative
Personally I arbitarialy declare the firewall as having the same time and use cron to update everyone from that. Since latency between machines is almost equal, everybody is out by the same amount.
Before anybody thinks it is silly to keep clocks tightly synchronised, try running NFS without it and you'll run into no end of problems. Even as little as one second will cause errors with make. The key is that all clocks must read the same, not that they need to be correct.
Oh, and don't get fooled into thinking you can accurately synchronise against those atomic clocks. The algorithms they use to average results make a number of incorrect assumptions that will result in you being out by a small constant amount, about as much as if you'd synchronised off an ordinary clock.
Think about this for a minute, MS sells xboxen for around $200 where it would cost you $400 to make it using intel hardware. This means you only need a cluster of a little over 1000 to recoup your losses.
And in case anybody doesn't think a beauwolf cluster of xboxen has any value beyond the geek factor, consider this: an xbox may not have an impressive CPU, but it does have an impressive graphics card. Now, what do graphics cards do? That's right, they do polygon operations, And what do polygon operations look like? Like matrix multiplications that's what. Please tell me I don't need to spell out what matricies are for (Hint: think `everything')
I.e. for a tiny amount of money you can build a cluster able to do amazing amounts of matrix manipulation. GPG cracking, neural net simulation, raytracing,... here we come!
The easiest way of seeing if this is true or just propoganda to justify your actions is to try substituing names and see if it still holds. Let's try that here:
OBL isn't in trouble for being a religious whacko, he's in trouble for instigating the murder of several thousand innocent people.
George Bush isn't in trouble for being a religious whacko, he's in trouble for instigating the murder of several thousand innocent people.(Think of the thousands of dead muslims civilians that were drafted against the American invasion of Afganastan.)
It doesn't matter whether he did it because he actually believes any of the tripe he spews to his cannon-fodder, or if he did it just for the financial gains from increasing arms sales. We need him dead, simply because he deserves it.
Well, to avoid being labelled a troll, judge for yourself.
[google]There is a very famous person by that name. Her official website is [here]. [Here] is a list of fansites and [here] are some other sites which discuss her. That name is mentioned in [these] sites, but it is unclear if they are talking about the same person. [Here] is a list of other people with that name.
[user]The person I am looking for isn't famous.
[google]Then you are probably looking for one of [these] people.
Actually, this isn't anywhere near as hard as you make it out. I have worked on a number of projects that have this as the goal. While perfect language understanding is a long way off, problems like this require only basic understanding.
Specificially, when you index pages you compare them using inverse document frequency (IDF). By iterating it is relativly trivial to split the related documents into pretty little folders, infogistics is one example, and I've used a demo of a different one that draws a graph of related documents. Interestingly, the complex boolean queries require AltaVista as a backend. Structured searches like this is the next step search engines, I'd be betting on seeing it outside the lab some time year depending on VC funding.
The level of interaction you're talking about is also obtainable. Here (for text users: infinity.otago.ac.nz:8080/teKaitito) we have a reasearch project that does exactly this and more, athough there isn't much use visiting it since we have most of the functionality hidden. A similar project provides an interactive tour of a museum.
A web search for resolution and inference engines should convince you this is a current research topic and should be widely available quite soon. So really all you're asking is for a good inference engine to be integrated into a state-of-the-art search engine. By far the hardest part will be getting the CPU requirements down. I'd say it will take two years at maximum
Yes they do seriously rival RedHat. In my experience a smaller product never succeeds unless it is better, at least for some people
Each of the distros has different advantages, and if you want to take RedHat as a basline I could say Debian has better package management, Slackware is more configurable, Mandrake is more polished, etc. etc. Of course, they all have disadvantages relative to redhat too - e.g. debian's installer.
As for competition shooting linux in the foot. I honestly think the reason Linux has been able to catch up to MS in the desktop market is because MS hasn't had enough competition to bother innovating. Besides, all the distros are based on the GNU public licence, so whenever one invents something the others can copy it freely.
Do you think KDE would be as good as it is if gnome wasn't around? having multiple choices means if one project fails then another succeeds.
Slashdot Mirror
When you print you mix ink in each pixel. However the cameras don't allow more than one pixel at each point. In order to get your 300dpi you need a 21.6MP camera.
I guess you're going to have to wait a few years before the fancy professional cameras can take advantage of consumer-level pritners.
How about a tripod and an external trigger? _I_ might not have nerves of steel but that doesn't mean the camera has to move a millimeter.
There are times when you want a high quality camera and can't set up the shot, but they are probably the minority of cases.
You're right though, faster reading of the light would be very useful. It isn't like we need to wait for the film to develop anymore.
Try scaling a photo up a bit, printing it to a decent printer, or doing any editing with photoshop/gimp and you'll really appreciate those 11MP.
I find that any resolution under 300dpi looks cheap, which is rarely the effect I'm trying to produce. Given that, a 2MP camera gives around 4"x4" which is or smaller than an ordinary photo.
This camera is a significant improvement, 11MP gives more like 10"^2, big enough for almost all uses.
FWIW, I'm still holding off buying a digital camera because my $300SLR is better than a $1000 digital camera. I can understand someone with a $2000SLR medium format camera saying much the same thing about this camera.
I've often wondered why people bother with ximian. Are the packages it releases any better than the ones released by gnome itself?
.... And if you install it then your installation seems to be not quite compatible with a standard gnome install.
Sure, it has a pretty autoupdate feature, but then so does debian and mandrake, and it can be added to redhat,
I can see why people would install gnome2 over kde3, although I personally prefer kde, but why would you install ximian gnome over normal gnome?
Is it yet another linux company that is going to crash and burn once it runs out of VC? Just what is there to encourage people to pay them money?
Corrin (sounding really like a troll...)
With [ADV] in the subject line it is trivial to write filter rules that trash the mail. Even Outlook can perform that sort of filtering (I think?).
Consider SIMD, very valuable in consoles. It is useful for practically every graphics operation, from shading to panning to zooming to ...
How many flops can an ordinary P4 do if you define the operations as changing a piece of data rather than seperate FPU instructions? 512 gigaflops if memory serves me right. That makes these less than ten times faster than current technology, which I find quite believable given it is a) not available for tw years and b) optimised for SIMD.
A simple (and cheap solution). Install a few quickcams and set them up on different webcam sites. Bandwidth requiremnts are within what DSL will give you.
Input can be handled by a normal switch, remote controlled via ethernet.
Now you say you want a changelog? And how exactly is someone to produce a changelog that shows you which bugfixes will apply to you? i386 only? Only affecting popular chipsets? Only affecting new hardware? Old hardware?
I guess my point is: There is no reason not to upgrade to 2.4.19. Unlike a major upgrade (e.g. 2.2 to 2.4) this is very, very unlikely to break anything.
Zimbabwe has a corrupt government: It drives white farmers off farms so there is little incentive for people with money to invest in agriculture; It rigs elections for political power. In many ways it makes Microsoft look like a friendly guy. It has nothing whatsoever to do with this decision.
Zimbabwe would love to accept the food. But just as Zimbabwe must feed its population, it must also protect its income and if a single farmer anywhere plants this corn it could destroy what remains of the country. If it sacrifices future earning potential in return for food then it has no chance of getting out of the third world ever.
This has nothing (directly) to do with Monsanto's patents on GE corn. A starving man will happily ignore his fears about GE being dangerous or his ideologies about patented food in order to feed his family. Perhaps the EU could be more tolerant about accepting GE imports, but then perhaps the US could be more willing to supply consumers with what they want.
Normal corn is not genetically engineered. It is crossbred but it has only ever been crossbred with other grains, never with soya beans or frogs. It may be that crossing it with these things doesn't make it any more dangerous and the EU's policy is unnecessary caution, or it might not be. Either way there is a difference between GM food and selective breeding.
I hope that clears up some of the FUD being posted. It still leaves open a number of possible solutions:
Oh, and if GE grain is shipped to NZ it is destroyed by customs; No sane exporter of food can dare import GE food. The market for GE food is just too small with the US on its own making ten times more than the total demand for GE food, it isn't just Zimbabwe that fears GE imports because of its export market.
Personally I arbitarialy declare the firewall as having the same time and use cron to update everyone from that. Since latency between machines is almost equal, everybody is out by the same amount.
Before anybody thinks it is silly to keep clocks tightly synchronised, try running NFS without it and you'll run into no end of problems. Even as little as one second will cause errors with make. The key is that all clocks must read the same, not that they need to be correct.
Oh, and don't get fooled into thinking you can accurately synchronise against those atomic clocks. The algorithms they use to average results make a number of incorrect assumptions that will result in you being out by a small constant amount, about as much as if you'd synchronised off an ordinary clock.
Think about this for a minute, MS sells xboxen for around $200 where it would cost you $400 to make it using intel hardware. This means you only need a cluster of a little over 1000 to recoup your losses.
... here we come!
And in case anybody doesn't think a beauwolf cluster of xboxen has any value beyond the geek factor, consider this: an xbox may not have an impressive CPU, but it does have an impressive graphics card. Now, what do graphics cards do? That's right, they do polygon operations, And what do polygon operations look like? Like matrix multiplications that's what. Please tell me I don't need to spell out what matricies are for (Hint: think `everything')
I.e. for a tiny amount of money you can build a cluster able to do amazing amounts of matrix manipulation. GPG cracking, neural net simulation, raytracing,
OBL isn't in trouble for being a religious whacko, he's in trouble for instigating the murder of several thousand innocent people.
George Bush isn't in trouble for being a religious whacko, he's in trouble for instigating the murder of several thousand innocent people.(Think of the thousands of dead muslims civilians that were drafted against the American invasion of Afganastan.)
It doesn't matter whether he did it because he actually believes any of the tripe he spews to his cannon-fodder, or if he did it just for the financial gains from increasing arms sales. We need him dead, simply because he deserves it.
Well, to avoid being labelled a troll, judge for yourself.
[google]There is a very famous person by that name. Her official website is [here]. [Here] is a list of fansites and [here] are some other sites which discuss her. That name is mentioned in [these] sites, but it is unclear if they are talking about the same person. [Here] is a list of other people with that name.
[user]The person I am looking for isn't famous.
[google]Then you are probably looking for one of [these] people.
Actually, this isn't anywhere near as hard as you make it out. I have worked on a number of projects that have this as the goal. While perfect language understanding is a long way off, problems like this require only basic understanding.
Specificially, when you index pages you compare them using inverse document frequency (IDF). By iterating it is relativly trivial to split the related documents into pretty little folders, infogistics is one example, and I've used a demo of a different one that draws a graph of related documents. Interestingly, the complex boolean queries require AltaVista as a backend. Structured searches like this is the next step search engines, I'd be betting on seeing it outside the lab some time year depending on VC funding.
The level of interaction you're talking about is also obtainable. Here (for text users: infinity.otago.ac.nz:8080/teKaitito) we have a reasearch project that does exactly this and more, athough there isn't much use visiting it since we have most of the functionality hidden. A similar project provides an interactive tour of a museum.
A web search for resolution and inference engines should convince you this is a current research topic and should be widely available quite soon. So really all you're asking is for a good inference engine to be integrated into a state-of-the-art search engine. By far the hardest part will be getting the CPU requirements down. I'd say it will take two years at maximum
Yes they do seriously rival RedHat. In my experience a smaller product never succeeds unless it is better, at least for some people
Each of the distros has different advantages, and if you want to take RedHat as a basline I could say Debian has better package management, Slackware is more configurable, Mandrake is more polished, etc. etc. Of course, they all have disadvantages relative to redhat too - e.g. debian's installer.
As for competition shooting linux in the foot. I honestly think the reason Linux has been able to catch up to MS in the desktop market is because MS hasn't had enough competition to bother innovating. Besides, all the distros are based on the GNU public licence, so whenever one invents something the others can copy it freely.
Do you think KDE would be as good as it is if gnome wasn't around? having multiple choices means if one project fails then another succeeds.