A suggestion for fixing this problem is to open the gates. Come right in. Anyone from a war torn country, middle east, wherever, come on over if you can get yourself here. Only catch: You must spend the first 15 years here in a rural outback town or property and not come within 200km of a city. It's thought to be the best way to 'solve' the country shopper / boat people / illegal immigrant problem.
It seems to me like it would just replace the "illegal immigrant" problem with an "illegal internal migrant" problem. Would you put passport controls at the city gates or something?!
Seriously though megafauna are relatively easy to control because they breed slowly and can't really hide. It's the small animals you have to worry about.
Specifying the "document character set" as unicode means that even if the charset you are writing your document in doesn't support the character you want you can still enter it as a numeric (or named if one is defined) entity, whether it will be displayed is mostly a matter of whether appropriate fonts are installed but generally i'd expect someone who writes Chinese to have Chinese fonts installed.
Generally it's the GUI system's job to handle input and output of text not an individual application. Is it reasonable to expect browsers to ship a massive font full mostly of characters that most of it's users will either fond meaningless or have already? Is it reasonable to expect browsers to implement their own input methods in case the operating system's one is defficient? Is it reasonable to expect them to implement their own font rendering for the same reason?
I'm pretty sure in HTML5 like in HTML4 the document is considered to be made up of unicode characters and other charsets are considered as encodings of unicode. Of course the HTML5 spec doesn't include all unicode characters explicitly that would be insane.
Valve has proven that while gamers will bitch about it they will put up with anti-resale measures. Heck if you make the service you are locking them into treat them well (unlimeted redownloads, massive discounts in sales etc) then they might even grow to like it (I'd say I see far more positive posts on steam round here nowadays than negative ones).
I guess the only questions left for devs are
1: which is more controversial and/or more profitable. Preventing resale altogether as valve and blizzard do or trying to cream some money off resale through content that is free to the first owner but chargable to subsequent owners. 2: how to balance strength of the protection measures verses inconviniance caused to user.
Note that there is a difference between wiring adaptors and actual signal converters.
A DVI-I connector (which is what you find on most equipment) has both analog and digital signals. The analog signals are compatible with VGA and the digital signals are compatible with HDMI. So you can connect a DVI-I device to either a VGA device or a HDMI device with cheap wiring adaptors.
But if you want to connect a VGA monitor to something that only has a HDMI output (or a DVI-D output) you will need a far more expensive active converter. Afaict those are sufficiently expensive that you can sometimes find whole monitors on special offer for not much more.
Yes, but the topic of discussion is power consumption not purchase price.
Power consumption is part of it but I don't think you can draw reasonable conclusions from power consumption alone. It's important but so are upfront cost and flexibility.
My point was that FPGAs do not solve the problem of power consumption - at least not yet. They are getting better but then so are ASICs.
Yes ASICs are the most power efficient way of performing a repetitive computation task because they have neither the data pushing overhead of CPUs/GPUs or the reconfigurable wiring overhead of FPGAs. However putting a design into an ASIC is expensive, so it's only practical if you want a lot of a copies of the design, plan to run each copy for a long time and are bloody sure you have got your design right. For many many workloads ASICs are out of the question.
So the more interesting question is how do FPGAs compare to CPUs and GPUs and that is going to be workload specific. IIRC the bitcoin miners found that FPGAs could give far more performance per watt than either CPUs or GPUs in their application BUT the upfront costs made it a hard sell. I'd expect crypto cracking to be similar.
It appears that, looking forward, the best solution will be a combination of the two techniques. Specialized ASIC components glued together with FPGA elements.
mmm, for example for floating point heavy tasks you could have an floating point blocks (adders, multipliers, dividers and so on) with FPGA like routing to move signals between them and some regular FPGA style register/logic units to control them.
Some simulations that we do today are performed thousands of times faster than "reality" while most others that we do today are performed thousands, or even millions of times slower. The speed of the simulation is irrelevant to their existence
For a simulation to be useful it must reach desired results in a reasonable amount of time. If you are simulating something that only takes a few milliseconds in real life then a simulation that runs a thousand times slower than reality will still feel basically instant and one a million times slower will be done in around an hour. OTOH if you are simulating something that takes years in real life then with a 1000 times slowdown your simulation will be running for millenia.
IMO email and the web are examples of how internet services should work. I can choose whether to use my ISP, a commercial provider, an advert supported provider or run it myself. Whichever I choose I can exchange mail with people who made other choices and those people can view my websites. If I pay a little for my own domain I can have email/website addresses which I get to keep as and when I choose to switch between the options.
I find the trend towards lockin based servies to be a disturbing change in internet communications. Competition should be based on quality of service provided, not on who can lock in the most users fastest.
Sure ultimately everything a uni pays for is going to be funded out of either tuition or research income. The point is individual academics and students don't directly pay for said journal or software access and the marginal cost of downloading an extra article or installing a copy of office on a new uni machine is zero so noone has any motivation to avoid doing it.
Elsevier tries very hard to extract subscription fees from their journals while keeping the cost invisiable to individual academics. Unless you or your institution buy a subscription to a load of them together you are likely to end up paying $10-30 or more for each paper that you might want to read. Most people, when they encounter this kind of paywall will just go elsewhere and read someone else's related work but other academics most likely won't run into that paywall and will keep citing your work.
FTFY
This gives the big publishers like Elsevier and the IEEE a huge advantage over either small closed access journals (which charge readers directly because they can't get the big institutional subs) and the open acces journals (which charge submitters since they can't get any money from readers).
Replacing an abandoned journal is rather different from trying to displace a journal by force. Setting up the website is easy, even finding reviewers is probablly not that hard. The difficult bit is convinving people to chose your journal over the established one. Oh and someone has to pay for your new journal (afaict reviewers do get paid even if it's only a nominal ammount) so if you are open access you will probablly have to charge authours to cover the cost of peer review. If you aren't open access and aren't affiliated with one of the big publishers (see below) you will have a hard time getting people to read your papers.
It's important to realise that individual academics and students within instituations don't directly pay for access to most papers from our budgets just like we don't directly pay for "core" software (we do pay for some more specialised software out of our own budgets but windows, office, matlab, endnote and so on are all covered centrally). Those things are paid for centrally as part of block subscriptions. If academics actually had to pay the prices that are shown to the general public I suspect there would be a very quick move towards open access journals.
How far does that £18 50ml pot actually go? I've seen similar products (pens and expoxies at least, not sure about paints) before but their downfall was they were silver based and hence expensive (tolerable for a repair job but you wouldn't want to do a whole board or similar with them).
Today, native code should be reserved for a few niche uses.
Unfortunately every platform seems to be pushing their own languages. Google is pushing a basterdised version of java. Apple is pushing objective C. MS is pushing.net*. Symbian uses a somewhat stripped down (no exception support) variant of C++. Afaict winCe/WM was traditionally programmed in C++ and VB though it seems later in it's life the platform got.net support too. The common ground between all these platforms is to write the core logic of your app in the well-supported subset of C++ (that is don't rely on exceptions) and then write the platform specific UI code in the platform's language of choice.
MS is basically telling all developers who want to port to WP7 from other mobile platforms that they will have to rewrite more-or-less from scratch. That seems like a suicidal move when you only have a tiny marketshare.
If MS offered a compiler that could take standard C++ and turn it into.net bytecode suitable for running on all their platforms (rather than only on platforms where running "unsafe" code is allowed) then the inability to run native code wouldn't be an issue but afaict they don't do that.
*yes I know.net is not technically a language but the main languages for.net are different from the main languages for other bytecode platforms.
Ferom TFA it seems the workers themselves are deliberately cricumventing the exposure measurements so they can earn more money before they are laid off for hitting their raditation quota.
It's not about "upstreams and peers", Internet transit is only a few dollars per month per megabit nowadays and due to the increasing dynamics of the web you can't cache all that much anyway. That is why proxies are so rare at fixed line providers nowadaysdays.
Afaict the main reason mobile providers run proxies is so that they can recompess images at lower quality and hence save bandwidth on the wireless side (aiui wireless bandwidth is FAR more expensive than internet bandwidth).
It seems to me like it would be a big win for ssl sites.
Consider a browser visits a ssl site. First they have to open a ssl connection and download the page. Then they have a few choices.
1: download all the items sequentially using the same connection (obviously slow and bad) 2: open multiple connections (wasting time and processing power on multiple ssl handshakes) 3: use pipelining and hope they don't hit an item that is slow to retrive and blocks everything else.
With spdy they can just open one ssl connection and do everything over it without having requests block each other.
You wouldn't print your mould directly, you'd print a model, then pack moulding sand round the model to form a mould. Then remove the model and pour in the metal.
Depending on the complexity of the part you could either go for a reusable model or a model that you melt out of the mould before using it. (like with lost wax casting)
Though if it's a mechanical part you would probablly need to do some machining after casting to get tight enough tolerances on the important surfaces.
Note that with a little tweaking of the position of various panel applets you can make current versions of XFCE act fairly similar to gnome 2. That is probablly the way i'm going to be going with my linux systems moving forward.
I wonder, can you go to individual authors asking for permission to distribute, like ZX Spectrum guys [worldofspectrum.org] do, or Nintendo has a say in it, so you might as well go and get fucked?
I don't know for sure but I would bet that for most games on their platform nintendo likely owns some rights to them either through explicit agreements or through the use of nintendo library code in the games (remember games in those days were built into a single big binary).
It takes effort to find a graphics card that supports it nowadays
Do you call going to the graphics card section on newegg and sorting by price effort?
I don't think i've ever seen a PC graphics card that couldn't drive an analog monitor (though I have seen onboard graphics without anlog output both on PCs and embedded devices). Low end cards tend to actually have a VGA port on the card itself. Higher end cards usually require you to plug a passive adaptor into their DVI ports to get at the analog signal.
I haven't seen a recent board with a parallel port.
heh, I ended up with paralell and serial ports on the back plate of my most recent board (a low end gigabyte Z68 ATX board, I don't remember the exact model offhand) without even explicitly looking for them.
One thing I do notice is that the higher end a board is the closer to "legacy free" it seems to be, presumablly because they want the space for other things.
It's a pretty basic RGB analog video signal with seperate H and V sync lines (hence the acronym RGBHV), modern ports also have some control lines but you can get a video signal through without them.
but doesn't higher monitor resolution require more (serial?) bandwidth?
The minimum pixel rate in an analog video signal to acheive a given resoloution would be (horizontal resoloution)*(vertical resoloution)*(refesh rate), in reality the pixel rate will be a bit higher to allow for blanking periods between lines and between frames. So the pixel rate for full HD at 60Hz will be arround 126 megasamples per second, that makes the worst case fundamental* of the signal 63Hz. For decent quality you probablly want to cover at least up to the third harmonic of the worst case fundamental so for 1080p you need a cable that is good from DC to arround 200 MHz. That is well within what pretty much any half decent coax can do.
There is really no technical reason that a "VGA" connection can't deliver a decent signal at very high resoloution provided all the components (source, target and cable) are of sufficient quality.
* That is what the fundamental frequency would be if the screen was disabling alternating black and white pixels.
Slashdot Mirror
A suggestion for fixing this problem is to open the gates. Come right in. Anyone from a war torn country, middle east, wherever, come on over if you can get yourself here. Only catch: You must spend the first 15 years here in a rural outback town or property and not come within 200km of a city.
It's thought to be the best way to 'solve' the country shopper / boat people / illegal immigrant problem.
It seems to me like it would just replace the "illegal immigrant" problem with an "illegal internal migrant" problem. Would you put passport controls at the city gates or something?!
Seriously though megafauna are relatively easy to control because they breed slowly and can't really hide. It's the small animals you have to worry about.
Specifying the "document character set" as unicode means that even if the charset you are writing your document in doesn't support the character you want you can still enter it as a numeric (or named if one is defined) entity, whether it will be displayed is mostly a matter of whether appropriate fonts are installed but generally i'd expect someone who writes Chinese to have Chinese fonts installed.
Generally it's the GUI system's job to handle input and output of text not an individual application. Is it reasonable to expect browsers to ship a massive font full mostly of characters that most of it's users will either fond meaningless or have already? Is it reasonable to expect browsers to implement their own input methods in case the operating system's one is defficient? Is it reasonable to expect them to implement their own font rendering for the same reason?
I'm pretty sure in HTML5 like in HTML4 the document is considered to be made up of unicode characters and other charsets are considered as encodings of unicode. Of course the HTML5 spec doesn't include all unicode characters explicitly that would be insane.
XFCE is also worth considering, it seems to be based round much the same principles as gnome2 was though it's not as pretty.
Valve has proven that while gamers will bitch about it they will put up with anti-resale measures. Heck if you make the service you are locking them into treat them well (unlimeted redownloads, massive discounts in sales etc) then they might even grow to like it (I'd say I see far more positive posts on steam round here nowadays than negative ones).
I guess the only questions left for devs are
1: which is more controversial and/or more profitable. Preventing resale altogether as valve and blizzard do or trying to cream some money off resale through content that is free to the first owner but chargable to subsequent owners.
2: how to balance strength of the protection measures verses inconviniance caused to user.
well, we've been there before (DVI to VGA).
Note that there is a difference between wiring adaptors and actual signal converters.
A DVI-I connector (which is what you find on most equipment) has both analog and digital signals. The analog signals are compatible with VGA and the digital signals are compatible with HDMI. So you can connect a DVI-I device to either a VGA device or a HDMI device with cheap wiring adaptors.
But if you want to connect a VGA monitor to something that only has a HDMI output (or a DVI-D output) you will need a far more expensive active converter. Afaict those are sufficiently expensive that you can sometimes find whole monitors on special offer for not much more.
Yes, but the topic of discussion is power consumption not purchase price.
Power consumption is part of it but I don't think you can draw reasonable conclusions from power consumption alone. It's important but so are upfront cost and flexibility.
My point was that FPGAs do not solve the problem of power consumption - at least not yet. They are getting better but then so are ASICs.
Yes ASICs are the most power efficient way of performing a repetitive computation task because they have neither the data pushing overhead of CPUs/GPUs or the reconfigurable wiring overhead of FPGAs. However putting a design into an ASIC is expensive, so it's only practical if you want a lot of a copies of the design, plan to run each copy for a long time and are bloody sure you have got your design right. For many many workloads ASICs are out of the question.
So the more interesting question is how do FPGAs compare to CPUs and GPUs and that is going to be workload specific. IIRC the bitcoin miners found that FPGAs could give far more performance per watt than either CPUs or GPUs in their application BUT the upfront costs made it a hard sell. I'd expect crypto cracking to be similar.
It appears that, looking forward, the best solution will be a combination of the two techniques. Specialized ASIC components glued together with FPGA elements.
mmm, for example for floating point heavy tasks you could have an floating point blocks (adders, multipliers, dividers and so on) with FPGA like routing to move signals between them and some regular FPGA style register/logic units to control them.
Some simulations that we do today are performed thousands of times faster than "reality" while most others that we do today are performed thousands, or even millions of times slower. The speed of the simulation is irrelevant to their existence
For a simulation to be useful it must reach desired results in a reasonable amount of time. If you are simulating something that only takes a few milliseconds in real life then a simulation that runs a thousand times slower than reality will still feel basically instant and one a million times slower will be done in around an hour. OTOH if you are simulating something that takes years in real life then with a 1000 times slowdown your simulation will be running for millenia.
IMO email and the web are examples of how internet services should work. I can choose whether to use my ISP, a commercial provider, an advert supported provider or run it myself. Whichever I choose I can exchange mail with people who made other choices and those people can view my websites. If I pay a little for my own domain I can have email/website addresses which I get to keep as and when I choose to switch between the options.
I find the trend towards lockin based servies to be a disturbing change in internet communications. Competition should be based on quality of service provided, not on who can lock in the most users fastest.
Sure ultimately everything a uni pays for is going to be funded out of either tuition or research income. The point is individual academics and students don't directly pay for said journal or software access and the marginal cost of downloading an extra article or installing a copy of office on a new uni machine is zero so noone has any motivation to avoid doing it.
Elsevier tries very hard to extract subscription fees from their journals while keeping the cost invisiable to individual academics. Unless you or your institution buy a subscription to a load of them together you are likely to end up paying $10-30 or more for each paper that you might want to read. Most people, when they encounter this kind of paywall will just go elsewhere and read someone else's related work but other academics most likely won't run into that paywall and will keep citing your work.
FTFY
This gives the big publishers like Elsevier and the IEEE a huge advantage over either small closed access journals (which charge readers directly because they can't get the big institutional subs) and the open acces journals (which charge submitters since they can't get any money from readers).
Replacing an abandoned journal is rather different from trying to displace a journal by force. Setting up the website is easy, even finding reviewers is probablly not that hard. The difficult bit is convinving people to chose your journal over the established one. Oh and someone has to pay for your new journal (afaict reviewers do get paid even if it's only a nominal ammount) so if you are open access you will probablly have to charge authours to cover the cost of peer review. If you aren't open access and aren't affiliated with one of the big publishers (see below) you will have a hard time getting people to read your papers.
It's important to realise that individual academics and students within instituations don't directly pay for access to most papers from our budgets just like we don't directly pay for "core" software (we do pay for some more specialised software out of our own budgets but windows, office, matlab, endnote and so on are all covered centrally). Those things are paid for centrally as part of block subscriptions. If academics actually had to pay the prices that are shown to the general public I suspect there would be a very quick move towards open access journals.
You'd probably get better and cheaper results by more conventional methods such as etching or milling (both of which CAN be done at home).
How far does that £18 50ml pot actually go? I've seen similar products (pens and expoxies at least, not sure about paints) before but their downfall was they were silver based and hence expensive (tolerable for a repair job but you wouldn't want to do a whole board or similar with them).
Today, native code should be reserved for a few niche uses.
Unfortunately every platform seems to be pushing their own languages. Google is pushing a basterdised version of java. Apple is pushing objective C. MS is pushing .net*. Symbian uses a somewhat stripped down (no exception support) variant of C++. Afaict winCe/WM was traditionally programmed in C++ and VB though it seems later in it's life the platform got .net support too. The common ground between all these platforms is to write the core logic of your app in the well-supported subset of C++ (that is don't rely on exceptions) and then write the platform specific UI code in the platform's language of choice.
MS is basically telling all developers who want to port to WP7 from other mobile platforms that they will have to rewrite more-or-less from scratch. That seems like a suicidal move when you only have a tiny marketshare.
If MS offered a compiler that could take standard C++ and turn it into .net bytecode suitable for running on all their platforms (rather than only on platforms where running "unsafe" code is allowed) then the inability to run native code wouldn't be an issue but afaict they don't do that.
*yes I know .net is not technically a language but the main languages for .net are different from the main languages for other bytecode platforms.
Ferom TFA it seems the workers themselves are deliberately cricumventing the exposure measurements so they can earn more money before they are laid off for hitting their raditation quota.
It's not about "upstreams and peers", Internet transit is only a few dollars per month per megabit nowadays and due to the increasing dynamics of the web you can't cache all that much anyway. That is why proxies are so rare at fixed line providers nowadaysdays.
Afaict the main reason mobile providers run proxies is so that they can recompess images at lower quality and hence save bandwidth on the wireless side (aiui wireless bandwidth is FAR more expensive than internet bandwidth).
It seems to me like it would be a big win for ssl sites.
Consider a browser visits a ssl site. First they have to open a ssl connection and download the page. Then they have a few choices.
1: download all the items sequentially using the same connection (obviously slow and bad)
2: open multiple connections (wasting time and processing power on multiple ssl handshakes)
3: use pipelining and hope they don't hit an item that is slow to retrive and blocks everything else.
With spdy they can just open one ssl connection and do everything over it without having requests block each other.
You wouldn't print your mould directly, you'd print a model, then pack moulding sand round the model to form a mould. Then remove the model and pour in the metal.
Depending on the complexity of the part you could either go for a reusable model or a model that you melt out of the mould before using it. (like with lost wax casting)
Though if it's a mechanical part you would probablly need to do some machining after casting to get tight enough tolerances on the important surfaces.
Note that with a little tweaking of the position of various panel applets you can make current versions of XFCE act fairly similar to gnome 2. That is probablly the way i'm going to be going with my linux systems moving forward.
I wonder, can you go to individual authors asking for permission to distribute, like ZX Spectrum guys [worldofspectrum.org] do, or Nintendo has a say in it, so you might as well go and get fucked?
I don't know for sure but I would bet that for most games on their platform nintendo likely owns some rights to them either through explicit agreements or through the use of nintendo library code in the games (remember games in those days were built into a single big binary).
It takes effort to find a graphics card that supports it nowadays
Do you call going to the graphics card section on newegg and sorting by price effort?
I don't think i've ever seen a PC graphics card that couldn't drive an analog monitor (though I have seen onboard graphics without anlog output both on PCs and embedded devices). Low end cards tend to actually have a VGA port on the card itself. Higher end cards usually require you to plug a passive adaptor into their DVI ports to get at the analog signal.
I haven't seen a recent board with a parallel port.
heh, I ended up with paralell and serial ports on the back plate of my most recent board (a low end gigabyte Z68 ATX board, I don't remember the exact model offhand) without even explicitly looking for them.
One thing I do notice is that the higher end a board is the closer to "legacy free" it seems to be, presumablly because they want the space for other things.
I'm not too technical with how VGA works
It's a pretty basic RGB analog video signal with seperate H and V sync lines (hence the acronym RGBHV), modern ports also have some control lines but you can get a video signal through without them.
but doesn't higher monitor resolution require more (serial?) bandwidth?
The minimum pixel rate in an analog video signal to acheive a given resoloution would be (horizontal resoloution)*(vertical resoloution)*(refesh rate), in reality the pixel rate will be a bit higher to allow for blanking periods between lines and between frames. So the pixel rate for full HD at 60Hz will be arround 126 megasamples per second, that makes the worst case fundamental* of the signal 63Hz. For decent quality you probablly want to cover at least up to the third harmonic of the worst case fundamental so for 1080p you need a cable that is good from DC to arround 200 MHz. That is well within what pretty much any half decent coax can do.
There is really no technical reason that a "VGA" connection can't deliver a decent signal at very high resoloution provided all the components (source, target and cable) are of sufficient quality.
* That is what the fundamental frequency would be if the screen was disabling alternating black and white pixels.