I just went to the google accelerated page for Reddit on my iphone safari browser. Guess what? It's no longer accessible in safari! google forces you to use chrome to reach it. Google is breaking the internet.
If you get a.dev page then after a year it's going to not respond to https. it's will be only a google accelerated page only and only vailble by Ghttp protocol.
I totally get the idea of open access. And it's laudable. It's also a choice. You have approximately three choices when you publish: 1. publish in an open access journal, and pay it's editorial and compositional page charges,
2. Publish in a pay-for access journal, which may have lower compositional page charges
3. publish in some open and free journal (are there any good ones?)
If you go to a pay-for-access journal and ask them to make all of your author's papers free to access you are basically insane. Sure you can ask but you are asking the journal to go bankrupt or at a minimum work at cost. If you add onto that a request to reduce page charges too, well.... At this point you should just ask for a pony as well.
One could imagine that journals should pay authors for their articles. THat makes some logical sense but it just shifts the cost to the access.
I like the pay-to-publish model myself because if there is barrier it can, if used correctly and not as a vanity press, result in a journal I'm more likely to want to read and more proud to publish in too.
The right argument is if in the age of digital publishing we could not find some less expensive process. But that's not what UC is asking.
But the key thing to keep sight of is that the editorial process should try to stamp out crap. That's the whole reason I'm willing to pay. I can't read everything and if every search term has a load of crap then it's useless.
However that's not hopeless. Google ate alta vista's lunch because it provided more relevant searches. So it is possible to beat down the cost and still beat down crap. However, when it comes to science articles I still prefer peer review to key term search as a way to beat down the crap.
Finally, UC should consider just requiring it's authors to put their articles in a non-copyrighted form on Xarciv before sending them to elsevier. They won't save money but if they genuinely want free access to all UC author pubs it's already available to them. I think it's all about money and not about free access
I love sky effects. There's lots of extra physics one can see in rainbows that most people never see. As described in the summary, this is a well known effect known as supe numary rainbows. They arise from multipathing in the dorplet as decribed. You see these all the time in thin clouds but you may not have noiced. Next time you see thin clouds look for oscilating purple and teal bands. Them's the supernumary bows. And next time you see a double rainbow look at it again. First he second rainbow going the reverse direction. This tells you the bow is not just a second rainbow in a different place than the first, but it's different physics. Second look at the gap between the bows. It's not only always the same size, it's also subtly darker than the rest of the sky. What's there is an infrared rainbow. You just can't see it. And on a cold day with some high very light clouds. Look off to the side of the sun. You might see bright spots in the sky on either side. THis tells you there are elingated hexagonal ice spikes in the air because it's cold up there. My favorite effect of all that makes my kids really wonder is the effect called "jesus rays" but I'm not going to spoil it for you cause it's cool when the penny drops.
I can't tell you why they were sloppy but I can tell you a few things that might contribute. first, voting systems have some rules they follow. It varies from state to state but typically you are not allowed to alter any code in these things within a certain number of months of the election. This has led, more than once, to situations where a known bug (e.g. overflows of vote counts, or vulnerabilities) exists in the code but they cannot legally patch it. FOr systems used in federal elections they are/were supposed to get the blessing of the federal election assistance commissions which mandates code quality reviews (these are a total joke as they look at programming styles not logic) and that adds more time. IN fact vendors in two states were caught fixing bugs prior to elections and they knew they were no supposed. The problem is what do you do if the bug is really really awful? That'a why the vendors risked getting caught. One assumes this happens a lot and they don't get caught too.
Second, voting machines don't make a lot of money compared to gaming. So there's a lot of corner cutting. More importantly, it's also a race to market. After states buy a brand of machine they are not going to switch to your brand for a decade. So late to market is death. Better to run with a crappy system. That's also presumably why these things have been built on things like windows CE and Windows Access data bases and commodity OEM computing platforms inside. Quicker to market. Security last.
This you can see even if they did catch the problem they were hosed about getting it fixed promptly.
How do they handle the heat? Many nvnm desktop solid state storage have heat sinks and they are bigger too.
As for memory versus storage this used to be a continuum . Old computers leaned heavily on virtual memory page swaps. And newer ones have a continuum of speeds for memory caches and dram and memory backing the ssd and then ssd fused with spinning disk.
On cellphones Apple initially only let one app be active at a time in memory .
And future designs show even more memory types . Many core machines have "near memory" managed by a cpu and far away memory that has to sent over a bus to another cpu
Example, you use a simple java swing text box to input some data. Then a new revision of java comes out and boom the text box gets new capabilies such as auto-fill or spell check.
This exact scenario happened in one particular touch screen voting system in which the windows CE form boxes would remember the previous use of the form and fill it it. Unfortutaley it was filling it in with the previous voter's vote! But it wasn't that the software designer overlooked this. When the software is written it did not do this. But after an update of the Windows CE it did.
Even changing things seeming innocuous like font definition files can introduce unanticipated changes post hoc.
This is true of anything that uses either late binding, or an OS API.
But you would be crazy to not use safe and validated things to be a window manager. Rolling your own would likely introduce even more prospects for security hazards.
and were annealed. It's an annealer. not exactly a quantum computer. But real QM computers have noise problems and waste almost all their bits on error correction methods. I wonder if there is any limit in which annealing and error correction strike some sort of equivalance or are two limit cases of something? I don't understand it well enough but I'm always trying to get more insight into the limits on the D-wave.
First gen is reference code. THen for a while people hack it for speed. Then some saint refactors it to generalize the issues that cause speed problems. And then the refactorization allows a consistent acceleration.
Since all aspects of the cassette tape response function from grain-level magnetic domain saturation, to wow and flutter in the the tape speed, to head alignment all can be numerically modeled. Even degaussing from tape stress over repeated plays and magnetic bleed through from tightly wound thin tapes. If all that gives it some j' Ne Sais Quoi that is sought, Why not just create a time domain filter for digital music and play that? lot cheaper than a cassette. Even a raspberry pi or an amazon dash button has the horsepower to do that kind of filtering. Moreover you don't even need to do it in real time, just preprocess it.
...much of my work for the last 10 years has been a process of cut/paste from my earlier work, or just Goolge a question, follow a link to Stackoverflow, read a few posts for 10 minutes, and only then copy/paste.
Huh, I hate to say it, but it sounds like you really should not have been employed for the past 10 years.
If I wrote code that way I would have for sure been fired LONG ago...
Rubbish. 90% of all coding these days is plumbing libraries. Computers have gotten so fast that glue code now is not inefficient. While at the same time big data wrangling, and numerical optimization and network management have become so complex it's literally dangerous to role your own. Much better to borrow code that has been demonstrated to work well and adapt it.
This is not to say that research algorithmic development, or scaling of research algorithms does require artisanal coding. But THAT is that 10% I mentioned. The rest is Plumbing in modern coding.
I used to worry intensely about how to manage garbage collection or resuse of arrays, which order I ran a loop over a 2D array, and if I should mulitply by 2 or add x+x. These it's much better to code for clarity than optimality and use as many libraries as possible to achieve the efficiency you need
What's the point of software development anyhow? presumably to solve problems not to develop software. Douglas Adams proposed the interface of the future would be a desk you work at trying to solve a problem. The computer would observe what you were doing, then write an algorithm to do it for you. At the time he meant this as a joke. But this is infact exactly the sort of problem that so-called Artifical Intelligence is getting good at. It's getting good at recognixing a start on something then completing it. For example deepFakes fills in a face into a removed face. Adobe's eraser removes defects and fills them back in. And combinatorial material ascience is having success in taking in some basic physics and examples of compounds that exhibit desired properties and then suggesting new molecules that might have similar properties.
AI is really crappy at figuring out what to do. It's really good at observing what you think is important then extrapolating that. Thus Douglas Adams desk interface is no long a joke concept.
How hard would it be to have a computer write a sorting algorithm just by watching someone sort numbers? It's plausible.
Are you sure you are an accountant or do you just play one on Slashdot? Kidding aside, what you just said that makes no sense at face value. It's like saying blockchain might replace type declarations in C-programming or validation of an XML document. That makes no sense right? Doubleentry is just a layout strategy for keeping books and it's a formal declaration process where one record backs another record just like a C-header file is to the C-code block declaration.
SO you have my interest now. Assuming you are serious how does Block chain replace a format specification?
Slashdot Mirror
Artificial Artificial Intelligence. It's the new genuine simulated leather.
been doing that for a long long long time.
https://en.wikipedia.org/wiki/...
You need D-sleep or you go insane and the world is destroyed.
I was going to say something, but I forgot what it was. Baby shark doo doo doo
I just went to the google accelerated page for Reddit on my iphone safari browser. Guess what? It's no longer accessible in safari! google forces you to use chrome to reach it. Google is breaking the internet.
If you get a .dev page then after a year it's going to not respond to https. it's will be only a google accelerated page only and only vailble by Ghttp protocol.
don't fall for it.
I totally get the idea of open access. And it's laudable. It's also a choice. You have approximately three choices when you publish: 1.
publish in an open access journal, and pay it's editorial and compositional page charges,
2. Publish in a pay-for access journal, which may have lower compositional page charges
3. publish in some open and free journal (are there any good ones?)
If you go to a pay-for-access journal and ask them to make all of your author's papers free to access you are basically insane. Sure you can ask but you are asking the journal to go bankrupt or at a minimum work at cost. If you add onto that a request to reduce page charges too, well .... At this point you should just ask for a pony as well.
One could imagine that journals should pay authors for their articles. THat makes some logical sense but it just shifts the cost to the access.
I like the pay-to-publish model myself because if there is barrier it can, if used correctly and not as a vanity press, result in a journal I'm more likely to want to read and more proud to publish in too.
The right argument is if in the age of digital publishing we could not find some less expensive process. But that's not what UC is asking.
But the key thing to keep sight of is that the editorial process should try to stamp out crap. That's the whole reason I'm willing to pay. I can't read everything and if every search term has a load of crap then it's useless.
However that's not hopeless. Google ate alta vista's lunch because it provided more relevant searches. So it is possible to beat down the cost and still beat down crap. However, when it comes to science articles I still prefer peer review to key term search as a way to beat down the crap.
Finally, UC should consider just requiring it's authors to put their articles in a non-copyrighted form on Xarciv before sending them to elsevier. They won't save money but if they genuinely want free access to all UC author pubs it's already available to them. I think it's all about money and not about free access
Once again google know where you go.
Volvo polestar anagrams
Love slot vapor
Volts A Love Pro
Volt Saver Loop
Maybe they will call the sport model the polecat and give it a white rally stripe. Eh Pepe?
I love sky effects. There's lots of extra physics one can see in rainbows that most people never see. As described in the summary, this is a well known effect known as supe numary rainbows. They arise from multipathing in the dorplet as decribed. You see these all the time in thin clouds but you may not have noiced. Next time you see thin clouds look for oscilating purple and teal bands. Them's the supernumary bows.
And next time you see a double rainbow look at it again. First he second rainbow going the reverse direction. This tells you the bow is not just a second rainbow in a different place than the first, but it's different physics. Second look at the gap between the bows. It's not only always the same size, it's also subtly darker than the rest of the sky. What's there is an infrared rainbow. You just can't see it.
And on a cold day with some high very light clouds. Look off to the side of the sun. You might see bright spots in the sky on either side. THis tells you there are elingated hexagonal ice spikes in the air because it's cold up there.
My favorite effect of all that makes my kids really wonder is the effect called "jesus rays" but I'm not going to spoil it for you cause it's cool when the penny drops.
I can't tell you why they were sloppy but I can tell you a few things that might contribute.
first, voting systems have some rules they follow. It varies from state to state but typically you are not allowed to alter any code in these things within a certain number of months of the election. This has led, more than once, to situations where a known bug (e.g. overflows of vote counts, or vulnerabilities) exists in the code but they cannot legally patch it. FOr systems used in federal elections they are/were supposed to get the blessing of the federal election assistance commissions which mandates code quality reviews (these are a total joke as they look at programming styles not logic) and that adds more time. IN fact vendors in two states were caught fixing bugs prior to elections and they knew they were no supposed. The problem is what do you do if the bug is really really awful? That'a why the vendors risked getting caught. One assumes this happens a lot and they don't get caught too.
Second, voting machines don't make a lot of money compared to gaming. So there's a lot of corner cutting. More importantly, it's also a race to market. After states buy a brand of machine they are not going to switch to your brand for a decade. So late to market is death. Better to run with a crappy system. That's also presumably why these things have been built on things like windows CE and Windows Access data bases and commodity OEM computing platforms inside. Quicker to market. Security last.
This you can see even if they did catch the problem they were hosed about getting it fixed promptly.
or Fed Ex?
How do they handle the heat? Many nvnm desktop solid state storage have heat sinks and they are bigger too.
As for memory versus storage this used to be a continuum . Old computers leaned heavily on virtual memory page swaps. And newer ones have a continuum of speeds for memory caches and dram and memory backing the ssd and then ssd fused with spinning disk.
On cellphones Apple initially only let one app be active at a time in memory .
And future designs show even more memory types . Many core machines have "near memory" managed by a cpu and far away memory that has to sent over a bus to another cpu
Example, you use a simple java swing text box to input some data. Then a new revision of java comes out and boom the text box gets new capabilies such as auto-fill or spell check.
This exact scenario happened in one particular touch screen voting system in which the windows CE form boxes would remember the previous use of the form and fill it it. Unfortutaley it was filling it in with the previous voter's vote!
But it wasn't that the software designer overlooked this. When the software is written it did not do this. But after an update of the Windows CE it did.
Even changing things seeming innocuous like font definition files can introduce unanticipated changes post hoc.
This is true of anything that uses either late binding, or an OS API.
But you would be crazy to not use safe and validated things to be a window manager. Rolling your own would likely introduce even more prospects for security hazards.
there isn't an easy answer.
Dilbert needs to explore the quantum boss: I'm observing you, and results don't seem to be improving.
and were annealed. It's an annealer. not exactly a quantum computer. But real QM computers have noise problems and waste almost all their bits on error correction methods. I wonder if there is any limit in which annealing and error correction strike some sort of equivalance or are two limit cases of something? I don't understand it well enough but I'm always trying to get more insight into the limits on the D-wave.
First gen is reference code. THen for a while people hack it for speed. Then some saint refactors it to generalize the issues that cause speed problems. And then the refactorization allows a consistent acceleration.
I agree!
What crappy misleading presentation. they say Even the apple was vulnerable, but oh wait, that was on the unpatched apple code, so nevermind.
Since all aspects of the cassette tape response function from grain-level magnetic domain saturation, to wow and flutter in the the tape speed, to head alignment all can be numerically modeled. Even degaussing from tape stress over repeated plays and magnetic bleed through from tightly wound thin tapes. If all that gives it some j' Ne Sais Quoi that is sought, Why not just create a time domain filter for digital music and play that? lot cheaper than a cassette. Even a raspberry pi or an amazon dash button has the horsepower to do that kind of filtering. Moreover you don't even need to do it in real time, just preprocess it.
...much of my work for the last 10 years has been a process of cut/paste from my earlier work, or just Goolge a question, follow a link to Stackoverflow, read a few posts for 10 minutes, and only then copy/paste.
Huh, I hate to say it, but it sounds like you really should not have been employed for the past 10 years.
If I wrote code that way I would have for sure been fired LONG ago...
Rubbish. 90% of all coding these days is plumbing libraries. Computers have gotten so fast that glue code now is not inefficient. While at the same time big data wrangling, and numerical optimization and network management have become so complex it's literally dangerous to role your own. Much better to borrow code that has been demonstrated to work well and adapt it.
This is not to say that research algorithmic development, or scaling of research algorithms does require artisanal coding. But THAT is that 10% I mentioned. The rest is Plumbing in modern coding.
I used to worry intensely about how to manage garbage collection or resuse of arrays, which order I ran a loop over a 2D array, and if I should mulitply by 2 or add x+x. These it's much better to code for clarity than optimality and use as many libraries as possible to achieve the efficiency you need
Also noted the common failing that you get what you ask for not what you want, which AI has made even more relevant than it was before
What's the point of software development anyhow? presumably to solve problems not to develop software.
Douglas Adams proposed the interface of the future would be a desk you work at trying to solve a problem. The computer would observe what you were doing, then write an algorithm to do it for you. At the time he meant this as a joke. But this is infact exactly the sort of problem that so-called Artifical Intelligence is getting good at. It's getting good at recognixing a start on something then completing it. For example deepFakes fills in a face into a removed face. Adobe's eraser removes defects and fills them back in. And combinatorial material ascience is having success in taking in some basic physics and examples of compounds that exhibit desired properties and then suggesting new molecules that might have similar properties.
AI is really crappy at figuring out what to do. It's really good at observing what you think is important then extrapolating that. Thus Douglas Adams desk interface is no long a joke concept.
How hard would it be to have a computer write a sorting algorithm just by watching someone sort numbers? It's plausible.
Are you sure you are an accountant or do you just play one on Slashdot? Kidding aside, what you just said that makes no sense at face value. It's like saying blockchain might replace type declarations in C-programming or validation of an XML document. That makes no sense right? Doubleentry is just a layout strategy for keeping books and it's a formal declaration process where one record backs another record just like a C-header file is to the C-code block declaration.
SO you have my interest now. Assuming you are serious how does Block chain replace a format specification?
If they suspect Huwaii isn't secure, then it's irrelevant how many years it sets them back.