HTML5 requires a more complex parser than XHTML ever will.
I guess that parsing time is a fraction of the whole time required to render the page: text rendering, displaying images and running JS.
As I have never wrote a HTML parser, I can only draw analogies with other fields. E.g. in source code compilation most of the time is spent in optimizer, not parser. Even when optimizer is off, the times are still dominated by (1) the reading of all the source files (in C - headers) from disk and (2) semantics validation and code generation.
This is an event in a way. For ages security folks have told Apple to do something about the vector of attack - at least somehow protect against malware which pretends to be Apple's own app (what is easy: just copy-paste an icon and no-one would be able to tell the difference). There were rumors about trojans on Mac OS for ages. Some of pirated Mac software on P2P networks is said to be infected. As that was probably the source before, Apple might have been reluctant to act.
Apple acknowledging existence of malware and actively doing something about it is an event.
MIX didn't have any VM. Not that VM is so much relevant to general CS. And it's not even about general CS: it is about computer and OS architecture. Even the RTFA's b-heap structure can be viewed as heap of heaps: main heap made of pages and sub-heaps inside of the every page. So much for novelty...
Not to mention the author's frame of reference is totally invalid, because in the real world, we make sure we have so much RAM in our servers that virtual memory never becomes that large of a factor.
You are wrong - about where he's wrong. The author essentially takes a very niche application, optimizes it and then goes to wildly extrapolate his finding on CS in general. That's where he's wrong.
Lots of server applications (RDBMSs, HTTPDs) would love to get better use of VM and load all the junk once and serve it as if from memory. The problem is that in RTFA's niche application (very dumb cache) one doesn't have to take care for things like coherency or consistency, something most other applications are bound to provide out of box and what contributes lion share of their complexity. And yes, instead of increasing the applications complexity, we'd rather throw some buck at RAM and keep it simple instead.
If a set of closely related nodes can fit in a cache line, this improves locality so that the CPU can do more work instead of twiddling its thumbs.
That's sort of... obvious? But off-topic.
RTFA: it revolves around I/O optimization. It was tested on system with 16GB RAM and 300GB mmaped storage. LOL. Essentially author claims that if you use binary tree for on-disk data look-ups it would be *surprise* very slow. Then he goes on to reinvent some b-trees, tuned for the task at hand, and shows how they are 10 times faster.
An artist or a business creates something and offers it for sale.
Selling reselling the same thing over and over again? Without any further investments?? And expect to be paid the same money forever???
It's rather stupid of anybody to expect that to work in long term.
Concert, shows, etc. That's how artists should earn their living. And in fact many do precisely that as very few can afford proper promotion for anything to sell.
Pirating commercially sold entertainment is not.
Entertainment is a service.
Service (or rather "a copy of service") can't be sold as a bottled water.
E.g. you can't rip a visit to concert or show. You either was there and seen it - or not.
It can't be WoW since the study found that consoles dominate among such players.
I usually make sarcastic remark that consoles would always win PCs in time spent playing metric, because in a usual console game one spends disproportional amount of time watching cut-scenes and flashy move animations or simply getting from A to B.
If you put in an order for a large number of identical machines chances are they'll have components from the same batch.
I'd say it depends.
Our supplier of storage solution "randomizes" the batches. IT was in the beginning checking the HDDs too, only to find that most deliveries contain drives from several batches.
Other company I have worked for got Dell servers and also RAID10. Few months later on Thursday one drive died. Dell provided new drive on Friday and recovery was started - only to find that another original drive failed over the weekend too, rendering the storage dead. All drives in the original RAID10 were from the same batch.
Some companies do get it. Some do not.
P.S. Having *all* drives from different batches, as was explained to me by data recovery specialists, is also bad if one later would want to try to recover information from the dead drive's platters: different batches might have different controllers with different configurations making them irreplaceable. If you have two drives from the same batch and one of them is dead - recovery would be relatively fast and cheap. Recovering information directly from platters is magnitude(s) more expensive.
How can anyone reasonably know what is and isn't copyrighted [...]
Every work by default is copyrighted.
The question here is that lots of content is passed around without a proper licenses, so technically one cannot really know under which terms they receive the work and what they are allowed to do with it.
It's kind of Internet age thing. Before, public domain and free content was scarce at best, so there was a little of legal confusion around it. Now it is quite commonplace and one may not expect that all people can tell the difference.
'I'm all for sharing, but I recognize the truly great things may not come from that environment.'
So let's put all the creators into a vacuum and see what they will come up with.
Probably he also should suggest closing schools, colleges, universities and public libraries: they are breeding grounds for knowledge and ideas sharing.
I would admit that long time ago I had very similar opinion. But then I got job in educational software company and worked with number of talented teachers and educators who have have enlightened me. Now it simply boggles my mind that somebody can even think about restraining sharing as it is the very foundation of our society and the backbone of our civilization.
it's entirely possible that battery life could be improved by recompiling the kernel with different flags or some equally esoteric maneuver.
Or probably not.
I did Ubuntu 8.04 tuning for low power consumption. My experience might be outdated, but I doubt it. Best I could do was to remove the Ubuntu, install like Xubuntu, then remove some more crap.
The problem is that the most power consuming device is the hard drive - but few of Linux desktop applications really care what/when/how they write to the disk. That leads to the effect that hard drive pretty much never sleeps.
After removing the offending applications, one finds that Linux also reads from disk speculatively nor applications have option to prefetch data from disk. E.g. reading a large PDF under Mac OS X, HDD remains silent all the time - under Linux HDD starts on every other page. And even scrolling thorough the document doesn't always help: to cache the data under Linux one has to really PgDown through every single page of the PDF. Likewise none of the Linux audio players has an option to prefetch the MP3s/etc into memory.
Kernel option are nice and all. But unless applications are implemented with power saving in mind, it's all for naught.
.... and in the times when I was trying to pilot Chrome + FlashBlock, the FlashBlock was appearing/disappearing on me sporadically. One week it is there and works - another week Flash is all over the pages.
I'm not sure about it, but I think that I have experienced the other side of the all-new "silent" browser auto-updates....
Chrome is a nice upgrade from IE. The problem to me that I never was an IE user. I got attached to the Netscape^W MozillaW FireFox's bells and whistles more than decade ago.
The Web without reliable AdBlock/FlashBlock/NoScript is a rather spooky place. Nor I have problem with Fx start-up: the five+ seconds wait is well worth it. Every second of it.
Having one of the Pirate parties directly associated in this way already reinforces perceived connections between the Pirate parties and outright software piracy to an extent that really isn't helpful.
And who said that piracy is bad?
I hear lots of businesses moaning and whining about it, but they would moan and whine about *any* problem. I do not see how it is different from say competition or financial crisis or natural disasters. Or the actually happening shift in the business model.
Avatar is the most pirated movie of all time - and it didn't prevent it from netting 1+Bln. Or could it be that the piracy actually helped? After seeing crappy copy off P2P many wanted to see it in all the 3D glory??
too often the expensive proprietary version is just that much better than the free version
With notable exception of M$Office 2003/earlier and CADs, this statement relates to the reality very loosely.
This is a fairly common problem with FOSS, and it's one of the downsides of the FOSS ideology- many FOSS projects often have great developers but tend to miss other things that proprietary vendors do not- good UI designers as well as investment into usability studies, good QA, etc.
WTF?! I use corpoware on the daily basis and what you try to advertise here is applicable optimistically to 5-10% of the said software. And the same share of FOSS is well polished and nice/easy to use.
A lot of FOSS software is developed for FOSS developers, anyone else be damned.
FOSS model is "egoistic development model" - everybody develops for himself. And many corporation also "get it" and assign developers to FOSS projects to make the adjustments - either locally or in mainline - to accommodate their business cases. What is pretty much the same as assignment of specialists to customize proprietary systems and maintain the customizations.
From a business perspective, there's often no point going free if you need more or higher paid specialists to look after said system, whilst the people who use the system are less productive.
This is the most stupid thing I have read in months.
I yet to see the aforementioned "productivity" anywhere else but marketing PowerPoint slides.
Business goes for proprietary software due to long term support contracts. And that's about 75% of reasons. The remaining 25% of reasons revolve around backward compatibility.
And assigning a specialist to "look after said system" is the same for proprietary software. With the notable difference that assigning a specialist to babysit a FOSS deployment might also result in the problems being fixed eventually - while with proprietary software that happens like... never. (Needless to mention that licensing costs often eclipse the IT wages: often it is cheaper to hire extra IT guy than to buy another proprietary corpoware.)
I could have called our IT for the examples, but I think it is redundant. The myth that proprietary software is somehow magically better for users is just that - myth. And was debunked many many times before.
Slashdot Mirror
HTML5 requires a more complex parser than XHTML ever will.
I guess that parsing time is a fraction of the whole time required to render the page: text rendering, displaying images and running JS.
As I have never wrote a HTML parser, I can only draw analogies with other fields. E.g. in source code compilation most of the time is spent in optimizer, not parser. Even when optimizer is off, the times are still dominated by (1) the reading of all the source files (in C - headers) from disk and (2) semantics validation and code generation.
Not so fast.
Big studios and record labels spoil artists with huge rewards. Or at least a promise of the huge reward. Breaking the bad habit might take long time.
I'm all for TV and movies learning to slim down their budgets and becoming more independent.
I wish to say "YES" too.
You want to destroy a country? Make it a democracy. A democracy is two wolves and a sheep voting on what's for dinner.
Democracy is the worst form of government, except for all those other forms that have been tried from time to time.
-- Sir Winston Leonard Spencer-Churchill
This is an event in a way. For ages security folks have told Apple to do something about the vector of attack - at least somehow protect against malware which pretends to be Apple's own app (what is easy: just copy-paste an icon and no-one would be able to tell the difference). There were rumors about trojans on Mac OS for ages. Some of pirated Mac software on P2P networks is said to be infected. As that was probably the source before, Apple might have been reluctant to act.
Apple acknowledging existence of malware and actively doing something about it is an event.
MIX didn't have any VM. Not that VM is so much relevant to general CS. And it's not even about general CS: it is about computer and OS architecture. Even the RTFA's b-heap structure can be viewed as heap of heaps: main heap made of pages and sub-heaps inside of the every page. So much for novelty...
Not to mention the author's frame of reference is totally invalid, because in the real world, we make sure we have so much RAM in our servers that virtual memory never becomes that large of a factor.
You are wrong - about where he's wrong. The author essentially takes a very niche application, optimizes it and then goes to wildly extrapolate his finding on CS in general. That's where he's wrong.
Lots of server applications (RDBMSs, HTTPDs) would love to get better use of VM and load all the junk once and serve it as if from memory. The problem is that in RTFA's niche application (very dumb cache) one doesn't have to take care for things like coherency or consistency, something most other applications are bound to provide out of box and what contributes lion share of their complexity. And yes, instead of increasing the applications complexity, we'd rather throw some buck at RAM and keep it simple instead.
If a set of closely related nodes can fit in a cache line, this improves locality so that the CPU can do more work instead of twiddling its thumbs.
That's sort of ... obvious? But off-topic.
RTFA: it revolves around I/O optimization. It was tested on system with 16GB RAM and 300GB mmaped storage. LOL. Essentially author claims that if you use binary tree for on-disk data look-ups it would be *surprise* very slow. Then he goes on to reinvent some b-trees, tuned for the task at hand, and shows how they are 10 times faster.
Was my first thought too: that's what you do not even need to do in Aperture, cause it does all that automatically for you.
Sadly it's Mac-only. And over-priced - if tagging is only what you want.
An artist or a business creates something and offers it for sale.
Selling reselling the same thing over and over again? Without any further investments?? And expect to be paid the same money forever???
It's rather stupid of anybody to expect that to work in long term.
Concert, shows, etc. That's how artists should earn their living. And in fact many do precisely that as very few can afford proper promotion for anything to sell.
Pirating commercially sold entertainment is not.
Entertainment is a service.
Service (or rather "a copy of service") can't be sold as a bottled water.
E.g. you can't rip a visit to concert or show. You either was there and seen it - or not.
i still think culture is culture and business is business, and we have to figure out where the border is, what the limits are.
Before I come to know US/etc laws, it was plain common sense to me: business is where money exchange is involved.
Private sharing -> no money involved -> not a business -> normal cultural information exchange.
Judging by the trailer I think the particular movie should be great.
It can't be WoW since the study found that consoles dominate among such players.
I usually make sarcastic remark that consoles would always win PCs in time spent playing metric, because in a usual console game one spends disproportional amount of time watching cut-scenes and flashy move animations or simply getting from A to B.
Apparently a ton of people disagree with you.
Yeah, the whole *4*% of all gamers.
IMO that would improve drastically quality of games and the game community.
That resonates with my own reading of the quote: all companies who are on the receiving end of M$' security investments praise the investor.
And obviously anti-virus companies would tell that Windows is better: without the swiss cheese OS they would be out of job.
If you put in an order for a large number of identical machines chances are they'll have components from the same batch.
I'd say it depends.
Our supplier of storage solution "randomizes" the batches. IT was in the beginning checking the HDDs too, only to find that most deliveries contain drives from several batches.
Other company I have worked for got Dell servers and also RAID10. Few months later on Thursday one drive died. Dell provided new drive on Friday and recovery was started - only to find that another original drive failed over the weekend too, rendering the storage dead. All drives in the original RAID10 were from the same batch.
Some companies do get it. Some do not.
P.S. Having *all* drives from different batches, as was explained to me by data recovery specialists, is also bad if one later would want to try to recover information from the dead drive's platters: different batches might have different controllers with different configurations making them irreplaceable. If you have two drives from the same batch and one of them is dead - recovery would be relatively fast and cheap. Recovering information directly from platters is magnitude(s) more expensive.
True. Again mixed it up with authorship...
How can anyone reasonably know what is and isn't copyrighted [...]
Every work by default is copyrighted.
The question here is that lots of content is passed around without a proper licenses, so technically one cannot really know under which terms they receive the work and what they are allowed to do with it.
It's kind of Internet age thing. Before, public domain and free content was scarce at best, so there was a little of legal confusion around it. Now it is quite commonplace and one may not expect that all people can tell the difference.
'I'm all for sharing, but I recognize the truly great things may not come from that environment.'
So let's put all the creators into a vacuum and see what they will come up with.
Probably he also should suggest closing schools, colleges, universities and public libraries: they are breeding grounds for knowledge and ideas sharing.
I would admit that long time ago I had very similar opinion. But then I got job in educational software company and worked with number of talented teachers and educators who have have enlightened me. Now it simply boggles my mind that somebody can even think about restraining sharing as it is the very foundation of our society and the backbone of our civilization.
Programming Python is much much less nightmare-ish than VB.
That's me Perl hardcore programmer saying.
it's entirely possible that battery life could be improved by recompiling the kernel with different flags or some equally esoteric maneuver.
Or probably not.
I did Ubuntu 8.04 tuning for low power consumption. My experience might be outdated, but I doubt it. Best I could do was to remove the Ubuntu, install like Xubuntu, then remove some more crap.
The problem is that the most power consuming device is the hard drive - but few of Linux desktop applications really care what/when/how they write to the disk. That leads to the effect that hard drive pretty much never sleeps.
After removing the offending applications, one finds that Linux also reads from disk speculatively nor applications have option to prefetch data from disk. E.g. reading a large PDF under Mac OS X, HDD remains silent all the time - under Linux HDD starts on every other page. And even scrolling thorough the document doesn't always help: to cache the data under Linux one has to really PgDown through every single page of the PDF. Likewise none of the Linux audio players has an option to prefetch the MP3s/etc into memory.
Kernel option are nice and all. But unless applications are implemented with power saving in mind, it's all for naught.
I'm not sure about it, but I think that I have experienced the other side of the all-new "silent" browser auto-updates....
Chrome is a nice upgrade from IE. The problem to me that I never was an IE user. I got attached to the Netscape^W MozillaW FireFox's bells and whistles more than decade ago.
The Web without reliable AdBlock/FlashBlock/NoScript is a rather spooky place. Nor I have problem with Fx start-up: the five+ seconds wait is well worth it. Every second of it.
Having one of the Pirate parties directly associated in this way already reinforces perceived connections between the Pirate parties and outright software piracy to an extent that really isn't helpful.
And who said that piracy is bad?
I hear lots of businesses moaning and whining about it, but they would moan and whine about *any* problem. I do not see how it is different from say competition or financial crisis or natural disasters. Or the actually happening shift in the business model.
Avatar is the most pirated movie of all time - and it didn't prevent it from netting 1+Bln. Or could it be that the piracy actually helped? After seeing crappy copy off P2P many wanted to see it in all the 3D glory??
too often the expensive proprietary version is just that much better than the free version
With notable exception of M$Office 2003/earlier and CADs, this statement relates to the reality very loosely.
This is a fairly common problem with FOSS, and it's one of the downsides of the FOSS ideology- many FOSS projects often have great developers but tend to miss other things that proprietary vendors do not- good UI designers as well as investment into usability studies, good QA, etc.
WTF?! I use corpoware on the daily basis and what you try to advertise here is applicable optimistically to 5-10% of the said software. And the same share of FOSS is well polished and nice/easy to use.
A lot of FOSS software is developed for FOSS developers, anyone else be damned.
FOSS model is "egoistic development model" - everybody develops for himself. And many corporation also "get it" and assign developers to FOSS projects to make the adjustments - either locally or in mainline - to accommodate their business cases. What is pretty much the same as assignment of specialists to customize proprietary systems and maintain the customizations.
From a business perspective, there's often no point going free if you need more or higher paid specialists to look after said system, whilst the people who use the system are less productive.
This is the most stupid thing I have read in months.
I yet to see the aforementioned "productivity" anywhere else but marketing PowerPoint slides.
Business goes for proprietary software due to long term support contracts. And that's about 75% of reasons. The remaining 25% of reasons revolve around backward compatibility.
And assigning a specialist to "look after said system" is the same for proprietary software. With the notable difference that assigning a specialist to babysit a FOSS deployment might also result in the problems being fixed eventually - while with proprietary software that happens like ... never. (Needless to mention that licensing costs often eclipse the IT wages: often it is cheaper to hire extra IT guy than to buy another proprietary corpoware.)
I could have called our IT for the examples, but I think it is redundant. The myth that proprietary software is somehow magically better for users is just that - myth. And was debunked many many times before.