Its oky YRO applies to anything related to rights, even if it isn't online. It isn't much of a stretch from there to apply it to anything related to online or you.
When I was still at Bearingpoint (before they went under), I worked on a project that was at least 400 million in the hole. And the client still wanted to keep changing requirements!
You are wrong. I am not going to say that I know much about optimization and performance of Python using c libraries, I don't. I do know that interpreted python is many many times slower than java. I do however know a lot about high performance java. Most of the time, it is comparable to writing in c/c++, a few cases run faster, most run slightly slower. One specific instance where java dramatically outperforms c is in allocation of memory on the heap, java does this in ~60 instructions vs c++ doing it in ~300.
The jvm does execute as native code, and the hotspot jvm can make optimizations at run time based on the current environment that will simply be unavailable to a c compiler.
Being written in c doesn't make it inherently faster than java. I've seen studies that have shown java running between 15% faster than c and less than 100% slower than c for specific algorithms.
It could go either way. Someone would have to do an actual comparison to see how these libraries stack up.
The argument against disclosure of personally identifiable voting records is that disclosing the vote record would allow a party to verify that a paid shill voted the way they were asked to.
Seriously, the performance gap between a hotspot optimized java implementation and python is generally a difference of about 20 fold for crunching numbers. For python to equal java in performance, the startup time of the vm has to represent the bulk of the work to be done. And then performance does not really matter that much because the size of your dataset is small enough it won't matter much.
I have a friend that "rolls his own" eCigs. He says the trick to make it more satisfying is to up the voltage to about 4.5v to get a hotter vapor and to use a variety of high quality flavors (red bull is one of his preferred ones). IIRC it can be done for less than $10 of electronics.
The number of countries with freedom of speech and the extent of that freedom in many countries is reducing every year. I fear that the concepts of freedom, liberty and democracy were nothing more than a passing fad and will be out of style within my lifetime.
As much as I believe software/math patents to be irredeemably evil and hope regularly that they are abolished, that may be a good idea unless the company you work for has a "We own all your patents" clause.
Opera isn't the only browser that needs extra care, its just that it needs a retarded amount of extra care.
Example: Take a select, input, button and textarea. Set them to have the same width, border, padding and margin (non-zero). Look at it in every browser. Do they appear the same width? NO! they do not. The margin of error is different between gecko, webkit, opera and ie.
java really really only has problems with startup time (that a web spider will never see) and the delay when a servlet|jsp is hit the first time. While doing web development, we see that startup and first load most of the time, giving an appearance of slowness, but it is much better on a production server with regular traffic.
How is this supposed to clear the wrongly convicted?
If you are wrongly convicted, you wont have much issue providing your own DNA to get free.
This has only a few applications:
With current technology a matching genetic pattern can be generated. This would make a great tool for acquiring genetic patterns for the fabrication of evidence and false convictions. Even without such fabrication, genetic evidence can be abused to implicate someone that just happened to have passed through the location of a crime days, weeks or months before the event.
Even so, the complexity of handling the huge number of exceptions and expected behavior for broken code is a vastly larger problem than validating and rendering xhtml.
actually it was. Specifically its handling of the alt attribute was bad.
There already existed a standard way for dealing with alternate display of tags that are not understood by the browser, that was to simply ignore the tag and render whatever was inside. By not following that paradigm, it made graceful multiple-level degradation all but impossible.
It is the same reason that the button element is better than input type=button|submit|etc.
That does not make any sense. XHTML is far easier to render than a random assortment of tag soup with no clear expectation of how it is supposed to actually render.
I would hardly call html5 an actual working implementation. Hell, html4 isn't perfectly implemented in a good chunk of browsers yet.
XHTML2 was an extensible format by definition. There was nothing there stopping you or anyone else from adding a video or canvas tag to it.
A good specification needs to be consistent, it needs to be logical and well formed, it needs to be minimal but specific, it needs to address known problems with the previous specification.
1: errors were treated like errors. That means that broken hacks made by graphic artists would result in an error message instead of a random attempt to render a broken document. This also made creating a partial implementation more difficult.
2: No one implemented a reference implementation. So that web browser vendors would have to do all the heavy lifting.
WHATWG formed and decided to take all the hacked errors and random implementations of browsers and make those errors the standard, then they added some cruft on top. Thus HTML5 was born. For some reason, W3C then abandoned the superior standard of XHTML2 and adopted the steaming pile that WHATWG dumped on them.
Slashdot Mirror
Its oky YRO applies to anything related to rights, even if it isn't online. It isn't much of a stretch from there to apply it to anything related to online or you.
mostly because if someone is going to pay you to sign a petition, they can do it right there when they get you to sign.
When I was still at Bearingpoint (before they went under), I worked on a project that was at least 400 million in the hole. And the client still wanted to keep changing requirements!
tell me about it. The only thing worse than a zombie is a zombie with stealth and combat training.
You are wrong. I am not going to say that I know much about optimization and performance of Python using c libraries, I don't. I do know that interpreted python is many many times slower than java. I do however know a lot about high performance java. Most of the time, it is comparable to writing in c/c++, a few cases run faster, most run slightly slower. One specific instance where java dramatically outperforms c is in allocation of memory on the heap, java does this in ~60 instructions vs c++ doing it in ~300.
The jvm does execute as native code, and the hotspot jvm can make optimizations at run time based on the current environment that will simply be unavailable to a c compiler.
Damn strait, we should be giving those malaria victims prosthetic feet instead!
Being written in c doesn't make it inherently faster than java. I've seen studies that have shown java running between 15% faster than c and less than 100% slower than c for specific algorithms.
It could go either way. Someone would have to do an actual comparison to see how these libraries stack up.
The argument against disclosure of personally identifiable voting records is that disclosing the vote record would allow a party to verify that a paid shill voted the way they were asked to.
Seriously, the performance gap between a hotspot optimized java implementation and python is generally a difference of about 20 fold for crunching numbers. For python to equal java in performance, the startup time of the vm has to represent the bulk of the work to be done. And then performance does not really matter that much because the size of your dataset is small enough it won't matter much.
I have a friend that "rolls his own" eCigs. He says the trick to make it more satisfying is to up the voltage to about 4.5v to get a hotter vapor and to use a variety of high quality flavors (red bull is one of his preferred ones). IIRC it can be done for less than $10 of electronics.
The number of countries with freedom of speech and the extent of that freedom in many countries is reducing every year. I fear that the concepts of freedom, liberty and democracy were nothing more than a passing fad and will be out of style within my lifetime.
As much as I believe software/math patents to be irredeemably evil and hope regularly that they are abolished, that may be a good idea unless the company you work for has a "We own all your patents" clause.
Opera isn't the only browser that needs extra care, its just that it needs a retarded amount of extra care.
Example: Take a select, input, button and textarea. Set them to have the same width, border, padding and margin (non-zero). Look at it in every browser. Do they appear the same width? NO! they do not. The margin of error is different between gecko, webkit, opera and ie.
I believe that copyright is more than sufficient IP protection for software.
Eliminate software patents completely.
If you don't do that, you should give up copyright protection and submit working source code to the patent office.
java really really only has problems with startup time (that a web spider will never see) and the delay when a servlet|jsp is hit the first time. While doing web development, we see that startup and first load most of the time, giving an appearance of slowness, but it is much better on a production server with regular traffic.
And how does that do anything other than increase your chances of getting convicted (both wrongfully and rightfully)?
How is this supposed to clear the wrongly convicted?
If you are wrongly convicted, you wont have much issue providing your own DNA to get free.
This has only a few applications:
With current technology a matching genetic pattern can be generated. This would make a great tool for acquiring genetic patterns for the fabrication of evidence and false convictions.
Even without such fabrication, genetic evidence can be abused to implicate someone that just happened to have passed through the location of a crime days, weeks or months before the event.
No it is not. I've spotted at least 5 of those items that are outright wrong.
Java is generally better than c#, but you don't need to make shit up to show that.
Even so, the complexity of handling the huge number of exceptions and expected behavior for broken code is a vastly larger problem than validating and rendering xhtml.
actually it was. Specifically its handling of the alt attribute was bad.
There already existed a standard way for dealing with alternate display of tags that are not understood by the browser, that was to simply ignore the tag and render whatever was inside. By not following that paradigm, it made graceful multiple-level degradation all but impossible.
It is the same reason that the button element is better than input type=button|submit|etc.
It would not be so bad if they handled local storage a lot like cookies except that they don't get sent back to the server with every request.
That does not make any sense. XHTML is far easier to render than a random assortment of tag soup with no clear expectation of how it is supposed to actually render.
I would hardly call html5 an actual working implementation. Hell, html4 isn't perfectly implemented in a good chunk of browsers yet.
XHTML2 was an extensible format by definition. There was nothing there stopping you or anyone else from adding a video or canvas tag to it.
A good specification needs to be consistent, it needs to be logical and well formed, it needs to be minimal but specific, it needs to address known problems with the previous specification.
HTML5 is and does none of these.
He is probably a disgruntled XHTML2 supporter, just like me.
What happened was that xhtml2 had two flaws.
1: errors were treated like errors. That means that broken hacks made by graphic artists would result in an error message instead of a random attempt to render a broken document. This also made creating a partial implementation more difficult.
2: No one implemented a reference implementation. So that web browser vendors would have to do all the heavy lifting.
WHATWG formed and decided to take all the hacked errors and random implementations of browsers and make those errors the standard, then they added some cruft on top. Thus HTML5 was born. For some reason, W3C then abandoned the superior standard of XHTML2 and adopted the steaming pile that WHATWG dumped on them.