The article seems to bemoan the lack of standards and APIs for relatively new technologies like blob storage, message queuing and the like.
In short, it's the same situtation everyone deals with when choosing a platform. Take full advantage of the specific platform and lose portablity, or code to a portable subset and tradeoff ease of implementation for platform independence.
You can minimize your dependency on a third-party API if you use an API of your own to code against. Creating a set of interfaces that provide the blob, queuing and other cloud features for your project not only isolates the cloud-specifc code to one place, but it makes it more testable (using test-doubles, fakes, etc).
This is a problem, but it is a well understood one with a time tested solution.
Thank you for this. There was so much commentary that was off the point it made my head spin.
What is so hard about understanding that people do work that makes them feel not so great yet have to do it anyway? And they may have opinions about what they do?
There was so much commentary with the undertone that these women should feel lucky to just stand there and look good to get paid, that their genetics is a ticket to lifelong success.
In the end, it is a job. It requires skills and discipline. Looking like that requires knowing how to dress, to stand and pose, makeup, how to communicate with lots of different people and to be (or appear) engaged. Hell, how to stand in place for long periods doing nothing but not being able to leave.
You have to take care of yourself, you have to make tough choices, and people will resent or vilify you for tapping into the existing market for attractive people.
If you make the choice, you often have to stick with it, because there are few options for you once you are pegged at "just a pretty face".
Oh, I should note that modeling is for men and women. Frankly, the idea of having to watch every calorie and workout 10-15 hours a week just for a hope at a photo shoot makes me happy that I am a programmer.
Look, certainly there are tons of people that are shallow and self-obsessed in modeling, but those people are everywhere.
As for knowing what you are getting into, well the job is promote a product. Imagine how fun it would be trying to impart information to the forty fifth time to some guy that is just staring at your breasts and steadfastly refusing to acknowledge you as a human being. Doesn't sound like fun?
Actually, there is good evidence that CBT treats depression as well as anxiety disorders. The British Department of Health approved CBT as a first line treatment for depression based on a review of available evidence, and is current expanding access to trained therapists.
The problem with many studies is that they are short-term and the placebo effect in the short term is large. The mere fact that somebody is "trying do something" by being in a trial does have a noticeable impact. What is more unclear is if the placebo effect really lasts. Some studies show that CBT and well some medications are indeed better in the long term in terms of overall mood and reduction in depressive episodes.
For me, it is sad that people often are forced into a SSRI because of costs when CBT or other focused therapy could bring about more long lasting changes and better outcomes without the risks of medications.
The underlying concept isn't as hard as the review makes it sound. Using some formalisms, the author creates a set of blocks upon which you can build patterns like those in the Gang of Four book. By doing so, you can see what commonalities occur and hopefully, this leads to deeper insight into what patterns really are.
The review mentions some detection of patterns in "legacy code" but the idea of refactoring the code to better support those concepts that is mentioned in the review seems too vague to me.
There was a good book on how you could use refactoring to implement design patterns in existing code, but it wasn't as grounded on theory and automation as refactoring itself. The theory mentioned is this work could maybe serve as a foundation for such an idea.
If this sparks a new set of tools that help automate the redesign of code to follow better practices, that'd be really great. But, I think there is still some real work left to be done before that happens. And while it remains to be seen, the formalism applied could reap great benefits in that regard.
No increase in demand for teacher and professors. For those of crazy and qualified enough to teach (because we are passionate about it), we can't. I begin to see why tenure isn't necessarily a great thing.
Looking at the discussion here, I can't help but notice something missing. The reason the article was racist is because it implies that blacks by nature have given traits.
The body of genomics provides no support for this. The genetic differences between races are so small that it many biologists and anthropologists claim that the term race has no scientific meaning when applied to humans as there is no statistically significant genetic variation between race populations, esp. compared to individual differences.
What can't get lost here is whatever is happening to groups in terms of violence, crime, etc, it is a cause of the environment. It is not innate and it is not inevitable. It has nothing to do with genetic makeup, but with the circumstances in which people live and are raised in.
This notion that "whites", faced with the same level of discrimination, lack of opportunity, and similar social factors would innately respond better to those challenges is so ridiculous that it beggars belief that people still place any credence in such arguments.
Seconded. No trolling here. They want reasonable payments for creating clever algorithms and techniques for dealing with interference. Sure, it was based on existing technologies, but choosing what technologies to use and combining them correctly and effectively was a difficult and unique proposition, and I think a patentable idea.
They even shopped the technology to actual companies to make products based on it, no go.
Also, it seems that their is a claim that the IEEE standards group was aware of the patent and used the technology. Happens all the time.
So, they finally to court to get royalties for use of the patented technology. Seems to me they exhausted other avenues. Companies were dismissive of their idea as "obvious". Well, in hindsight, why wouldn't it be? So, they went to court.
Good for them. Sure, they venue shopped, but they accepted a pretty reasonable settlement from a large group of companies that greatly profited from the invention.
Code to standards (with appropriate polyfills) and ye shall prosper.
Yea, that'd be fine if there wasn't a ton of unspecified or "up-to-vendor" behavior in the HTML, CSS and DOM standards as they stand now. The reality is that any moderately complex JavaScript page has to be tested against all the browsers. The advantage of things like jQuery, etc. is they do a lot of that and hide the inconsistencies from you.
In this case, it seems tinyMCE tickled odd behavior in Gecko's window.onload event. But, as with a lot of DOM events, figuring out what should fire and in what order is a heck of a challenge, hence things like $(function() {}) in JQuery.
HTML5 may place this on solid ground, but right now HTML5 is a work in progress. In my mind, HTML5 will still have cross-browser issues, because it is really hard to get multiple vendors to truly implement a complex standard correctly and consistently. I worked on the UML2.0 standards committee, and that's exactly what happened.
Hopefully, the HTML5 test suite will be a massive regression suite, on the order of what is used for compilers that touches everything and these inconsistences will become much less common.
Firstly, he can be sentenced up to ten years. Sentencing hasn't occurred yet. In fact, the article notes the time spent will likely be less because of the nature of his background and lack of criminal history.
Also, he was offered a plea deal that included no prison time. By, rejecting this deal, he decided to take his case to the jury and accepted the chance of a harsher sentence if found guilty on the charges.
As for the motivated by bias factor that made him guilty of a hate crime, certainly, these laws are controversial and this case may lead to their re-examination.
But, it is the law of the state he was in, he was found guilty of violating it. If the jury thought he violated the law, then good for them for putting aside their personal objections to it and doing what is required of them.
If you don't like these kinds of laws, you lobby to change it. Via the courts or legislation. Maybe this case will be a basis for challenging the law in this state, for example.
All in all, this seems very simple. Don't spy on people. Don't violate their privacy. There are consequences for such actions, and those may be legal in nature.
Not every problem leads itself to GPGPU solutions. I'm no expert, but looking at a paper on a similar idea for optical reconstruction, I'd bet that the creation and update of the model via analysis (comparsion between predicted and actual results) is really hard to make parallel, and that process has a lot of non-localized memory access.
I'm sure you could use GPU acceleration for the CG calculations in the reconstruction phases, but I'm not sure that's the limiting factor here.
In short, there are really good reasons to use a lot of fast CPUs, and this sounds like one of them.
Seriously, I credit a lot of my success in CS with being taught Scheme. There are free environments that are great for teaching, and you can start with a very reasonable subset of the full language and still do some fun stuff. And the fact that it isn't widely used is a big bonus in my opinion. Languages change, library change, but the fundamentals slowly change. Scheme makes it clear that programming isn't just about a language, it's about abstraction.
Not saying that you give high schoolers a copy of SICP of course, but the fact that it is a stepping stone to that is really kind of neat in my opinion.
Let me make it clear, I find Javascript's quirks and behaviors to be very frustrating at times, so I am biased against it. And, really, there are things that need to be fixed even if that means making a updated version that breaks some older scripts, so be it. Add a attribute to script that says "Hey, use the less-brain damaged version of JS" and be done with it.
As we know, char* means a string. We aren't passing a reference to a single character. The overloading of pointers for arrays is a complicating factor. You are correct in that C does make you decide passing semantics, but it can't truly enforce them. I can happy cast away const and destroy the source string if I want.
Compare to a C# version:
string StrCopy(string src, out string dest);
Here, src is passed by value and we can't get around that, but by using the out keyword, we can get pass by reference semantics on the value. With this and auto-boxing support, I don't don't think you need pointer syntax. C++ added the concept of a reference explicitly to also address the issues of pointer semantic overloading in C. So, pointers aren't the final solution. Not a bad thing at all (when you need them, you need them), but there are other solutions to the problems you raised.
What frustrates me about web security or security in general is alluded in the review; that there is really not a good idea of what security is.
More specifically, the idea of security as a binary property. For me, it seems a more realistic approach is how much information and resource access do I have to gain to perform an action, what are the paths to do so, and how likely are those paths to occur, and what is the cost of the breach. The web makes this analysis harder without question, but still possible.
For example, I have a web site that contains account ids in a post to change something for that account. If the account id is an email, then forging the request is trivial. If the id is an opaque token, but sent in the clear (HTTP), it is less trivial, but still relatively easy. If the id is sent via https POST, sniffing is harder, but replay attacks may occur.
And so on. The point is to decide when the advantage gained is overwhelmed by the cost and risk of the attack.
I remember a time when somebody was worried about GUIDs in a URL because you could guess a valid one. Of course, it is much easier (understatement) to capture a valid one than guess one. This book will have value if it helps you avoid red herrings like that and focus on the real threats.
Of course, I am not a security specialist, so this may be naive at best.
Nothing wrong with a good diet, and proper D3 levels are already tested by a lot of neurologists and supplements are recommended to maintain them in MS suffers.
But, of course, the authors claims that good food is all you need. It all falls down when I saw the recommendation not to take interferons, which, unlike this diet, has a huge amount of research that shows that it reduces relapses for those with relapsing-remitting MS, and lengthens the periods between replases.
Given how little is understood about MS, it is greatly specious to claim that MS can be cured by a diet. And to assume that the doctors behind this work don't have their own biases and agendas just because they don't toe the "party line" in research seems a bit naive.
Finally, given that MS has periods of remission, how can you claim that MS is reversed vs being in a period of remission? Here, the authors are quite slient on this question as well.
If I remember correctly, the project was code-named Motor, and Microsoft partnered with Corel to release a commerical version of it. Didn't go very far.
The compiler did create MSIL (it was targeted for the CLR, and MSIL is the low-level assembly language of the CLR), and had the basic.Net libraries, which were much smaller then. This core is still quite portable, but it is a much smaller part of the the whole.Net platform now.
The basic point does remain, that there is a significant effort in bringing the whole.Net platform to another OS. The libraries are non-trivial to port for sure. I mean, the whole WCF/WWF/WPF (and Sliverlight) are just huge. Add in the ASP.Net, WinForms (as you mentioned) and well, it gets crazy. And who knows what WinRT will change.
I think the CLR has some real advantages as a langauge VM, and is a worthy alternative to the JVM in this sense, but it coming from Microsoft worries many open-source developers.
I think it'd be great if Mono focused on becoming a target VM for lots of langauges (Python, Ruby, Scala, Perl, etc) that worked really well in the Linux/Unix environment.
Doesn't reinvent all the wheels, just the ones to make programs work great in Linux/Unix, and then bring in that ecosystem to Windows.
I've always thought that Alice would be an awesome remake. More modern graphics would really add to the overall vibe of that game and storyline. It was one of the few games that really had a creepy vibe for me, and I miss that kind of game making.
The process can slow down if Oracle and IBM decided to apply their efforts in different directions. If one is pulling one way, and the other pulls the other way, then nothing happens, just a lot of strain. Now, it could be different, but so be it. Having seen IBM and Oracle reps work in as part of a OMG standardization process, it could definitely go either way. They could really get some work done in the JCP, or they can just bog it down to nothing.
I know when C# and.Net come up, it is hard to split them out from Visual Studio and the larger ecosystem, but if I look at the core languages, there seems to be a lot of cross pollination between C# and Java. And looking at the planned features for JDK 7 and 8, I'd be chomping at the bit for an implementation of them to see how they can impact existing code bases and new projects.
If you look at the C++0x efforts, a lot of vendors are adopting draft features already. Visual Studio 2010, GCC and Intel C++ all have lambda support for example and some other draft standard features. If you told me that C++ might beat Java to a standard on things like type inference and lambdas just a year or two ago, I'd scarely believe it possible. In the current political climate, who knows.
Here's hoping the good ideas get unblocked and back on track.
I really don't think bodes very well for OpenJDK. Both Oracle and IBM have tons of resources but often don't see eye to eye and aren't above using this and the JCP as a proxy for competitive battling. I see this drowning in politics and little in Java being improved in a timely manner. Meanwhile, things like.Net will accelerate with Microsoft firmly at the helm, and other open source options that are more agile (Ruby on Rails, etc.) and have more benevolent or open-minded stewards will become more popular as well.
I see less and less hope for Java adopting the positive language and library features from the C# and Ruby worlds. I am currently working on a C# project, and things like LINQ, anonymous types, extension methods (haven't used dynamic yet) and the functional/fluent programming styles they enabled enhances my productivity compared to Java.
Sometimes, I feel that Java is destined to become like COBOL, still widely used, but with the language mostly frozen in its current form, some codes with generics and annotations, some without, and with new features eschewed for backwards compatibility with older codes. I could even see codes that have generics taken out to match the Java 1.1 code to be better maintained or understood (that ? wildcard can be tricky to some).
If you read the article, it makes a very good case on why user education is not the problem. Teaching users to create really long strings of passwords doesn't help: It makes them irritable when they have to type their passwords over and over because of typos. These little costs (externalites in the article) are weighed against the distant cost of a break in and in many cases, the cost isn't worth it. I'd never use the above as a password, it'd drive me nuts. Also, the sentence itself is good enough, so why encode it?
Thanks for the reference. I should have known that Adobe would have published something on this in SIGGRAPH (as somebody noted, they publish a lot). I see I was wrong about using perceptual guidelines. But the idea of constraint regions combined with the algorithm improvements for patch finding made for nice jump in processing time and results. And in GPU acceleration and you get a great new tool that is efficient enought to support exploratory workflows.
I wish there was a paper on the core algorithms behind it (cmon Adobe, SIGGRAPH), but I could see why Adobe may be sitting on every aspect of this tool; because it sure seems to bring some real photoshop wizardry to the common user. It was really an example of "delete this thing" and it just works. Takes a common complex task and massively simplifies it. One of the most impressive marketing demos I've seen in a while.
Sure, there are some cases in which I doubt it works, but from what I could see, it seemed to have some vision and perceptual rules built in to guide how to fill in the deleted area. And frankly, it's a feature that for professionals, makes the price tag for the upgrade worth it. For some tasks, it'd pay for itself in labor alone. What would take a expert hours to do, this could do in minutes. If I was Adobe, I'd seriously consider taking this and make a Photoshop Elements Extended Edition (or whatever) and add about 79-99 bucks to this price for this feature alone. Arguably, it'd be worth it for many.
The problem is that I feel the backgrounds of many here don't get social science research and tend to over estimate what the conclusions of these studies are. Certainly, my science requirements for my CS degree were physics and chemistry. It wasn't until my postdoc that I was exposed to research design in this area.
Things like aggression and anti-social behaviors are difficult to operationalize, but there are reasonable measures. Sure, these measures are never as good as those available to hard sciences and it weakens the conclusions, but too many are quick to dismiss all studies as flawed without understanding what makes good research design in social science and what doesn't (hint, it isn't as easy as it looks).
Overall, this does seem to support that there is a weak effect on VVG on aggressive behaviors and anti-social tendencies. Why is the idea that younger children may be negatively impacted by content that is not meant for them such a hard thing to accept here? Just because some would seek to use these studies as tools for wider censorship doesn't invalidate the studies. And noted, it is a call for parents and society to have a discussion. In the end, controlling what children are exposed is a role of society as a whole. The internet doesn't make this magically go away.
I don't think doing a one-sided hypothesis test in a paper is evidence of bias in itself. In fact, the one-sided hypothesis is harder to prove (but has a larger effect size if proven).
These studies seems to measure effects on agressive behaviors or antisocial behaviors, both of which are reasonably operationalized. It seems that this meta-analysis holds that there is a weak, but significant effect in the positive. Sadly, no links to the corpus of papers in the meta-analysis, but I'd have a hard time imaging anybody would try to publish such an analysis if it wasn't at least possible to control for the major validity threats.
At the start, the open source developer got hit with a large SLAPP fine (urgh), and finally got the judgement reversed and was awarded damages, but the article notes that: "This doesn't fully compensate Jacobsen for all of his time and expense over 5 years, but it was the best he could get." So, by not using the right OSI license, the developer opened himself to years of legal hassles and woes.
Also, one wonders if by proactively suing, he ended up being worse off than by not waiting and then countersuing. Finally, it is noteworthy that since the DMCA was used on behalf of the open source developer, this may not be seen by opponents of that law as a victory at all, as it provides validation (if weak) of it's existence.
If this is victory for the little guy, I'd really hate to see what defeat is like.
But gained a lot anyway. Sure, the LSE has moved on, but the fact that a.Net application on Windows Server 2003/SQL Server 2000 could handle the LSE at all isn't a total loss for MS. I'm sure they learned a lot in this failure. Seriously,.Net 1.0 came out in 2002. In five/six years, the VM held up to a pretty high standard. Sure, the damn thing melted down for a day and getting five/six nines out of any Windows system is/will be a total black art, but doing so for a Linux system isn't a walk in the park either. There is plenty of room for this new deployment to crash. If it does, it's not an indicment of Linux; it is a statement on how hard such systems are at all levels.
Rock solid systems can be bit on both platforms. It is just a matter of if the costs and benefits are worth it. In this case, it doesn't seem the Windows solution held up. But to call this as proof on how bad MS software is seems to be hyperbole that misses the fact for a good while, it did work. Given how new all the software involved is, I'm surprised.
Slashdot Mirror
The article seems to bemoan the lack of standards and APIs for relatively new technologies like blob storage, message queuing and the like.
In short, it's the same situtation everyone deals with when choosing a platform. Take full advantage of the specific platform and lose portablity, or code to a portable subset and tradeoff ease of implementation for platform independence.
You can minimize your dependency on a third-party API if you use an API of your own to code against. Creating a set of interfaces that provide the blob, queuing and other cloud features for your project not only isolates the cloud-specifc code to one place, but it makes it more testable (using test-doubles, fakes, etc).
This is a problem, but it is a well understood one with a time tested solution.
Thank you for this. There was so much commentary that was off the point it made my head spin.
What is so hard about understanding that people do work that makes them feel not so great yet have to do it anyway? And they may have opinions about what they do?
There was so much commentary with the undertone that these women should feel lucky to just stand there and look good to get paid, that their genetics is a ticket to lifelong success.
In the end, it is a job. It requires skills and discipline. Looking like that requires knowing how to dress, to stand and pose, makeup, how to communicate with lots of different people and to be (or appear) engaged. Hell, how to stand in place for long periods doing nothing but not being able to leave.
You have to take care of yourself, you have to make tough choices, and people will resent or vilify you for tapping into the existing market for attractive people.
If you make the choice, you often have to stick with it, because there are few options for you once you are pegged at "just a pretty face".
Oh, I should note that modeling is for men and women. Frankly, the idea of having to watch every calorie and workout 10-15 hours a week just for a hope at a photo shoot makes me happy that I am a programmer.
Look, certainly there are tons of people that are shallow and self-obsessed in modeling, but those people are everywhere.
As for knowing what you are getting into, well the job is promote a product. Imagine how fun it would be trying to impart information to the forty fifth time to some guy that is just staring at your breasts and steadfastly refusing to acknowledge you as a human being. Doesn't sound like fun?
Well, maybe it's real work after all.
Actually, there is good evidence that CBT treats depression as well as anxiety disorders. The British Department of Health approved CBT as a first line treatment for depression based on a review of available evidence, and is current expanding access to trained therapists.
The problem with many studies is that they are short-term and the placebo effect in the short term is large. The mere fact that somebody is "trying do something" by being in a trial does have a noticeable impact. What is more unclear is if the placebo effect really lasts. Some studies show that CBT and well some medications are indeed better in the long term in terms of overall mood and reduction in depressive episodes.
For me, it is sad that people often are forced into a SSRI because of costs when CBT or other focused therapy could bring about more long lasting changes and better outcomes without the risks of medications.
The underlying concept isn't as hard as the review makes it sound. Using some formalisms, the author creates a set of blocks upon which you can build patterns like those in the Gang of Four book. By doing so, you can see what commonalities occur and hopefully, this leads to deeper insight into what patterns really are.
The review mentions some detection of patterns in "legacy code" but the idea of refactoring the code to better support those concepts that is mentioned in the review seems too vague to me.
There was a good book on how you could use refactoring to implement design patterns in existing code, but it wasn't as grounded on theory and automation as refactoring itself. The theory mentioned is this work could maybe serve as a foundation for such an idea.
If this sparks a new set of tools that help automate the redesign of code to follow better practices, that'd be really great. But, I think there is still some real work left to be done before that happens. And while it remains to be seen, the formalism applied could reap great benefits in that regard.
No increase in demand for teacher and professors. For those of crazy and qualified enough to teach (because we are passionate about it), we can't. I begin to see why tenure isn't necessarily a great thing.
Looking at the discussion here, I can't help but notice something missing. The reason the article was racist is because it implies that blacks by nature have given traits.
The body of genomics provides no support for this. The genetic differences between races are so small that it many biologists and anthropologists claim that the term race has no scientific meaning when applied to humans as there is no statistically significant genetic variation between race populations, esp. compared to individual differences.
What can't get lost here is whatever is happening to groups in terms of violence, crime, etc, it is a cause of the environment. It is not innate and it is not inevitable. It has nothing to do with genetic makeup, but with the circumstances in which people live and are raised in.
This notion that "whites", faced with the same level of discrimination, lack of opportunity, and similar social factors would innately respond better to those challenges is so ridiculous that it beggars belief that people still place any credence in such arguments.
Seconded. No trolling here. They want reasonable payments for creating clever algorithms and techniques for dealing with interference. Sure, it was based on existing technologies, but choosing what technologies to use and combining them correctly and effectively was a difficult and unique proposition, and I think a patentable idea.
They even shopped the technology to actual companies to make products based on it, no go.
Also, it seems that their is a claim that the IEEE standards group was aware of the patent and used the technology. Happens all the time.
So, they finally to court to get royalties for use of the patented technology. Seems to me they exhausted other avenues. Companies were dismissive of their idea as "obvious". Well, in hindsight, why wouldn't it be? So, they went to court.
Good for them. Sure, they venue shopped, but they accepted a pretty reasonable settlement from a large group of companies that greatly profited from the invention.
Bonus, the money goes back to basic research.
Code to standards (with appropriate polyfills) and ye shall prosper.
Yea, that'd be fine if there wasn't a ton of unspecified or "up-to-vendor" behavior in the HTML, CSS and DOM standards as they stand now. The reality is that any moderately complex JavaScript page has to be tested against all the browsers. The advantage of things like jQuery, etc. is they do a lot of that and hide the inconsistencies from you.
In this case, it seems tinyMCE tickled odd behavior in Gecko's window.onload event. But, as with a lot of DOM events, figuring out what should fire and in what order is a heck of a challenge, hence things like $(function() {}) in JQuery.
HTML5 may place this on solid ground, but right now HTML5 is a work in progress. In my mind, HTML5 will still have cross-browser issues, because it is really hard to get multiple vendors to truly implement a complex standard correctly and consistently. I worked on the UML2.0 standards committee, and that's exactly what happened.
Hopefully, the HTML5 test suite will be a massive regression suite, on the order of what is used for compilers that touches everything and these inconsistences will become much less common.
Firstly, he can be sentenced up to ten years. Sentencing hasn't occurred yet. In fact, the article notes the time spent will likely be less because of the nature of his background and lack of criminal history.
Also, he was offered a plea deal that included no prison time. By, rejecting this deal, he decided to take his case to the jury and accepted the chance of a harsher sentence if found guilty on the charges.
As for the motivated by bias factor that made him guilty of a hate crime, certainly, these laws are controversial and this case may lead to their re-examination.
But, it is the law of the state he was in, he was found guilty of violating it. If the jury thought he violated the law, then good for them for putting aside their personal objections to it and doing what is required of them.
If you don't like these kinds of laws, you lobby to change it. Via the courts or legislation. Maybe this case will be a basis for challenging the law in this state, for example.
All in all, this seems very simple. Don't spy on people. Don't violate their privacy. There are consequences for such actions, and those may be legal in nature.
Not every problem leads itself to GPGPU solutions. I'm no expert, but looking at a paper on a similar idea for optical reconstruction, I'd bet that the creation and update of the model via analysis (comparsion between predicted and actual results) is really hard to make parallel, and that process has a lot of non-localized memory access.
I'm sure you could use GPU acceleration for the CG calculations in the reconstruction phases, but I'm not sure that's the limiting factor here.
In short, there are really good reasons to use a lot of fast CPUs, and this sounds like one of them.
Seriously, I credit a lot of my success in CS with being taught Scheme. There are free environments that are great for teaching, and you can start with a very reasonable subset of the full language and still do some fun stuff. And the fact that it isn't widely used is a big bonus in my opinion. Languages change, library change, but the fundamentals slowly change. Scheme makes it clear that programming isn't just about a language, it's about abstraction.
Not saying that you give high schoolers a copy of SICP of course, but the fact that it is a stepping stone to that is really kind of neat in my opinion.
Let me make it clear, I find Javascript's quirks and behaviors to be very frustrating at times, so I am biased against it. And, really, there are things that need to be fixed even if that means making a updated version that breaks some older scripts, so be it. Add a attribute to script that says "Hey, use the less-brain damaged version of JS" and be done with it.
Not quite. Take the prototype for strcpy:
char *strcpy(char *dest, const char *src);
As we know, char* means a string. We aren't passing a reference to a single character. The overloading of pointers for arrays is a complicating factor. You are correct in that C does make you decide passing semantics, but it can't truly enforce them. I can happy cast away const and destroy the source string if I want.
Compare to a C# version:
string StrCopy(string src, out string dest);
Here, src is passed by value and we can't get around that, but by using the out keyword, we can get pass by reference semantics on the value. With this and auto-boxing support, I don't don't think you need pointer syntax. C++ added the concept of a reference explicitly to also address the issues of pointer semantic overloading in C. So, pointers aren't the final solution. Not a bad thing at all (when you need them, you need them), but there are other solutions to the problems you raised.
What frustrates me about web security or security in general is alluded in the review; that there is really not a good idea of what security is.
More specifically, the idea of security as a binary property. For me, it seems a more realistic approach is how much information and resource access do I have to gain to perform an action, what are the paths to do so, and how likely are those paths to occur, and what is the cost of the breach. The web makes this analysis harder without question, but still possible.
For example, I have a web site that contains account ids in a post to change something for that account. If the account id is an email, then forging the request is trivial. If the id is an opaque token, but sent in the clear (HTTP), it is less trivial, but still relatively easy. If the id is sent via https POST, sniffing is harder, but replay attacks may occur.
And so on. The point is to decide when the advantage gained is overwhelmed by the cost and risk of the attack.
I remember a time when somebody was worried about GUIDs in a URL because you could guess a valid one. Of course, it is much easier (understatement) to capture a valid one than guess one. This book will have value if it helps you avoid red herrings like that and focus on the real threats.
Of course, I am not a security specialist, so this may be naive at best.
Nothing wrong with a good diet, and proper D3 levels are already tested by a lot of neurologists and supplements are recommended to maintain them in MS suffers.
But, of course, the authors claims that good food is all you need. It all falls down when I saw the recommendation not to take interferons, which, unlike this diet, has a huge amount of research that shows that it reduces relapses for those with relapsing-remitting MS, and lengthens the periods between replases.
Given how little is understood about MS, it is greatly specious to claim that MS can be cured by a diet. And to assume that the doctors behind this work don't have their own biases and agendas just because they don't toe the "party line" in research seems a bit naive.
Finally, given that MS has periods of remission, how can you claim that MS is reversed vs being in a period of remission? Here, the authors are quite slient on this question as well.
If I remember correctly, the project was code-named Motor, and Microsoft partnered with Corel to release a commerical version of it.
Didn't go very far.
The compiler did create MSIL (it was targeted for the CLR, and MSIL is the low-level assembly language of the CLR), and had the basic .Net libraries, which were much smaller then. This core is still quite portable, but it is a much smaller part of the the whole .Net platform now.
The basic point does remain, that there is a significant effort in bringing the whole .Net platform to another OS. The libraries are non-trivial to port for sure. I mean, the whole WCF/WWF/WPF (and Sliverlight) are just huge. Add in the ASP .Net, WinForms (as you mentioned) and well, it gets crazy. And who knows what WinRT will change.
I think the CLR has some real advantages as a langauge VM, and is a worthy alternative to the JVM in this sense, but it coming from Microsoft worries many open-source developers.
I think it'd be great if Mono focused on becoming a target VM for lots of langauges (Python, Ruby, Scala, Perl, etc) that worked really well in the Linux/Unix environment.
Doesn't reinvent all the wheels, just the ones to make programs work great in Linux/Unix, and then bring in that ecosystem to Windows.
I've always thought that Alice would be an awesome remake. More modern graphics would really add to the overall vibe of that game and storyline. It was one of the few games that really had a creepy vibe for me, and I miss that kind of game making.
The process can slow down if Oracle and IBM decided to apply their efforts in different directions. If one is pulling one way, and the other pulls the other way, then nothing happens, just a lot of strain. Now, it could be different, but so be it. Having seen IBM and Oracle reps work in as part of a OMG standardization process, it could definitely go either way. They could really get some work done in the JCP, or they can just bog it down to nothing.
I know when C# and .Net come up, it is hard to split them out from Visual Studio and the larger ecosystem, but if I look at the core languages, there seems to be a lot of cross pollination between C# and Java. And looking at the planned features for JDK 7 and 8, I'd be chomping at the bit for an implementation of them to see how they can impact existing code bases and new projects.
If you look at the C++0x efforts, a lot of vendors are adopting draft features already. Visual Studio 2010, GCC and Intel C++ all have lambda support for example and some other draft standard features. If you told me that C++ might beat Java to a standard on things like type inference and lambdas just a year or two ago, I'd scarely believe it possible. In the current political climate, who knows.
Here's hoping the good ideas get unblocked and back on track.
I really don't think bodes very well for OpenJDK. Both Oracle and IBM have tons of resources but often don't see eye to eye and aren't above using this and the JCP as a proxy for competitive battling. I see this drowning in politics and little in Java being improved in a timely manner. Meanwhile, things like .Net will accelerate with Microsoft firmly at the helm, and other open source options that are more agile (Ruby on Rails, etc.) and have more benevolent or open-minded stewards will become more popular as well.
I see less and less hope for Java adopting the positive language and library features from the C# and Ruby worlds. I am currently working on a C# project, and things like LINQ, anonymous types, extension methods (haven't used dynamic yet) and the functional/fluent programming styles they enabled enhances my productivity compared to Java.
Sometimes, I feel that Java is destined to become like COBOL, still widely used, but with the language mostly frozen in its current form, some codes with generics and annotations, some without, and with new features eschewed for backwards compatibility with older codes. I could even see codes that have generics taken out to match the Java 1.1 code to be better maintained or understood (that ? wildcard can be tricky to some).
If you read the article, it makes a very good case on why user education is not the problem. Teaching users to create really long strings of passwords doesn't help: It makes them irritable when they have to type their passwords over and over because of typos. These little costs (externalites in the article) are weighed against the distant cost of a break in and in many cases, the cost isn't worth it. I'd never use the above as a password, it'd drive me nuts. Also, the sentence itself is good enough, so why encode it?
Thanks for the reference. I should have known that Adobe would have published something on this in SIGGRAPH (as somebody noted, they publish a lot). I see I was wrong about using perceptual guidelines. But the idea of constraint regions combined with the algorithm improvements for patch finding made for nice jump in processing time and results. And in GPU acceleration and you get a great new tool that is efficient enought to support exploratory workflows.
I wish there was a paper on the core algorithms behind it (cmon Adobe, SIGGRAPH), but I could see why Adobe may be sitting on every aspect of this tool; because it sure seems to bring some real photoshop wizardry to the common user. It was really an example of "delete this thing" and it just works. Takes a common complex task and massively simplifies it. One of the most impressive marketing demos I've seen in a while.
Sure, there are some cases in which I doubt it works, but from what I could see, it seemed to have some vision and perceptual rules built in to guide how to fill in the deleted area. And frankly, it's a feature that for professionals, makes the price tag for the upgrade worth it. For some tasks, it'd pay for itself in labor alone. What would take a expert hours to do, this could do in minutes. If I was Adobe, I'd seriously consider taking this and make a Photoshop Elements Extended Edition (or whatever) and add about 79-99 bucks to this price for this feature alone. Arguably, it'd be worth it for many.
The problem is that I feel the backgrounds of many here don't get social science research and tend to over estimate what the conclusions of these studies are. Certainly, my science requirements for my CS degree were physics and chemistry. It wasn't until my postdoc that I was exposed to research design in this area.
Things like aggression and anti-social behaviors are difficult to operationalize, but there are reasonable measures. Sure, these measures are never as good as those available to hard sciences and it weakens the conclusions, but too many are quick to dismiss all studies as flawed without understanding what makes good research design in social science and what doesn't (hint, it isn't as easy as it looks).
Overall, this does seem to support that there is a weak effect on VVG on aggressive behaviors and anti-social tendencies. Why is the idea that younger children may be negatively impacted by content that is not meant for them such a hard thing to accept here? Just because some would seek to use these studies as tools for wider censorship doesn't invalidate the studies. And noted, it is a call for parents and society to have a discussion. In the end, controlling what children are exposed is a role of society as a whole. The internet doesn't make this magically go away.
I don't think doing a one-sided hypothesis test in a paper is evidence of bias in itself. In fact, the one-sided hypothesis is harder to prove (but has a larger effect size if proven).
These studies seems to measure effects on agressive behaviors or antisocial behaviors, both of which are reasonably operationalized. It seems that this meta-analysis holds that there is a weak, but significant effect in the positive. Sadly, no links to the corpus of papers in the meta-analysis, but I'd have a hard time imaging anybody would try to publish such an analysis if it wasn't at least possible to control for the major validity threats.
At the start, the open source developer got hit with a large SLAPP fine (urgh), and finally got the judgement reversed and was awarded damages, but the article notes that: "This doesn't fully compensate Jacobsen for all of his time and expense over 5 years, but it was the best he could get." So, by not using the right OSI license, the developer opened himself to years of legal hassles and woes.
Also, one wonders if by proactively suing, he ended up being worse off than by not waiting and then countersuing. Finally, it is noteworthy that since the DMCA was used on behalf of the open source developer, this may not be seen by opponents of that law as a victory at all, as it provides validation (if weak) of it's existence.
If this is victory for the little guy, I'd really hate to see what defeat is like.
But gained a lot anyway. Sure, the LSE has moved on, but the fact that a .Net application on Windows Server 2003/SQL Server 2000 could handle the LSE at all isn't a total loss for MS. I'm sure they learned a lot in this failure. Seriously, .Net 1.0 came out in 2002. In five/six years, the VM held up to a pretty high standard. Sure, the damn thing melted down for a day and getting five/six nines out of any Windows system is/will be a total black art, but doing so for a Linux system isn't a walk in the park either. There is plenty of room for this new deployment to crash. If it does, it's not an indicment of Linux; it is a statement on how hard such systems are at all levels.
Rock solid systems can be bit on both platforms. It is just a matter of if the costs and benefits are worth it. In this case, it doesn't seem the Windows solution held up. But to call this as proof on how bad MS software is seems to be hyperbole that misses the fact for a good while, it did work. Given how new all the software involved is, I'm surprised.