...but on the other hand, it has all but killed newspapers' cash cow, the classifieds. You can say, "who cares, newspapers are dying anyway," but the resources newspapers have - trained journalists, editors, the whole damn infrastructure - is quite valuable to our society as a whole.
I think there's a flaw in this logic. It only considers the entity who is receiving less money (newspapers) and the entity that is receiving very little money (Craigslist). It doesn't consider the true beneficiary of the Craigslist services, the people saving money on the service. That money doesn't just vanish into thin air, it gets spent in other ways or, in seemingly rare cases, saved. The positive effect of newspapers needs to be balanced against the positive effect that this money will create when spent in other ways.
Not that everything you've said isn't quite interesting (on the contrary, I found it quite interesting), but it's predicated on the notion that the most likely path to monetization of CL's popularity is to increase prices to "collect monopoly rents." However the path to increased profits that's been suggested most often is for CL to add AdSense to some or all of their pages. By some estimates, this could be as much as $500m/year that they'd take in from simply showing text ads. It's arguable whether the text ad nuisance would create a situation where a competitor could undercut them by offering essentially the same service without ads, but the prevailing sentiment is that users wouldn't be motivated to use a competitor that doesn't have the same community/momentum that CL has.
What would be truly interesting would be to see their community's reaction to monitizing the site in a way that was consistant with the current ideals. Would the community actually welcome text ads if they knew that 80% of the revenue was being donated to charity? What if there was an option to turn remove ads for users that felt they were annoying? Would some user-centric strategy towards increasing profits be seen as a money-grab on CL's part or as an effort to use ads as a vehicle towards positive change?
The modularity lacking in Java is packaging and importing. As another poster mentioned, toolkits create the illusion that JavaScript has these, but it really doesn't. Somewhat less important is that there really isn't a true inheritance model. You can inherit another object's prototype, but that doesn't give you the same flexibility as true inheritance.
If all you want is to run LISP with Java's speed, JDK 1.6 doesn't really change anything for you. There have been LISP implemntations (here are a couple of options) for some time now.
What JSR-223 means is that it's now possible to integrate scripting languages into Java programs without having to use a project-specific API. As far as I can tell, the only LISP(ish) implementation that supports JSR-223 is SISC. This means that you can write your Java in such a way that you can very simply switch scripting language or framework without much modification to your Java code. It also means there is a standard way for scripting languages to interact with a running Java program. This is nice for anyone who wants to write a hybrid application where some parts naturally lend themselves to being implemented in Java and some parts don't.
Microsoft spends $7 billion dollars annually on research and development and a large number, albeit much smaller than that, defending themselves against patent-related lawsuits (Eolas ring any bells?)
While Gates may want software patents abolished so that his company, with vastly more resources than anyone else, can develop whatever they want and react to new software trends without fear of someone having patented the idea, the net result would be the same as what so many open-source fans have been clamoring for...doing away with (or at least significantly raising the bar for) software patents.
What we really need to do is stop electing old-money millionaires. New-money [m|b]illionaires often have proved their skills in attaining their wealth. Things like fiscal responsibility and ability to predict how the world will change as time passes are often key ingredients in people amassing a large fortune without receiving a large part of it from parents.
While I'm about the biggest Microsoft critic there is here, would it really be so bad to have a president who: a) Understands how bad an unbalanced budget really is. b) Sees the value of research and development as a means of staying ahead of the competition. c) Has no need to be beholden to any campaign contributors (why would Gates even bother to waste his time raising money when the cost of a campaign is a tiny fraction of his net worth?) d) Understands that the internet isn't just a series of tubes. e) Actually consider the position of hardware manufacturers when content companies decide they need some incredibly short-sighted piece of legislation to retain their monopoly. f) Understands the problems inherant in the US patent situation and how it stifles innovation by both small companies / individuals as well as for large companies like Microsoft.
It would seem that regardless of political affiliation, a Gates presidency would mean a positive difference in a number of the slashdot-discussed political issues.
Another option for creating XML-based GUIs in java is XSWT. I've found that it makes laying out a GUI no harder than laying out an HTML document (or it would be no harder if they'd improve their docs...grr), yet it still gives you the ability to manage all behvavior in Java.
However, it's not Swing, which may be a problem for some (I personally like SWT better than Swing, but understand that some believe the opposite).
The problem with calling devices "cellular", whether or not it is apt, is that it's a bad name. When naming something you want to describe its function, not its implementation. "Mobile Phone" is a much better name since it describes how the phone is used rather than how that use is achieved. That way, you can have cellular mobile phone, satellite mobile phone and even some-unknown-future-technology mobile phone. The "mobile phone" part never changes even when the underlying technology does.
OT I know, but the last time I was there was in 1988, about 9 months before the invasion of Kuwait. It was a really nice place and the people there were very friendly. Can you say the same? By most accounts, Iraq is now a primitive, war-torn country where American's are feared, distrusted and demonized. Saddam was no angel, but from everything that I've read, life has gotten worse, not better, for Iraqis since he was deposed.
You're free to believe whatever you want, including that you were serving your country. But you also need to allow for the possibility that many people out there will feel that you've done something that hasn't been in either the best interests of our country (yes, I'm American) or the best interests of Iraq.
Heh...I guess it wasn't exactly clear what generation was supposed to mean, but I wasn't really referring to a sociological generation per se, but to the dichotomy between those who embrace technology vs. those who fear it. I used the term generation because, to a large extent, the boundary between these two groups is governed by age. Of course there are exceptions to this and I'm guessing that since you're reading slashdot, you identify more with younger generations where technology is concerned.
But I do believe that the days of policy being dictated by technophobes are winding down. Even if we have a decade or more to live through with the baby-boomers in power, the population (and thereby the electorate) will become increasingly technologically savvy. That will mean that those in power who wish to stay in power will need to have aides who can better make technology understandable and relatable to them rather than the situation we're currently seeing (e.g. the now infamous "tubes" analogy). Equally so, I also feel that the days of the technophobe pundit are also winding down. Rants like O'Reilly's will become increasingly unpopular as elderly listeners die off and are replaced by listeners from the (to use a term I hate) "myspace generation."
Even languages that would end up as mostly 3 byte characters tend to benefit from the savings on single byte characters for control and formatting markup.
To further your point, compression techniques like gzip make the difference between UTF-16 and UTF-8 even smaller since they recognize which characters or character sequences are being used and allow them to be encoded using less bytes. A 3-byte UTF-8 character and a 2-byte UTF-16 character (as part of a document of any reasonable size) are represented by gzip in basically the same number of bytes.
But even this is a minor concern...with the size of hard drives and the bandwidth capabilities we have these days, the number of bytes used to represent text is pretty insignificant. Even in the worst case scenario of UTF-8 (IIRC, a 5-byte character), my broadband connection can still transfer 160,000 of those characters per second. Trading the backwards-compatible nature of UTF-8 for the few bytes saved by UTF-16 makes very little sense.
We should be using this opportunity to actively get rid of languages.
What you don't seem to be groking is that language is an integral part of people's identities. The french are proud of their language...losing it would be like losing a part of themselves. Your link even mentions that Chinese people take the difficulty of their language as a badge of honor. Every language has some measure of this.
Ignoring how foolish it would be to try to ensure that there is only one language, the natural candidate for "universal language" would be the one that was created specifically for that purpose, Esperanto. It was specifically designed to be easy to learn and pronounce for as many people as possible. And, more importantly, adopting it would mean that everyone would be forced to learn a new language. It would make far more sense to try to ensure that eveyrone speaks two languages, their native langauge and Esperanto. This would allow people to contextually use language depending on whether it was intended for consumption by everyone on earth (passport, street sign, technical manual, etc) or whether it was intended specifically for people who speak their language. Most sci-fi visions of the future where a common language is realized adopt this scenario of a single, common language that everyone speaks without eliminating the other languages entirely.
However my personal feeling is that we should be actively using pre-school and early grade school (ages 3-7 or so) to teach only languages and reading to children. Children are so receptive to learning languages at that age, that it's very possible to teach 4+ languages during that time span. And when people speak that many languages natively, they can pick up languages far more easily as adults. I only wish I had been given this sort of education...I know it would have made it far more easy for me to get to the point where I am now (can speak and read 4 languages fluently and a number of others with varying degrees of ability).
Reality is people from Bill's generation selling our country to corporate interests, destroying our environment, sending our generation off to be killed in Iraq, spending umpteen trillion dollars of money they don't actually have with the full realization that they'll all be dead long before the time comes to pay up and pointlessly banning activities and substances that they only don't partake in because their bodies are too old and frail to allow them to enjoy.
Reality is that his generation is using their last gasps at power to fuck everything up for our generation. Is it any wonder that we want to divorce ourselves from his so-called reality?
When projects like this get going and the emphasis should be on subject matter experts (SME), the projects usually get expendable high-level highly paid deadweight -- I've seen it too many times.
I think you've hit the nail on the head with this statement. The government made a poor decision on who should run the project and he, in turn, made quite possibly the most bone-headed decisions he could possibly make.
From the article:
Granger commissioned the management consulting company McKinsey to do a study...[that]... concluded... that no single existing vendor was big enough to act as prime contractor on the countrywide, multibillion-dollar initiative the NHS was proposing....The result: He divided England into five regions--London; Eastern; Northeast; Northwest with West Midlands; and Southern--each with a population of about 10 million.
That last sentence made my jaw drop. How someone in his position could so blatantly avoid consulting anyone with any technical acumen is beyond me. Yes, it's possible, that no single vendor was capable of creating such a system alone. But the vast majority of a project like this is about creating a single process for every use case that the system is designed to handle. As such, the project shouldn't ever be broken down into groupings like number of patients in the system. Computers are great at handling really large numbers when the software is designed up-front to scale to really large numbers. The system should have been broken down into separate processes for which individual vendors would handle that single process (or grouping of processes) for everyone in the country.
The X-Ray example is a perfect one. Why would anyone in their right mind have 5 separate vendors all attempt to implement a solution for the problem that was only applied to the region they managed? At best, one region would end up with a solution that was better than every other region. However a competant management decision would have been to look for a vendor that could handle *only* the process of integrating the country's X-ray facilities with the country's high-speed data network. Another vendor would be responsible for supplying and maintaing that network. Still another vendor would be responsible for maintaining the huge data center (or centers) where information was housed. Just off the top of my head, GE could be responsible for the X-Ray integration (I know they have the necessary expertise), BT could handle high-speed network (among others, but why use foreign expertise when a UK company could handle it). And there are any number of competant vendors that could handle a high-availability server environment with a massive database.
Basically, had they had anyone with have an ounce of technical acumen, they would have devided the project up along functional boundaries of the application rather than regional boundaries of the country. That way, even if some of the projects went horribly over-budget, at least some of the project would be useful. Now, because of the inept management decisions, the whole thing is a train wreck.
The thing that MySQL does better than any other database, and the reason why it continues to be supported by so many open source projects, is how well it works in a shared environment. Namely, the $5/mo webhosting environment. That doesn't mean that it can't be done with other databases, but mysql makes everything really simple from the hosting provider's standpoint. And most open source web applications expect to be installed in this kind of environment at least some of the time.
However the benefit of MySQL drops off substantially for custom-built applications that run on one or more dedicated servers. When you have full control over everything is when MySQL becomes more pain than it's worth. An all this would be fine, except that some people making the decision on which database to use when they have full control over everything often mistake MySQL's success for quality and reliability.
...but because it's just XML, which means you can create data-based SVG images "out of thin air" with PHP or the scripting language of your choice.
This was never the reason why SVG was promising. There have been imagine manipulation libraries available to scripting languages for a long time. I remember using GD in Perl to create GIF images "out of thin air" as you put it. That was in 1996.
The strength of SVG is that it's vector-based. This allows developers to produce images that can be resized on-the-fly by the user's browser with no loss in image quality. This becomes a particularly good thing when you start to design CSS-based webpages that make no assumptions about the user's monitor resolution. So long as dimensions of elements are specified as percentages rather than pixels, SVG allows a webpage to look great at everything from 800x600 to the insanely high resolutions that monitors/video cards are capable of these days. And it's not just for webpages. SVG makes a great technology for icons in any UI. OS X already has SVG rendering built into Aqua and I believe that there are Linux window managers that have also integrated it into their offering.
SVG is the key for enabling us to use higher resolutions without having to have 20/10 vision to see everything that's based around the concept of a pixel.
The thing to remember when talking about JPEG files is that you're really not talking about JPEG. You're talking about JFIF (the file format) and not JPEG (the compression algorithm). JFIF, similar to what mp3 does for audio, only specifies how images are decoded, not how they are encoded. That means the encoding algorithm can differ between implementations with some performing better than others. From everything I've heard, the baseline JPEG compression (read: worst possible performance) has been patent-free since the beginning. The algorithm in question is probably one that produces smaller files while maintaining the same image quality. The Gimp is probably either using the baseline algorithm or one that incorporates their own improvements.
WMV and Quicktime are a whole different beast. In those cases, the entire file format is proprietary.
P.S. The above is from what I was able to gather in about a day of research into what it would take to write my own jpeg library. If anyone notices anything in this that isn't correct, please point it out.
Ugh. I'm a native English speaker and I shudder to think of what life would be like if English became the universal language.
But then again, I've gone to a lot of effort to learn 3 other languages and I believe that many of the wonderful experiences I've had in other countries have been partly facilitated by the people I meet knowing that I've gone to a lot of effort to be able to try to experience their culture as more than just a casual tourist. Even in countries where I don't speak the language, it's amazing how much more people will accept you if you just take the simple step of learning a few phrases in their language. I'm always annoyed when other Americans talk about how people in other countries are rude to tourists when they've really been rude to them by having the arrogance to assume that their hosts should go the extra mile to learn English.
I'm not sure this is really a case I need to make, since I think most people can recognize the value of having so many languages.
As far as internationalized URLs, there is really no reason why we shouldn't be able to do it quite simply. So long as UTF-8 is used, all existing domains would need absolutely no change since I don't believe extended ASCII characters are used in domain names.
He also doesn't take into account that some exercises can actually be detrimental to depressive people. Exercises that cause the body to produce adrenaline can actually make someone's depression worse. When the body releases adrenaline it also releases cortisol. While the adrenaline levels subside shortly after you finish exercising, cortisol levels take far longer to normalize. While the causes of depression aren't fully understood, there have been studies that have shown a connection between cronically-elevated cortisol levels and depression.
So while exercise is generally one recommended activity for depressives as it can help create an environment where the other treatments can be more effective, you have to be careful which exercises you engage in. And, like you said, exercise isn't a magic bullet, it's only part of eliminating one of the obstacles to getting better.
The problem with IE5 for Mac is that it was released on a staggered development cycle from IE on the PC. This led to radically different functionality depending on the platform. IE for the Mac was a huge step forward (for IE at the time, that is) when it comes to standards compliance. But it doesn't help developers to comply to standards in one place and not the other. So when the next version of IE came out on Windows, it was in some ways more standards compliant than IE for Mac. But it was also somewhat less compliant in other areas. Had they chosen to do a straight port to the Mac--bugs, foibles and all--it would have been preferable. Even if it was less standards compliant than IE 5 for the Mac ended up being.
There's nothing magic about the standards that makes them better than any other browser target. The only reason developers are constantly clamoring for standards compliance is that is presents a single target to code to. We don't care which target we code to, we just don't want to have to support multiple targets. And that's why Microsoft's refusal to comply with standards is so frustrating. Every new IE release creates a new target to code for (though hopefully the release of IE7 will mean we can finally stop testing for IE 5.5).
This is why Hachamovitch's mention of needing to maintain compatability with other products at Microsoft is somewhat disingenuous. That compatability is only a problem because they've chosen to deviate from web standards both in IE and in the other products. Standards compliance can become a mechanism for allowing projects to work independently from each other without fear that they will need anything more than a couple of tweaks before launching to ensure that the two products work well together. But since they've chosen to deviate from standards (I won't go so far as to say that they've intentionally deviated, but it does seem that way at times), all those project teams need to work closely together to ensure that changes to IE don't break everyone else's code.
Not that I agree with removing any of the ancient Egyptian antiquities from Egypt, but there is a huge difference between removing them for display at a prominent museum like the British Museum and removing them to be sold on the black market such that they will likely never again be available to be seen by the public.
A good example of this is the mummy of Ramses I. If this had been pillaged by archaeologists on behalf of the British Museum, it would be in much better shape that it is currently. However, because it was unearthed by tomb robbers, it spent over 100 years at a museum in Niagra Falls with very little concern for maintaining it and absolutely no indication given to visitors that it was, in fact, the mummy of a Pharoah. An "expert" grave robber would have followed much a much more strict procedure to ensure that it was properly cared for and properly catalogued (if only to increase the value, but still).
That said, the Ramses I mummy did end up in Egypt, which almost makes up for the shoddy maintenance it received over the course of its post-excavation life (museums around the world should follow the example of the Carlos Museum at Emory University and return everything that was stolen from Egypt)...
According to the Oxford American Dictionary, you're wrong. They define intelligence as "The ability to acquire and apply knowledge and skills." (I don't have an OED on hand, but I'd imagine their definition is similar). By that definition, a person's musical talents would be considered a form of intelligence.
What the IQ tests measure is Intellect (defined as "The faculty of reasoning and understanding objectively, esp. with regard to abstract or academic matters"). Both words come from the same Latin root, but their meanings are subtly different.
But I understand where you're coming from...somehow I think/. types are better able to understand the separation of these traits because we, at one point in our lives, determined these characteristics by rolling dice for each of them separately.
Whether the assertion of "production grade" is true or not is completely irrelevant when it comes time to sell the company. So long as you can hack it together well enough that you can scale it up to a number of users that will convince someone to buy you out, long-term sustainability of the code isn't necessary. If you can hit that performance metric, you can throw around the "production grade" terminology all you want.
Engineers too often forget that technical superiority often means very little in the short term. Time to market, marketing and sales often contribute far more towards short-term success than sound engineering does. Long term is an entirely different matter, but most of these Bubble 2.0 buyouts happen so quickly that there isn't time for the technical mistakes to become a problem. And it's pretty rare that the buying company is purchasing the smaller company for their actual technology, what they're really buying is the brand and userbase.
Take the case of the YouTube buyout for example...Google paid $1.6b for it and there's absolutely nothing there that Google couldn't rewrite in under a year in manner that's maintainable going forward. It might take 10-15 engineers on to do it, but that's roughly 3-4 million on an investment roughly 500 times that much.
Slashdot Mirror
Not that everything you've said isn't quite interesting (on the contrary, I found it quite interesting), but it's predicated on the notion that the most likely path to monetization of CL's popularity is to increase prices to "collect monopoly rents." However the path to increased profits that's been suggested most often is for CL to add AdSense to some or all of their pages. By some estimates, this could be as much as $500m/year that they'd take in from simply showing text ads. It's arguable whether the text ad nuisance would create a situation where a competitor could undercut them by offering essentially the same service without ads, but the prevailing sentiment is that users wouldn't be motivated to use a competitor that doesn't have the same community/momentum that CL has.
What would be truly interesting would be to see their community's reaction to monitizing the site in a way that was consistant with the current ideals. Would the community actually welcome text ads if they knew that 80% of the revenue was being donated to charity? What if there was an option to turn remove ads for users that felt they were annoying? Would some user-centric strategy towards increasing profits be seen as a money-grab on CL's part or as an effort to use ads as a vehicle towards positive change?
The modularity lacking in Java is packaging and importing. As another poster mentioned, toolkits create the illusion that JavaScript has these, but it really doesn't. Somewhat less important is that there really isn't a true inheritance model. You can inherit another object's prototype, but that doesn't give you the same flexibility as true inheritance.
If all you want is to run LISP with Java's speed, JDK 1.6 doesn't really change anything for you. There have been LISP implemntations (here are a couple of options) for some time now.
What JSR-223 means is that it's now possible to integrate scripting languages into Java programs without having to use a project-specific API. As far as I can tell, the only LISP(ish) implementation that supports JSR-223 is SISC. This means that you can write your Java in such a way that you can very simply switch scripting language or framework without much modification to your Java code. It also means there is a standard way for scripting languages to interact with a running Java program. This is nice for anyone who wants to write a hybrid application where some parts naturally lend themselves to being implemented in Java and some parts don't.
Microsoft spends $7 billion dollars annually on research and development and a large number, albeit much smaller than that, defending themselves against patent-related lawsuits (Eolas ring any bells?)
While Gates may want software patents abolished so that his company, with vastly more resources than anyone else, can develop whatever they want and react to new software trends without fear of someone having patented the idea, the net result would be the same as what so many open-source fans have been clamoring for...doing away with (or at least significantly raising the bar for) software patents.
Well, when you consider it, all readers are technically literate.
However, technical acumen is a whole different matter...
What we really need to do is stop electing old-money millionaires. New-money [m|b]illionaires often have proved their skills in attaining their wealth. Things like fiscal responsibility and ability to predict how the world will change as time passes are often key ingredients in people amassing a large fortune without receiving a large part of it from parents.
While I'm about the biggest Microsoft critic there is here, would it really be so bad to have a president who:
a) Understands how bad an unbalanced budget really is.
b) Sees the value of research and development as a means of staying ahead of the competition.
c) Has no need to be beholden to any campaign contributors (why would Gates even bother to waste his time raising money when the cost of a campaign is a tiny fraction of his net worth?)
d) Understands that the internet isn't just a series of tubes.
e) Actually consider the position of hardware manufacturers when content companies decide they need some incredibly short-sighted piece of legislation to retain their monopoly.
f) Understands the problems inherant in the US patent situation and how it stifles innovation by both small companies / individuals as well as for large companies like Microsoft.
It would seem that regardless of political affiliation, a Gates presidency would mean a positive difference in a number of the slashdot-discussed political issues.
Another option for creating XML-based GUIs in java is XSWT. I've found that it makes laying out a GUI no harder than laying out an HTML document (or it would be no harder if they'd improve their docs...grr), yet it still gives you the ability to manage all behvavior in Java.
However, it's not Swing, which may be a problem for some (I personally like SWT better than Swing, but understand that some believe the opposite).
The problem with calling devices "cellular", whether or not it is apt, is that it's a bad name. When naming something you want to describe its function, not its implementation. "Mobile Phone" is a much better name since it describes how the phone is used rather than how that use is achieved. That way, you can have cellular mobile phone, satellite mobile phone and even some-unknown-future-technology mobile phone. The "mobile phone" part never changes even when the underlying technology does.
You're free to believe whatever you want, including that you were serving your country. But you also need to allow for the possibility that many people out there will feel that you've done something that hasn't been in either the best interests of our country (yes, I'm American) or the best interests of Iraq.
Heh...I guess it wasn't exactly clear what generation was supposed to mean, but I wasn't really referring to a sociological generation per se, but to the dichotomy between those who embrace technology vs. those who fear it. I used the term generation because, to a large extent, the boundary between these two groups is governed by age. Of course there are exceptions to this and I'm guessing that since you're reading slashdot, you identify more with younger generations where technology is concerned.
But I do believe that the days of policy being dictated by technophobes are winding down. Even if we have a decade or more to live through with the baby-boomers in power, the population (and thereby the electorate) will become increasingly technologically savvy. That will mean that those in power who wish to stay in power will need to have aides who can better make technology understandable and relatable to them rather than the situation we're currently seeing (e.g. the now infamous "tubes" analogy). Equally so, I also feel that the days of the technophobe pundit are also winding down. Rants like O'Reilly's will become increasingly unpopular as elderly listeners die off and are replaced by listeners from the (to use a term I hate) "myspace generation."
Even languages that would end up as mostly 3 byte characters tend to benefit from the savings on single byte characters for control and formatting markup.
To further your point, compression techniques like gzip make the difference between UTF-16 and UTF-8 even smaller since they recognize which characters or character sequences are being used and allow them to be encoded using less bytes. A 3-byte UTF-8 character and a 2-byte UTF-16 character (as part of a document of any reasonable size) are represented by gzip in basically the same number of bytes.
But even this is a minor concern...with the size of hard drives and the bandwidth capabilities we have these days, the number of bytes used to represent text is pretty insignificant. Even in the worst case scenario of UTF-8 (IIRC, a 5-byte character), my broadband connection can still transfer 160,000 of those characters per second. Trading the backwards-compatible nature of UTF-8 for the few bytes saved by UTF-16 makes very little sense.
Ignoring how foolish it would be to try to ensure that there is only one language, the natural candidate for "universal language" would be the one that was created specifically for that purpose, Esperanto. It was specifically designed to be easy to learn and pronounce for as many people as possible. And, more importantly, adopting it would mean that everyone would be forced to learn a new language. It would make far more sense to try to ensure that eveyrone speaks two languages, their native langauge and Esperanto. This would allow people to contextually use language depending on whether it was intended for consumption by everyone on earth (passport, street sign, technical manual, etc) or whether it was intended specifically for people who speak their language. Most sci-fi visions of the future where a common language is realized adopt this scenario of a single, common language that everyone speaks without eliminating the other languages entirely.
However my personal feeling is that we should be actively using pre-school and early grade school (ages 3-7 or so) to teach only languages and reading to children. Children are so receptive to learning languages at that age, that it's very possible to teach 4+ languages during that time span. And when people speak that many languages natively, they can pick up languages far more easily as adults. I only wish I had been given this sort of education...I know it would have made it far more easy for me to get to the point where I am now (can speak and read 4 languages fluently and a number of others with varying degrees of ability).
Reality is that his generation is using their last gasps at power to fuck everything up for our generation. Is it any wonder that we want to divorce ourselves from his so-called reality?
From the article:
That last sentence made my jaw drop. How someone in his position could so blatantly avoid consulting anyone with any technical acumen is beyond me. Yes, it's possible, that no single vendor was capable of creating such a system alone. But the vast majority of a project like this is about creating a single process for every use case that the system is designed to handle. As such, the project shouldn't ever be broken down into groupings like number of patients in the system. Computers are great at handling really large numbers when the software is designed up-front to scale to really large numbers. The system should have been broken down into separate processes for which individual vendors would handle that single process (or grouping of processes) for everyone in the country.
The X-Ray example is a perfect one. Why would anyone in their right mind have 5 separate vendors all attempt to implement a solution for the problem that was only applied to the region they managed? At best, one region would end up with a solution that was better than every other region. However a competant management decision would have been to look for a vendor that could handle *only* the process of integrating the country's X-ray facilities with the country's high-speed data network. Another vendor would be responsible for supplying and maintaing that network. Still another vendor would be responsible for maintaining the huge data center (or centers) where information was housed. Just off the top of my head, GE could be responsible for the X-Ray integration (I know they have the necessary expertise), BT could handle high-speed network (among others, but why use foreign expertise when a UK company could handle it). And there are any number of competant vendors that could handle a high-availability server environment with a massive database.
Basically, had they had anyone with have an ounce of technical acumen, they would have devided the project up along functional boundaries of the application rather than regional boundaries of the country. That way, even if some of the projects went horribly over-budget, at least some of the project would be useful. Now, because of the inept management decisions, the whole thing is a train wreck.
The thing that MySQL does better than any other database, and the reason why it continues to be supported by so many open source projects, is how well it works in a shared environment. Namely, the $5/mo webhosting environment. That doesn't mean that it can't be done with other databases, but mysql makes everything really simple from the hosting provider's standpoint. And most open source web applications expect to be installed in this kind of environment at least some of the time.
However the benefit of MySQL drops off substantially for custom-built applications that run on one or more dedicated servers. When you have full control over everything is when MySQL becomes more pain than it's worth. An all this would be fine, except that some people making the decision on which database to use when they have full control over everything often mistake MySQL's success for quality and reliability.
The strength of SVG is that it's vector-based. This allows developers to produce images that can be resized on-the-fly by the user's browser with no loss in image quality. This becomes a particularly good thing when you start to design CSS-based webpages that make no assumptions about the user's monitor resolution. So long as dimensions of elements are specified as percentages rather than pixels, SVG allows a webpage to look great at everything from 800x600 to the insanely high resolutions that monitors/video cards are capable of these days. And it's not just for webpages. SVG makes a great technology for icons in any UI. OS X already has SVG rendering built into Aqua and I believe that there are Linux window managers that have also integrated it into their offering.
SVG is the key for enabling us to use higher resolutions without having to have 20/10 vision to see everything that's based around the concept of a pixel.
Thanks for the clarification. I had been trusting the information on this page to be accurate...glad I didn't actually go a head with implementing it.
The thing to remember when talking about JPEG files is that you're really not talking about JPEG. You're talking about JFIF (the file format) and not JPEG (the compression algorithm). JFIF, similar to what mp3 does for audio, only specifies how images are decoded, not how they are encoded. That means the encoding algorithm can differ between implementations with some performing better than others. From everything I've heard, the baseline JPEG compression (read: worst possible performance) has been patent-free since the beginning. The algorithm in question is probably one that produces smaller files while maintaining the same image quality. The Gimp is probably either using the baseline algorithm or one that incorporates their own improvements.
WMV and Quicktime are a whole different beast. In those cases, the entire file format is proprietary.
P.S. The above is from what I was able to gather in about a day of research into what it would take to write my own jpeg library. If anyone notices anything in this that isn't correct, please point it out.
Ugh. I'm a native English speaker and I shudder to think of what life would be like if English became the universal language.
But then again, I've gone to a lot of effort to learn 3 other languages and I believe that many of the wonderful experiences I've had in other countries have been partly facilitated by the people I meet knowing that I've gone to a lot of effort to be able to try to experience their culture as more than just a casual tourist. Even in countries where I don't speak the language, it's amazing how much more people will accept you if you just take the simple step of learning a few phrases in their language. I'm always annoyed when other Americans talk about how people in other countries are rude to tourists when they've really been rude to them by having the arrogance to assume that their hosts should go the extra mile to learn English.
I'm not sure this is really a case I need to make, since I think most people can recognize the value of having so many languages.
As far as internationalized URLs, there is really no reason why we shouldn't be able to do it quite simply. So long as UTF-8 is used, all existing domains would need absolutely no change since I don't believe extended ASCII characters are used in domain names.
He also doesn't take into account that some exercises can actually be detrimental to depressive people. Exercises that cause the body to produce adrenaline can actually make someone's depression worse. When the body releases adrenaline it also releases cortisol. While the adrenaline levels subside shortly after you finish exercising, cortisol levels take far longer to normalize. While the causes of depression aren't fully understood, there have been studies that have shown a connection between cronically-elevated cortisol levels and depression.
So while exercise is generally one recommended activity for depressives as it can help create an environment where the other treatments can be more effective, you have to be careful which exercises you engage in. And, like you said, exercise isn't a magic bullet, it's only part of eliminating one of the obstacles to getting better.
The problem with IE5 for Mac is that it was released on a staggered development cycle from IE on the PC. This led to radically different functionality depending on the platform. IE for the Mac was a huge step forward (for IE at the time, that is) when it comes to standards compliance. But it doesn't help developers to comply to standards in one place and not the other. So when the next version of IE came out on Windows, it was in some ways more standards compliant than IE for Mac. But it was also somewhat less compliant in other areas. Had they chosen to do a straight port to the Mac--bugs, foibles and all--it would have been preferable. Even if it was less standards compliant than IE 5 for the Mac ended up being.
There's nothing magic about the standards that makes them better than any other browser target. The only reason developers are constantly clamoring for standards compliance is that is presents a single target to code to. We don't care which target we code to, we just don't want to have to support multiple targets. And that's why Microsoft's refusal to comply with standards is so frustrating. Every new IE release creates a new target to code for (though hopefully the release of IE7 will mean we can finally stop testing for IE 5.5).
This is why Hachamovitch's mention of needing to maintain compatability with other products at Microsoft is somewhat disingenuous. That compatability is only a problem because they've chosen to deviate from web standards both in IE and in the other products. Standards compliance can become a mechanism for allowing projects to work independently from each other without fear that they will need anything more than a couple of tweaks before launching to ensure that the two products work well together. But since they've chosen to deviate from standards (I won't go so far as to say that they've intentionally deviated, but it does seem that way at times), all those project teams need to work closely together to ensure that changes to IE don't break everyone else's code.
Not that I agree with removing any of the ancient Egyptian antiquities from Egypt, but there is a huge difference between removing them for display at a prominent museum like the British Museum and removing them to be sold on the black market such that they will likely never again be available to be seen by the public.
A good example of this is the mummy of Ramses I. If this had been pillaged by archaeologists on behalf of the British Museum, it would be in much better shape that it is currently. However, because it was unearthed by tomb robbers, it spent over 100 years at a museum in Niagra Falls with very little concern for maintaining it and absolutely no indication given to visitors that it was, in fact, the mummy of a Pharoah. An "expert" grave robber would have followed much a much more strict procedure to ensure that it was properly cared for and properly catalogued (if only to increase the value, but still).
That said, the Ramses I mummy did end up in Egypt, which almost makes up for the shoddy maintenance it received over the course of its post-excavation life (museums around the world should follow the example of the Carlos Museum at Emory University and return everything that was stolen from Egypt)...
According to the Oxford American Dictionary, you're wrong. They define intelligence as "The ability to acquire and apply knowledge and skills." (I don't have an OED on hand, but I'd imagine their definition is similar). By that definition, a person's musical talents would be considered a form of intelligence.
/. types are better able to understand the separation of these traits because we, at one point in our lives, determined these characteristics by rolling dice for each of them separately.
What the IQ tests measure is Intellect (defined as "The faculty of reasoning and understanding objectively, esp. with regard to abstract or academic matters"). Both words come from the same Latin root, but their meanings are subtly different.
But I understand where you're coming from...somehow I think
Whether the assertion of "production grade" is true or not is completely irrelevant when it comes time to sell the company. So long as you can hack it together well enough that you can scale it up to a number of users that will convince someone to buy you out, long-term sustainability of the code isn't necessary. If you can hit that performance metric, you can throw around the "production grade" terminology all you want.
Engineers too often forget that technical superiority often means very little in the short term. Time to market, marketing and sales often contribute far more towards short-term success than sound engineering does. Long term is an entirely different matter, but most of these Bubble 2.0 buyouts happen so quickly that there isn't time for the technical mistakes to become a problem. And it's pretty rare that the buying company is purchasing the smaller company for their actual technology, what they're really buying is the brand and userbase.
Take the case of the YouTube buyout for example...Google paid $1.6b for it and there's absolutely nothing there that Google couldn't rewrite in under a year in manner that's maintainable going forward. It might take 10-15 engineers on to do it, but that's roughly 3-4 million on an investment roughly 500 times that much.