Maybe it's the only book worth reading for an experienced SQL programmer, but SQL for Smarties is not for SQL beginners. These are the first words of the Introduction:
This book, like the first edition, is for the working SQL programmer who wants to pick up some advanced programming tips and techniques. It assumes that the reader is an SQL programmer with a year of actual experience.
I don't think the incidence of birth defects is caused by heritable defects in the parents' genes. Rather, it seems that the IVF procedure can damage an embryo.
My wife and I have noticed that many of the IVF children of our friends have some kind of congenital problem, often a subtle developmental deficiency, such as a mild learning disability. The problems were not inherited from the parents. Maybe someday the children will inherit their parents' infertility, but that's not the kind of problem we've noticed.
Science still has a very incomplete understanding of the processes of embryonic development. The variations between animals and their cloned offspring show that the development of specific characteristics depends not just on the genetic "blueprint" -- which is actually more like a bill of (protein) material than a detailed design of the end-product organism -- but also very heavily on the whole construction process, and particularly on the environment in which it takes place. I'm not convinced that early embryonic development in a Petri dish is the same as development in a uterus, or even that a uterus in which an embryo is implanted is the same as a uterus in which an egg was fertilized. Some of the chemistry may be different, so some of the development may be different.
Unfortunately, I haven't yet seen research on specific abnormalities or developmental problems of IVF babies. This must be an extremely sensitive subject for the infertility industry.
Of course, even if IVF is riskier for children than natural conception, most IVF parents would probably prefer a higher-risk pregnancy over no pregnancy at all. But we need more research so those parents can base their decisions on fuller information.
You have it exactly wrong. Hilary showed the Indians that they should off-shore their own jobs to Buffalo! The departure of Buffalo's manufacturing base did bring wages down to near-Third World levels a couple decades ago, but that can be a competitive advantage for the City of No Illusions. Buffalo may yet become the American Bangalore!
This trend has nothing to do with free software. Expensive, proprietary software is good for its owners, not its developers.
The off-shoring trend is all about the big companies discovering the global infrastructure and a large enough talent pool in the third world to allow cost savings by moving coding jobs. Didn't you notice that Microsoft is one of the companies that is off-shoring development? Listen to the audiotape linked from the NYT article. The IBM HR guy specifically cites the competitive challenge of Microsoft's directive that every one of Microsoft's departments should off-shore some project.
India will soon be full of.Net coders. Your future is not safe with Microsoft, or HP, or IBM. Your future, if it's in IT, may require you to stay clear of the multinationals and in fact try to compete against them, or around them, with nimbleness and with the advantages of the U.S. infrastructure. And one of those advantages is access to free software that can serve as a robust, but low-cost, platform for your development.
People get the same way when they're cooped up and take up all sorts of repetitive psychotic behaviors. It's a self-protection method for the brain I believe, keeping itself occupied in some endless task rather than concentrating on its continuously uninteresting environment and going crazy.
The demos mostly show the figures being hit, blasted, or shot. Maybe that's to appeal to game makers, but it suggests that the figures still have a rather limited repertoire. I'd like to see more gymnastics and other complex body movement, not just the physics model of a jointed body crumpling.
Check out The Teaching Company. They have a nice catalog of high-quality college lecture courses available on cassette, CD, VHS, or DVD. Their selection leans toward the humanities, but they offer a little of everything. I know people who are addicted to these courses.
I used to think the process described in the article was the way to build a home network, and that's why I didn't have a home network. Like a big LEGO sculpture, it's cool to look at, and some geeks will make a hobby of it, but it's not a project for most people who just want a useful end-product.
Now there's WiFi, and even the cheap 802.11b hardware is fine for sharing files, printers, and broadband. Buy a USB adapter for each remote computer and you don't even need to open the cases. You can have everyone connected in an hour. Now THAT is practical home networking.
Re:They pulled MySQL out!
on
PHP 5 Beta 1
·
· Score: 1
This is a big deal, but (1) it appears to be an effort to light a fire under MySQL's licensing staff and (2) losing built-in MySQL support would not be all bad. It might cause PHP developers to give a little thought to their websites' back ends. This little SQLite database that is going into PHP 5 actually looks pretty neat. If a developer needs a more powerful back end, let him evaluate MySQL along with other alternatives. MySQL is increasingly profit-driven, while SQLite is good old-fashioned public domain open source, and deserves to be PHP's default database.
Re: Just use PEAR/DB
on
PHP 5 Beta 1
·
· Score: 4, Informative
Instead of ODBC, you'd be better off using the pear/db module as middleware. It supports more databases (mysql, odbc, sqlite, pgsql, etc.) and if it isn't the future standard for database access in PHP, something like it will be.
I've been using PHP's built-in (until now) MySQL functions, because they're faster than pear/db, but this licensing dust-up has convinced me that portability among database vendors is worth a performance hit. And the pear/db module is getting increasing attention and is likely to get faster.
Err, now that I've RTFA'd, I see that SCO has asked for a permanent injunction rather than a preliminary injunction, so they don't need to show their cards yet, and there will be no impact on IBM or its customers until the case is over. SCO's failure to ask for a preliminary injunction looks like uncertainty about its case, but I suppose it shows enough intelligence to choose a slow method of suicide over a quick method.
The court will only grant SCO's injunction if SCO shows a likelihood that it will ultimately prevail on the merits of its complaint. This question gives IBM its first opportunity to fire its legal guns, which IBM has been putting into position for months. Good-bye, SCO.
I've been using WordPerfect since 1986, and Paradox since 1991, so I'm a customer. But bugs haven't been getting fixed in the last few years. The products are pretty stable as they are, but there are well-known bugs that keep showing up in every new version.
A few weeks ago I needed a spreadsheet with a formula in 1-2 million cells. Since Quattro Pro was on my machine, I fired it up, defined my rows and columns, created my formula, gave the copy command, and Quattro Pro crashed. I checked the docs for any limitation I'd exceeded, but couldn't find anything. I tried copying one column at a time, and that worked for a few columns, then it crashed again. It just didn't work.
So I downloaded Open Office, installed it, and made my spreadsheet. No problem.
If Corel can't get its products up to the quality standards of open source software, it has no business charging money for its products.
Almost. The capitol is the building where the legislature meets, named after the Capitol in Washington, which was named after the Capitol (a temple to Jupiter) in ancient Rome. The city where the government is based is the capital city.
The author of the Inquirer article wrote an earlier article three weeks ago about what a jerk Marc Fleury, the head of JBoss, was. There was also an article a few days ago in Open Enterprise trends, about new profit-sharing and stock-purchase plans at JBoss, which might have been either a cause or an effect of dissension there.
This "open data format" proposal is impractical and counter-productive.
Impractical
Most computer programs are useful because they produce human-readable output, usually by arranging pixels on a screen or on a sheet of paper in ways that correspond with natural-language representations (words and numbers). Every purchaser of software is free to decide before purchase whether a program's output is sufficient for his needs, including any potential need for data archiving or transfer.
But until the data is output, it is encoded for the convenience of the program, not for the user. Even an ASCII file is not really "text," but numeric bytes whose values correspond to alphanumeric characters only through an artificial, though widely adopted, convention.
A data file is just a data structure stored in a persistent medium. As a developer, I shudder to imagine the burden of documenting every bit of data my programs store. I'd need to convert them from binary format to human readable. Not only that, but binary formats are necessary for some functionality. Have you ever benchmarked the performance of a large conventional database against a stored-as-XML database?
I guess I couldn't compress data, or maybe I could only compress it in a "standard" format - but which one, and who decides which one? Graphics and sound? I suppose I couldn't use a proprietary format for those either. And I guess I couldn't encrypt any data, at least not in any way a competing program couldn't decrypt. Well, if my competion can decrypt the data, so can anyone else. Sorry, but this is a non-starter for many business applications.
Counter-productive
Currently, a user has his software, he has his data (stored in whatever format the software's author found efficient and secure enough for the application), and he has a legal right to make copies of them so he won't lose any data he has bargained for. So this law doesn't really gain him data security; it gains interoperability, the possibility that competing software vendors will be able to serve his needs. There's an implicit hope that small software vendors could use open data formats to compete more equally against large vendors.
Alas, that hope is a pipe dream. Because any software vendor would be able to extend a published data format, the publisher who could extend it most and fastest would always lead the market. That means the biggest company would lead the market.
It's happened before. Microsoft's president is on record, telling his developers to "embrace and extend" industry standards. That's how Microsoft maintains de facto leadership of many standards. They can embrace whatever is open to be embraced, and extend it so it becomes Microsoft's. Examples include HTML (Microsoft's progress slowed by developer inertia), web services (Microsoft looking strong), Java (thwarted, so far, by Sun's license), and even ".doc" files (we old-timers remember when they were plain ASCII).
An open source author has the protections of copyright and licensing terms against Microsoft stealing his code. But the author of a program that keeps its data open has no such protection. The data belongs to the program's user, not the program's author. Open data is ripe for the picking by a bigger company, and a small-scale author of successful open-data software will soon be following instead of leading in his market.
Let's leave software authors and users free to decide whether human-readable output is enough for them, or if they want data structures stored persistently in human-readable form. The perversion of the patent system should teach us this much: laws that are meant to help the little guy, may help the big guy even more.
Don't be so sure you'll never get to take that course. MIT - the birthplace of GNU, after all - is leading the academic world in its "OpenCourseWare" initiative. The syllabi, lecture notes, assignments, labs and exams for some courses have already been published on the web, and MIT plans to publish almost all its course materials by 2007.
Whether mankind will survive long enough to achieve this kind of technology is not the right question. The question is whether our technological civilization will survive long enough. Even species survival is in some doubt, but taking that for granted, it still seems certain from history that civilizations do not last more than a few hundred years. Typically, they deplete the resources they thrived on, and internal disorder makes them vulnerable to external assaults, which cause further internal disorder, and a spiral down into poverty and ignorance. And the extreme specialization of skills, on which our economy and especially our technology rely, is itself a vulnerability.
Slashdot Mirror
I don't think the incidence of birth defects is caused by heritable defects in the parents' genes. Rather, it seems that the IVF procedure can damage an embryo.
My wife and I have noticed that many of the IVF children of our friends have some kind of congenital problem, often a subtle developmental deficiency, such as a mild learning disability. The problems were not inherited from the parents. Maybe someday the children will inherit their parents' infertility, but that's not the kind of problem we've noticed.
Science still has a very incomplete understanding of the processes of embryonic development. The variations between animals and their cloned offspring show that the development of specific characteristics depends not just on the genetic "blueprint" -- which is actually more like a bill of (protein) material than a detailed design of the end-product organism -- but also very heavily on the whole construction process, and particularly on the environment in which it takes place. I'm not convinced that early embryonic development in a Petri dish is the same as development in a uterus, or even that a uterus in which an embryo is implanted is the same as a uterus in which an egg was fertilized. Some of the chemistry may be different, so some of the development may be different.
Unfortunately, I haven't yet seen research on specific abnormalities or developmental problems of IVF babies. This must be an extremely sensitive subject for the infertility industry.
Of course, even if IVF is riskier for children than natural conception, most IVF parents would probably prefer a higher-risk pregnancy over no pregnancy at all. But we need more research so those parents can base their decisions on fuller information.
You have it exactly wrong. Hilary showed the Indians that they should off-shore their own jobs to Buffalo! The departure of Buffalo's manufacturing base did bring wages down to near-Third World levels a couple decades ago, but that can be a competitive advantage for the City of No Illusions. Buffalo may yet become the American Bangalore!
This trend has nothing to do with free software. Expensive, proprietary software is good for its owners, not its developers.
.Net coders. Your future is not safe with Microsoft, or HP, or IBM. Your future, if it's in IT, may require you to stay clear of the multinationals and in fact try to compete against them, or around them, with nimbleness and with the advantages of the U.S. infrastructure. And one of those advantages is access to free software that can serve as a robust, but low-cost, platform for your development.
The off-shoring trend is all about the big companies discovering the global infrastructure and a large enough talent pool in the third world to allow cost savings by moving coding jobs. Didn't you notice that Microsoft is one of the companies that is off-shoring development? Listen to the audiotape linked from the NYT article. The IBM HR guy specifically cites the competitive challenge of Microsoft's directive that every one of Microsoft's departments should off-shore some project.
India will soon be full of
The demos mostly show the figures being hit, blasted, or shot. Maybe that's to appeal to game makers, but it suggests that the figures still have a rather limited repertoire. I'd like to see more gymnastics and other complex body movement, not just the physics model of a jointed body crumpling.
They're replacing mainframes, not Windows servers.
1. Map important stuff.
2. Indulge paranoid fantasies of security-industrial complex.
3. Sell them "exclusive access" to your map.
4. Profit!!!
Check out The Teaching Company. They have a nice catalog of high-quality college lecture courses available on cassette, CD, VHS, or DVD. Their selection leans toward the humanities, but they offer a little of everything. I know people who are addicted to these courses.
I used to think the process described in the article was the way to build a home network, and that's why I didn't have a home network. Like a big LEGO sculpture, it's cool to look at, and some geeks will make a hobby of it, but it's not a project for most people who just want a useful end-product.
Now there's WiFi, and even the cheap 802.11b hardware is fine for sharing files, printers, and broadband. Buy a USB adapter for each remote computer and you don't even need to open the cases. You can have everyone connected in an hour. Now THAT is practical home networking.
This is a big deal, but (1) it appears to be an effort to light a fire under MySQL's licensing staff and (2) losing built-in MySQL support would not be all bad. It might cause PHP developers to give a little thought to their websites' back ends. This little SQLite database that is going into PHP 5 actually looks pretty neat. If a developer needs a more powerful back end, let him evaluate MySQL along with other alternatives. MySQL is increasingly profit-driven, while SQLite is good old-fashioned public domain open source, and deserves to be PHP's default database.
Instead of ODBC, you'd be better off using the pear/db module as middleware. It supports more databases (mysql, odbc, sqlite, pgsql, etc.) and if it isn't the future standard for database access in PHP, something like it will be.
I've been using PHP's built-in (until now) MySQL functions, because they're faster than pear/db, but this licensing dust-up has convinced me that portability among database vendors is worth a performance hit. And the pear/db module is getting increasing attention and is likely to get faster.
Orrin will have a rude awakening from his corporatist wet dream when the Linux kernel hackers start taking down SCO's computers.
Err, now that I've RTFA'd, I see that SCO has asked for a permanent injunction rather than a preliminary injunction, so they don't need to show their cards yet, and there will be no impact on IBM or its customers until the case is over. SCO's failure to ask for a preliminary injunction looks like uncertainty about its case, but I suppose it shows enough intelligence to choose a slow method of suicide over a quick method.
The court will only grant SCO's injunction if SCO shows a likelihood that it will ultimately prevail on the merits of its complaint. This question gives IBM its first opportunity to fire its legal guns, which IBM has been putting into position for months. Good-bye, SCO.
I've been using WordPerfect since 1986, and Paradox since 1991, so I'm a customer. But bugs haven't been getting fixed in the last few years. The products are pretty stable as they are, but there are well-known bugs that keep showing up in every new version.
A few weeks ago I needed a spreadsheet with a formula in 1-2 million cells. Since Quattro Pro was on my machine, I fired it up, defined my rows and columns, created my formula, gave the copy command, and Quattro Pro crashed. I checked the docs for any limitation I'd exceeded, but couldn't find anything. I tried copying one column at a time, and that worked for a few columns, then it crashed again. It just didn't work.
So I downloaded Open Office, installed it, and made my spreadsheet. No problem.
If Corel can't get its products up to the quality standards of open source software, it has no business charging money for its products.
The author of the Inquirer article wrote an earlier article three weeks ago about what a jerk Marc Fleury, the head of JBoss, was. There was also an article a few days ago in Open Enterprise trends, about new profit-sharing and stock-purchase plans at JBoss, which might have been either a cause or an effect of dissension there.
This "open data format" proposal is impractical and counter-productive.
Impractical
Most computer programs are useful because they produce human-readable output, usually by arranging pixels on a screen or on a sheet of paper in ways that correspond with natural-language representations (words and numbers). Every purchaser of software is free to decide before purchase whether a program's output is sufficient for his needs, including any potential need for data archiving or transfer.
But until the data is output, it is encoded for the convenience of the program, not for the user. Even an ASCII file is not really "text," but numeric bytes whose values correspond to alphanumeric characters only through an artificial, though widely adopted, convention.
A data file is just a data structure stored in a persistent medium. As a developer, I shudder to imagine the burden of documenting every bit of data my programs store. I'd need to convert them from binary format to human readable. Not only that, but binary formats are necessary for some functionality. Have you ever benchmarked the performance of a large conventional database against a stored-as-XML database?
I guess I couldn't compress data, or maybe I could only compress it in a "standard" format - but which one, and who decides which one? Graphics and sound? I suppose I couldn't use a proprietary format for those either. And I guess I couldn't encrypt any data, at least not in any way a competing program couldn't decrypt. Well, if my competion can decrypt the data, so can anyone else. Sorry, but this is a non-starter for many business applications.
Counter-productive
Currently, a user has his software, he has his data (stored in whatever format the software's author found efficient and secure enough for the application), and he has a legal right to make copies of them so he won't lose any data he has bargained for. So this law doesn't really gain him data security; it gains interoperability, the possibility that competing software vendors will be able to serve his needs. There's an implicit hope that small software vendors could use open data formats to compete more equally against large vendors.
Alas, that hope is a pipe dream. Because any software vendor would be able to extend a published data format, the publisher who could extend it most and fastest would always lead the market. That means the biggest company would lead the market.
It's happened before. Microsoft's president is on record, telling his developers to "embrace and extend" industry standards. That's how Microsoft maintains de facto leadership of many standards. They can embrace whatever is open to be embraced, and extend it so it becomes Microsoft's. Examples include HTML (Microsoft's progress slowed by developer inertia), web services (Microsoft looking strong), Java (thwarted, so far, by Sun's license), and even ".doc" files (we old-timers remember when they were plain ASCII).
An open source author has the protections of copyright and licensing terms against Microsoft stealing his code. But the author of a program that keeps its data open has no such protection. The data belongs to the program's user, not the program's author. Open data is ripe for the picking by a bigger company, and a small-scale author of successful open-data software will soon be following instead of leading in his market.
Let's leave software authors and users free to decide whether human-readable output is enough for them, or if they want data structures stored persistently in human-readable form. The perversion of the patent system should teach us this much: laws that are meant to help the little guy, may help the big guy even more.
Don't be so sure you'll never get to take that course. MIT - the birthplace of GNU, after all - is leading the academic world in its "OpenCourseWare" initiative. The syllabi, lecture notes, assignments, labs and exams for some courses have already been published on the web, and MIT plans to publish almost all its course materials by 2007.
Whether mankind will survive long enough to achieve this kind of technology is not the right question. The question is whether our technological civilization will survive long enough. Even species survival is in some doubt, but taking that for granted, it still seems certain from history that civilizations do not last more than a few hundred years. Typically, they deplete the resources they thrived on, and internal disorder makes them vulnerable to external assaults, which cause further internal disorder, and a spiral down into poverty and ignorance. And the extreme specialization of skills, on which our economy and especially our technology rely, is itself a vulnerability.