on how you are using each platform. The biggest problems I have seen deal with propietary features in the different Unices. For instance, I worked as a Solaris Admin using NIS+ and while it supported authenicating other Unices that could just use NIS, it don't work well. But that was a years ago.
Things that help include creating branchs in your login scripts (.profile or.cshrc) that set your preferences on the different boxes using uname.
There is a good O'Rielly book called "Unix for Oracle DBAs" that is a really good cross Unix reference that you should consider picking up.
The scripts you use to create and populate the database should go into CVS and then you can run them to create or recreate your database and data.
I have also used CVS to store a file that is a series of database alters or a series of files that are a bunch of db alters that are to applied together to form a kind of DB release or tag.
DDL and load scripts are just code, treat them as code.
Design Patterns don't speed up development or execution speed of a system. They are tool used to describe a solution to a well known problem. Using a pattern "correctly" will only provide effecency in describing a problem/solution to another person versed in tht pattern.
Patterns and anti-patterns are valuable, important tools for software engineering but please don't make them to be more than they are.
Sure there are a few really good people in India, sure they can compete with US developers, sure you can overcome the different country/time zone issue (to some degree). How long before these good Indian developers start wanting more money because the are "worth" it? No long. Business is about making money and with limited talent pools like good IT people, margin is the way to go not volume.
Why do you think the best of the Indian talent comes to the US? More money and a better life. How long with the best still in India talent work for $10/hour?
I have written web application using Perl and XSLT (and a few other technologies). The biggest issue with XSLT using XSL files as the templates and XML as the data (how else would you do it;-)) is that XSLT is meant to recurse over XML DOM objects translating the values of the DOM objects into another DOM (HTML in this case). Most web developers want (rightly so) to just use simple replacement variables and do string replacement in very limited places on a template.
Perl is great for doing this kind of template based approach, so is PHP and Cold Fusion and other programming environments. XSL files and the XSLT process are fairly heavy weight for a simple find and replace the 75% of web applications want. Originally XSLT was meant to be an off-line process, i.e. not on the fly, where the extra flexibility but heavy weight process was a plus, not a minus. But this idea of the right tool for the right job is lost on most people.
There are a number of companies that plan on coming out with Wireless IM/E-Mail/Web much like RIM does wireless, always e-mail and with VoiceStream you can get Wireless Always on AIM on your cell phone. The Hiptop is one thing that comes to mind. Soon you will be able to IM and E-Mail personal stuff at work all day long and it won't pass over your employer's Network all for $20-40 dollars a month.
You have a great point and there is a great deal of real world history that can play in to this. Such as the stock market, ponzi and pyrimid schemas.
In the end it will get down to reputation and playability. If a game marker tries to make games that they can profit from the secondary market in some way that people don't think it right, they will become less viable. This happens in the real world, like at restaurants that don't give customers a good experience, they don't come back and tell their friends the restaurant isn't any good.
As far as flooding the market, if players spend a great deal of time and money getting items and the company running the game floods the market, the game will loose customers.
I think blizard is taking the right stand, not promoting or demoting selling of characters or items. Their business is to sell games and get people to play. That is what they care about.
Come on, Linus is god. He is everything. Life won't be worth living without him. There was no Open Source before home and there will be no open source after him.
in the future. I have Borland C++ 3.1 for Windows 3.1 that I purchased 8 or 9 years ago. I haven't used it for 6 years or more. The same for Turbo Pascal version 3.0, 4.0 and 6.0. I can't even remember the last time I programmed in Pascal, maybe something like 10 years ago.
I don't see the rate of technology slowing and I don't see Universities giving up on Unix anytime soon. With that said, both MS tools and Unix as we know it today are going to be very different 10 years from now. So when your average MS indocratened 12-year actually becomes a programmer (at something like 22) ever MS tool or Unix tool they learned could be long gone.
come to think of it, 2004 is 2^30 seconds after 1970! so time_t may be a signed 31-bit or unsigned 30-bit type!
The General solution to this problem is to just turn the signed int into an unsigned int and extent the problem to 2068. Given that in 2000 the world is looking at making the 64-bit transition on the desktop in the next 5 years and at that time time_t will most likely be a signed 64-bit int, I don't think this will be a problem.
signed int (the base type for time_t is a signed int) is -2^31 to 2^31-1 to take zero into account.
an unsigned int is 0 to 2^32-1.
These number ranges are a by product of 32-bit CPU registers. 31-bit signed int and 30-bit unsigned don't make very much sense in a 32-bit CPU world.
Slashdot Mirror
on how you are using each platform. The biggest problems I have seen deal with propietary features in the different Unices. For instance, I worked as a Solaris Admin using NIS+ and while it supported authenicating other Unices that could just use NIS, it don't work well. But that was a years ago.
.cshrc) that set your preferences on the different boxes using uname.
Things that help include creating branchs in your login scripts (.profile or
There is a good O'Rielly book called "Unix for Oracle DBAs" that is a really good cross Unix reference that you should consider picking up.
The scripts you use to create and populate the database should go into CVS and then you can run them to create or recreate your database and data.
I have also used CVS to store a file that is a series of database alters or a series of files that are a bunch of db alters that are to applied together to form a kind of DB release or tag.
DDL and load scripts are just code, treat them as code.
Design Patterns don't speed up development or execution speed of a system. They are tool used to describe a solution to a well known problem. Using a pattern "correctly" will only provide effecency in describing a problem/solution to another person versed in tht pattern.
Patterns and anti-patterns are valuable, important tools for software engineering but please don't make them to be more than they are.
Sure there are a few really good people in India, sure they can compete with US developers, sure you can overcome the different country/time zone issue (to some degree). How long before these good Indian developers start wanting more money because the are "worth" it? No long. Business is about making money and with limited talent pools like good IT people, margin is the way to go not volume.
Why do you think the best of the Indian talent comes to the US? More money and a better life. How long with the best still in India talent work for $10/hour?
I have written web application using Perl and XSLT (and a few other technologies). The biggest issue with XSLT using XSL files as the templates and XML as the data (how else would you do it ;-)) is that XSLT is meant to recurse over XML DOM objects translating the values of the DOM objects into another DOM (HTML in this case). Most web developers want (rightly so) to just use simple replacement variables and do string replacement in very limited places on a template.
Perl is great for doing this kind of template based approach, so is PHP and Cold Fusion and other programming environments. XSL files and the XSLT process are fairly heavy weight for a simple find and replace the 75% of web applications want. Originally XSLT was meant to be an off-line process, i.e. not on the fly, where the extra flexibility but heavy weight process was a plus, not a minus. But this idea of the right tool for the right job is lost on most people.
There are a number of companies that plan on coming out with Wireless IM/E-Mail/Web much like RIM does wireless, always e-mail and with VoiceStream you can get Wireless Always on AIM on your cell phone. The Hiptop is one thing that comes to mind. Soon you will be able to IM and E-Mail personal stuff at work all day long and it won't pass over your employer's Network all for $20-40 dollars a month.
You have a great point and there is a great deal of real world history that can play in to this. Such as the stock market, ponzi and pyrimid schemas.
In the end it will get down to reputation and playability. If a game marker tries to make games that they can profit from the secondary market in some way that people don't think it right, they will become less viable. This happens in the real world, like at restaurants that don't give customers a good experience, they don't come back and tell their friends the restaurant isn't any good.
As far as flooding the market, if players spend a great deal of time and money getting items and the company running the game floods the market, the game will loose customers.
I think blizard is taking the right stand, not promoting or demoting selling of characters or items. Their business is to sell games and get people to play. That is what they care about.
Musicmaker.com is a failed company that dropped out. I am sure many of the other 27 a losers too.
I'm saying SDMI is good or bad but don't read too much into 27 out of business companies not be part of SDMI.
Come on, Linus is god. He is everything. Life won't be worth living without him. There was no Open Source before home and there will be no open source after him.
in the future. I have Borland C++ 3.1 for Windows 3.1 that I purchased 8 or 9 years ago. I haven't used it for 6 years or more. The same for Turbo Pascal version 3.0, 4.0 and 6.0. I can't even remember the last time I programmed in Pascal, maybe something like 10 years ago.
I don't see the rate of technology slowing and I don't see Universities giving up on Unix anytime soon. With that said, both MS tools and Unix as we know it today are going to be very different 10 years from now. So when your average MS indocratened 12-year actually becomes a programmer (at something like 22) ever MS tool or Unix tool they learned could be long gone.
come to think of it, 2004 is 2^30 seconds after 1970! so time_t may be a signed 31-bit or unsigned 30-bit type!
The General solution to this problem is to just turn the signed int into an unsigned int and extent the problem to 2068. Given that in 2000 the world is looking at making the 64-bit transition on the desktop in the next 5 years and at that time time_t will most likely be a signed 64-bit int, I don't think this will be a problem.
signed int (the base type for time_t is a signed int) is -2^31 to 2^31-1 to take zero into account.
an unsigned int is 0 to 2^32-1.
These number ranges are a by product of 32-bit CPU registers. 31-bit signed int and 30-bit unsigned don't make very much sense in a 32-bit CPU world.