Actually, that’s not true. For many software developers, it is an ASCII world. But, if you’re making software for other developers to use, please keep character encodings in mind. I’ve had to jump through hoops with many packages along the way to get them using UTF-8.
Luckily, when you use open source libraries, the lack of non-ASCII support is usually not a big problem, because it’s easy enough to add the support. Not so with a closed-source third party library. Though, anyone offering a commercial library would have to deal with encodings once they start offering the product internationally.
Entries (RSS)
February 4th, 2005 at 8:57 am
My current work project is a Java application that must run on mainframes and support sundry EBCDIC variants. Another way to be non-ASCII…
February 4th, 2005 at 9:06 am
It’s been quite a while since I’ve seen any mention of EBCDIC. Yeah, I guess those mainframes are still out there, aren’t they?
In truth, the “world” hasn’t really been ASCII for a while. Internationalized GUIs have had to do something else for some time. I am thankful for the effort of the Unicode people to try and pull everything under one umbrella.
July 30th, 2005 at 4:42 am
What is particularly dangerous is that most APIs (e.g., .NET and Java) don’t force the caller to explicitly define which encoding is to be applied. So some people won’t consider the consequences of using the default-encoding, while others might not even know what encodings are all about.
Default-Encoding Considered Harmful: http://arnosoftwaredev.blogspot.com/2005/03/default-encoding-considered-harmful.html