langreiter.com plain, simple: comments-2003-10-13

LANGREITER.COM plain, simple

START

Montag, 13. Oktober 2003

Joel Spolsky: The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!)

www.joelonsoftware.com/articles/Unicode.html

Certainly the best introduction to that semi-mess I've ever come across.

2003-10-13-uniknow

andi 7695 days AGO
i guess joel never did more with unicode-data than displaying it. otherwise i don't know why he thinks this is sooo easy.. especially not if you work with non-english languages.

chris 7695 days AGO
can you expand some more?

btw, I think there's a reason he included "absolute minimum" in the title ;-)

hns 7694 days AGO
Or you could just use Java.

rainman tsr 7694 days AGO
"I should warn you that if you are one of those rare people who knows about internationalization, you are going to find my entire discussion a little bit oversimplified."

pointless...

David Ness 7694 days AGO
Joel's piece has good and bad features. I suppose the simple-minded explanation may be necessary---and mildly informative---but he surely brushes under the rug most of the complexity of encoding issues. Things like ligatures represent complexities in encoding that arise well _before_ we have even started to concern ourselves with the even greater complexity of internationalization. These issues have considerable depth. They are decidedly _not_ simple. Thus, on balance, I find his piece more misleading than helpful.

chris 7694 days AGO
a kind of roadmap to more encompassing enlightenment at the end of that introductory article wouldn't be a bad idea. david, do you know of any resources that go beyond joel's article, while at the same time being similarly readable? or is the only option to get the tome?

David Ness 7694 days AGO
I'll see if I can think of one. For example, TeX's need for virtual fonts suggests some of the depth of the problem. My principal argument was with Joel's (in my view incorrect) idea that ultimately this is pretty simple. I think that's wrong. The TeX people have been grappling with the issue unsatisfactorily (again in my view) for more than 20 years (since TeX `82 at least) and while there's only one Knuth, this means that many pretty good minds have been hitting their heads on the wall over this one for decades and still there are important problems to solve.

andi 7693 days AGO
sorry for replying late, i'm currently on holiday.
i thought of things like this: blogs.gotdotnet.com/raymondc/PermaLink.aspx/3cfdaedf-2060-4697-b7bf-19205d8448aa .
also i remember jeffrey zeldman using one of the miscellaneus symbols (www.unicode.org/charts/PDF/U2600.pdf) on his website which crashed some browsers while others just would not show the correct one.
personally i find it quite hard to work with unicode data as not all of my favourite development tools support it.

manuel 7692 days AGO
Ich finde die hier ganz gut:

UTF-8 and Unicode FAQ for Unix/Linux
http://www.cl.cam.ac.uk/~mgk25/unicode.html

Hello World...
http://www.cs.bell-labs.com/sys/doc/utf.html

Please log in (you may want to register first) to post comments!