Re: Revised proposal for UTF-16

[Dan Kegel <dank@alumni.caltech.edu>]

At 09:10 AM 5/25/98 +0200, Harald Alvestrand wrote:
>I think it's wrong to say anything about how people who insist on
>doing the Wrong Thing should behave; I was trying to craft language
>that said what the Right Thing is, and hint at how to tackle the
>Wrong Thing.
Yes.

>What about adding:
>
>"Note: There is no way to 100% reliably detect little-endian data that does
>not use the BOM."
Good.  Still, there has the whiff of 'dark secret' about it that
might be confusing to newcomers.

>Just telling people who try to deal with the mess that they are getting
>ever deeper into a mess.....
No, I think there's a clear way - somehow - to write this.

You know, they say that if it's too hard to document, maybe
there's something wrong with it.

Perhaps the key is to ALWAYS send a BOM.  
Then the language becomes exceedingly clear and simple.
- Dan