From: Brad Templeton (brad@templetons.com)
Date: Wed Jan 02 2002 - 13:38:27 CST
On Wed, Jan 02, 2002 at 12:23:47PM -0500, Bill Davidsen wrote:
> On Mon, 31 Dec 2001, Brad Templeton wrote:
>
> I think trying to get rid of X- headers at this point would be a waste of
> time and effort. They are existing practice, and I see absolutely no
> benefit from taking the "we could have done that better" approach.
>
> The benefit is, well, if you see some benefit from trying to change the
> way things are done, let me know. The cost is that people will just make
> up header names using no rules at all, won't understand registry, will
> have no quick way to add a local header, etc.
>
> In other words the horse is out of the barn, let's concentrate on things
> with a better cost/benefit ratio.
The problem is that right now some people put an X- on their new headers
and some people don't. It's relatively balanced. The standard says X-
and many people correctly ignore it, because it's a bad idea. When people
are doing things two ways, the standard should answer it.
For now, I would just remove the X- langauge, and worry about the registry
later. It can work remarkably well even without a registry. Defining a new
experimental header is relatively rare.