Re: gzip/deflate compression/encoding

On Jun 28 2005, Harald Tveit Alvestrand wrote:
> 
> I'd very much like to see the deflate-8bit and deflate-base64 C-T-Es 
> formally defined.

Does it have to be a content encoding, or would a small number of
media types serve the same purpose, ie to try the scheme out?

> 
> If someone identifies a niche for which they can be introduced sensibly, 
> they will be able to deploy them, rather than inventing Yet Another 
> Compression Scheme. If nobody starts using them seriously, well.... we've 
> lost some time, and an RFC number or two. Big deal. We've done that any 
> number of times, for any number of things (remember the MODEL top level 
> type? Who's hurt by having it around?)

I think I can guarantee that, whether it's useful or not, it will be used
by spammers to hide their junk. That means mail software will have to handle 
it just on the off chance it's legitimate, and pretty soon all
mail software will implement it, whether it has a useful place or not. 

Now if it's a media type rather than a transfer encoding, it can be safely
ignored if it doesn't find a use.

-- 
Laird Breyer.