[Tfug] Knoppix

[Anthony Steckman]

Well, I was really only incorrect to the extent that I tried to argue 
HTTP transfers were handled in either a binary or an ASCII mode and that 
the MIME type would effect this.

I'd still argue that any file type offered on a server should have a 
defined MIME type. I can think of at least three ways in which this is 
beneficial:

(1) If the end user's client is reporting a content type, the type 
reported to the end user is correct,

(2) If the end user's client offeres the ability to configure file 
handling behaviors (automatically opening certain file types within 
certain applications, for example), then the client will correctly 
identify the file type and process it accordingly, and

(3) It helps to prevent confusion.


John Gruenenfelder wrote:
> On Sun, Dec 14, 2003 at 06:51:40PM -0700, Anthony Steckman wrote:
> 
>>
>>Yeah, after Jeremy's last post I ran some tests using a smaller ISO file 
>>I created for testing purposes. I repeated the tests with a few other 
>>binary file types. It seems it doesn't matter if the browser or download 
>>manager thinks it's plain text or not, the file will still download 
>>succesfully.
> 
> 
> While it's not the case here, this problem *is* present with browsers in
> Windows.  Even Mozilla/Firebird will happily destroy a transfer if the remote
> end says it is text/plain.
> 
>