[e2e] RFI: Microsoft accused of TCP standards violation

Jianping Pan jpan at fla.fujitsu.com
Mon Jan 6 18:11:50 PST 2003

On Monday 06 January 2003 05:49 pm, Neil Carpenter wrote:
> <..snip...>
> > Initially we thought it might be a bug in Netscape. Now
> > seems that it was *designed* to compete Microsoft's IE
> > with Microsoft's IIS.
> Can you back up that assertion with references or is that simply more
> disinformation?

Seems NO to me now since I can not find tcpdumps show IE
can get benefit by doing this and there is no point for 
Netscape to compete something might not exist.

Any tcpdump files between IE/IIS are highly appreciated.

> This thread started with a significant piece of disinformation (there is
> nothing in 'violation' of RFC's by sending data on a half-open connection,

This might be out of the end2end scope. In the tcpdump we
collected, the browser claimed itself "HTTP/1.0" and the 
server closed its outgoing data flow after the first 
request. There is no reason for the client to send the 
second request through the half-close TCP connection, 
especially it anticipates the reply data from the server.

> nor is there any presentable evidence that IIS responds to these).  Let's
> stick to the realm of the reasonable rather than making more unsupportable
> assertions.

Thank for the reminding.

Best regards,


> Neil

> All opinions presented are my own and not those of my employer.

The same claims apply to my messages too.

More information about the end2end-interest mailing list