Board index » delphi » Re: Next Generation Web Protocol

Re: Next Generation Web Protocol


2004-07-29 03:03:58 PM
delphi86
"Daniel Becroft" writes
Quote
Erik Springelkamp writes:
>
>From the HTTP/1.1 RFC:
>www.w3.org/Protocols/rfc2616/rfc2616.html
>
>HTTP communication usually takes place over TCP/IP connections. The
>default port is TCP 80 [19], but other ports can be used. This does not
>preclude HTTP from being implemented on top of any other protocol on
>the Internet, or on other networks. HTTP only presumes a reliable
>transport; any protocol that provides such guarantees can be used;

And that, in no way, contradicts what I have said. That paragraph states
how HTTP is used in
relation to TCP connections. If you wanted, you could use HTTP over a raw
IP connection, or a UDP
connection. However, that does not eliminate the fact that HTTP is part of
the TCP/IP Protocol
Suite.
Wel, dear Daniel, you are wrong again. It clearly states
it needs reliable transport, and UDP is not reliable.
Please, do your homework.
Let's say, write simple netbios web server during weekend.
Quote

www.protocols.com/pbook/tcpip1.htm

That website lists the protocols included in the TCP/IP Protocol suite,
grouped in the 7 layers of
the OSI model.

And please, don't mention OSI model is serious discussion.
It is one of biggest disasters in history, ran over by running
TCP/IP code so many times that by now it is a dead meet.
 
 

Re: Next Generation Web Protocol

Hrvoje Brozovic writes:
Quote
"Daniel Becroft" writes
>Erik Springelkamp writes:
>>
>>From the HTTP/1.1 RFC:
>>www.w3.org/Protocols/rfc2616/rfc2616.html
>>
>>HTTP communication usually takes place over TCP/IP connections. The
>>default port is TCP 80 [19], but other ports can be used. This does not
>>preclude HTTP from being implemented on top of any other protocol on
>>the Internet, or on other networks. HTTP only presumes a reliable
>>transport; any protocol that provides such guarantees can be used;
>
>And that, in no way, contradicts what I have said. That paragraph states
how HTTP is used in
>relation to TCP connections. If you wanted, you could use HTTP over a raw
IP connection, or a UDP
>connection. However, that does not eliminate the fact that HTTP is part of
the TCP/IP Protocol
>Suite.

Wel, dear Daniel, you are wrong again. It clearly states
it needs reliable transport, and UDP is not reliable.
I stand corrected. However, my statement still stands: HTTP is part of the TCP/IP Protocol Suite.
Quote
>www.protocols.com/pbook/tcpip1.htm
>
>That website lists the protocols included in the TCP/IP Protocol suite,
grouped in the 7 layers of
>the OSI model.
>

And please, don't mention OSI model is serious discussion.
It is one of biggest disasters in history, ran over by running
TCP/IP code so many times that by now it is a dead meet.
It may be, but it is still useful to understanding data communications, and how the TCP/IP package
suite. But that is besides the point. I was merely citing another source that says HTTP is a member
of the TCP/IP Protocol suite.
--
Daniel Becroft
; =================================
"Real computer scientists don't comment their code. The identifiers are so long they can not afford
the disk space."
"Blue sparks and white smoke, the two most expensive components of any electrical system, and once
used up will cost a fortune to replace."
 

Re: Next Generation Web Protocol

Maybe my previos post was to hash,
but you took it gracefully. Thanks for that.
As for your primary quastion,
what do you find lacking in this portocols.
Biggest problem is running out of address space,
and IPv6 is slow comming, much slower than
expected, but in the end, it will come.
As for the rest, SMTP is most troubled with
problems directly connected with its age, but
it is refreshed not so long ago, and modern
implementations take common problems into
account.
As for other application level protocols,
what is you find wrong with them?
Except that they are no so brand new,
as .NET applications, you alos mentioned, are?
 

Re: Next Generation Web Protocol

Hrvoje Brozovic writes:
<snipped rest />
Quote
As for other application level protocols,
what is you find wrong with them?
It isn't that I find anything 'wrong' with them, I was just after other people's thoughts and
opinions as to whether they would ever be completely replaced in the near future.
<snipped rest />
--
Daniel Becroft
; =================================
"Real computer scientists don't comment their code. The identifiers are so long they can not afford
the disk space."
"Blue sparks and white smoke, the two most expensive components of any electrical system, and once
used up will cost a fortune to replace."
 

Re: Next Generation Web Protocol

"Daniel Becroft" writes
Quote

It isn't that I find anything 'wrong' with them, I was just after other
people's thoughts and
opinions as to whether they would ever be completely replaced in the near
future.
Some of they may be clumsy, but none is broken.
Their age and wide acceptance is big plus.
There are some akward stuff nowerdays,
since relative cost of computer storage, memory and CPU
and network bandwith and latency is different than it was
in a time of their invention, but clever modern implementation
can cope with it.
MS would like if internet was full of their proprietary
protocols, but even their give up that idea more or less.