[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [debian-sparc] Which is best distro for Ultra 10



On Wed, Jun 04, 2003 at 01:28:23PM +0100, Martin wrote:
> > > > Do you mean to tell me that XML is _not_ taking over the world?
> > > In some applications.  I'm more interested in lower level software.
> > > There is no XML version of machine code or IP - and something tells me
> > > there never will be :-)
> > Would RFC3252 change your mind?
> I stand corrected *bows head to your supirior knowledge*
> 
> BUT
> 
> I think this RFC is aimed in completely the wrong direction.  A neat and
> standards compliant form of notation - I can see the use.  Althought I
> would say that tcpdump / ethereal do it quite well already.  Perhaps as
> a way of storing record of TCP/IP... then maybe... XML is good for
> storing structured data so... perhaps.

IP over XML is good for specifying NAT rules in xsl ;-)
I think this RFC is mainly aimed at its release date 

Frank

> 
> But not as an actual working protocol as suggested by this RFC.  A 20 to
> 30 fold increase in bandwidth useage for the same degree of service,
> having to replace all switching and routing equipment - it's just not
> practical.
> 
> Plus I don't see that any of the problems that it claims to address are
> actually problems:
> "dependencies on proprietary and hard-to-understand binary protocols" -
> IMHO TCP/IP is neither,
> "eliminating developer time spent figuring out the intricacies of each
> new protocol" - how many people routinely write software for an average
> machine connected to a network that actually needs to play with TCP/IP
> at this level?
> "mitigates concerns over "network vs. host" byte ordering which is at
> the root of many network application bugs." - it's simple enough really,
> use htonl, etc. or Sun RPC or ... this is not the size of problem that
> needs this size of solution - it's not really that big a problem at all
> - there is already a standard for this sort of thing.
> 
> Finally "Security considerations that apply to IP, TCP and UDP also
> likely apply to BLOAT as it does not attempt to correct for issues not
> related to message format." - IE use IPv6
> 
> My initial example was wrong but thank you for providing a fine example
> of why XML is not suitable for all tasks - even in an ideal world.
> 
> Sorry about being completely OT.  I again suggest anyone wanting to
> continue this joins me in doing so off list.
> 
> Sweet Dreams,
>  - Martin
>  
> -- 
> Martin
> inkubus@interalpha.co.uk
> "Seasons change, things come to pass"
> 
> 
> -- 
> To UNSUBSCRIBE, email to debian-sparc-request@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org



Reply to: