[asterisk-biz] Experimental/new VoIP rate search engine.

[Trixter aka Bret McDanel]

On Mon, 2009-01-05 at 09:11 -0500, Kristian Kielhofner wrote:
>  How many pieces of equipment can't implement the core SIP
> RFCs properly?  

most? :)  In all fairness there are different versions of SIP so you
have to ensure that everyone is referring to the same one, but on top of
that the specs are not well written and are vague in some areas.
RFC2833/4733 are even worse (yeah yeah rfc1889 component, bleh).  But
yeah there are a lot of quirks that people do and it causes some
compatibility problems.

Many vendors also insist their interpretation is the valid one and will
not work to support anything additional.  

As for sdp wanna see breakage?  Look up what RFC2833 requires for codecs
of different sample rates.  Lets say you have a 8k and a 16k codec
offered, you have to offer 2 different RFC2833 payloads, because the
2833 rate must match the sample rate.  And since you can flip back and
forth between anything offered at any time that makes it even more
complicated.  Some vendors cheat and offer only one telephone-event
payload with a 8k sample rate, and when its in 16k mode they cheat (lie
in essence), and you have to know that they deviated and how so you can
interop with it.  This is almost like no one thought of multiple sample
rates offered and just kluge hacked it in.  Its quite ugly on this
front.

So yeah, the real question is who is compatible not who isnt since
almost everyone isnt :)

-- 
Trixter http://www.0xdecafbad.com     Bret McDanel
pgp key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x8AE5C721

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
Url : http://lists.digium.com/pipermail/asterisk-biz/attachments/20090106/da2ec5e6/attachment.pgp