BLF Presence Specification for FreeSWITCH

[FASTDEVICE]

I'm working with several offshore IP phone manufactures to widen our selection of compatible phones w/ 2600hz's network. However, they are all having the same difficulty in implementing BLF presence messaging for FreeSWITCH. 

Can anyone point me to a definitive and well documented whitepaper on BLF presence for FreeSWITCH and any addendum that may be required for how 2600hz may affect that implementation?  

[lazedo]

Hi,

FYI, kazoo does not rely on FreeSWITCH for blf.
you can check the following rfc's
https://www.ietf.org/rfc/rfc3265.txt
https://tools.ietf.org/html/rfc4235

hope this helps

[FASTDEVICE]

Thanks @lazedo. Does Asterisk and Broadsoft follow a different standard from the ones you are referring? Even Yealink, with their new line of phones, spent months implementing a working version of BLF that is compatible with FreeSWITCH based PBXs.

[Darren Schreiber]

This line of questioning is a bit suspect. Are you asking for a definitive way the packets work?

The problem here is in RFCs in general, specifically because of RFC 2119:

The all-capital terms "MUST", "SHOULD", "MAY", "SHOULD NOT", "MUST
NOT", and "RECOMMENDED" are used as defined in RFC 2119 [5].

The word "MAY" is the root of all evil. In fact, the FreeSWITCH team has a running joke about it when advertising their conference, ClueCon - you SHOULD attend, you MAY learn something. ;-) Or something like that.

Anyway the joke is pretty simple. Because the word MAY is used in the RFC in various spots that are really ambiguous, different vendors end up implementing whatever they interpret that word to mean. Thus, you end up with different implementations. The word MAY exists in the RFC for the goal of ensuring people can use the specification in differing ways, but it ends up serving as something that creates issue when two people trying to implement the same functionality do it in different ways. Thus, a mess.

Thus SIP :-)

So, if you're asking "what packets exactly can be transmitted", the answer is read Luis's specified RFCs . EVERYONE - including Broadsoft and Asterisk - MUST be following them. But not everyone MAY agree on exactly how the RFC reads, thus, the implementations can take a while to get right on each platform.

Trial & error FTW

[FASTDEVICE]

Thanks for the clarification, but yikes what a mess. Back to snooping with wireshark.

[Darren Schreiber]

Yes, it is a mess. I know BLF seems like "just a light" and that is why everyone gets so mad at when it doesn't work, but yeah, what you're looking at is literally why it is such a mess. :-/   I'm sure it seemed like a good idea at the time when those people implemented it.