« Images of the Day - Scottsdale, AZ | Main | pulver.com PARTY at Supercomm 2005: June 7th in Chicago »

May 01, 2005

Guest Blogger: Henry Sinnreich - re: Enough SIP / Skype Foolishness

Stephen Smith has posted an excellent summary and conclusion about the "Enough SIP / Skype Foolishness" debate on April 30. See the excellent summary at: http://www.bloglines.com/blog/StephenSmith?id=22.

So there is very little to add, except two items:

Skype is a great and welcome challenge for SIP implementers:

- To provide an equal or better user experience, all the way from downloading the application and subscribing for service to the simple use of presence, IM and voice. To assure plugging in a new SIP phone out of the box, and have it work after only entering some ID.

- To remember the P2P roots of SIP and that all call control features can be implemented using P2P SIP without an expensive VoIP network infrastructure. Not adding infrastructure costs that VoIP providers can ill afford and users should not have to pay for. Examples are expensive network based service components such as the softswitch (I don't mean the IP-PSTN gateway which is a must), the Centrex server and various other application servers, the media servers, the session border controllers, policy servers and more.

Users may even have to pay so that Skype gets blocked!

It's time to remember that on the stupid Internet the applications reside in the endpoints. There are some exceptions though confirming the rule, such as centralized conference service and PSTN gateways as mentioned.

Though private enterprise IP networks may have more and stricter requirements than consumer VoIP has, there is no reason to forget the Internet basics when buying enterprise solutions.

The topics of QoS and MOS suggest another wake-up call from Skype. Skype has demonstrated that expensive QoS enabled "VoIP network infrastructure" is not required and has arguably the best sound quality nevertheless. Wow! The best sound quality without using MPLS!

BTW: Most public VoIP providers have no QoS equipment anyway, since they don't control the network. One has to wonder who makes those buying decisions for QoS? Have they heard that QoS does not produce bandwidth (only revenue for purveyors of QoS enabled networks), nor can MPLS avoid congestion on the access links.

In conclusion, Skype is not a killer for SIP, but an excellent challenge for implementers of SIP. Skype is also a challenge when budgeting for network infrastructure, except for bandwidth.

As for security, that's another story, for some other time.

Henry Sinnreich
Richardson, Texas
USA

Share this post:

Digg | del.icio.us | Reddit | Newsvine | Google Bookmark | Yahoo MyWeb | StumbleUpon

Posted by jeff on May 1, 2005 05:37 PM | Permalink

Additional resources: Watch PrimeTime TV Shows | Watch the Jeff Pulver Show | Jeff's Qik Videos

Comments

Posted by: injection molding at June 16, 2009 09:34 PM

Posted by: Free Stuff at May 10, 2007 07:23 PM

Hi Henry,

and what if the somewhat bellheadish SIP developers and implementors do not respond to the challange? You we are talking about simple SIP since one year - I have not seen anything yet.

And you also know that the major problem of SIP is the public URI. As long as VoIP providers do not offer public SIP URIs, the average users sees not differnce between SIP or Skype. See also
http://voipandenum.blogspot.com/2005/04/what-of-sip-is-dead-and-what-not-dead.html

And I am currnetly in Geneva to whatch the NGN marriage of ITU-T and IETF ;-)

see http://www.itu.int/ITU-T/worksem/ngn/200505/program.html

Posted by: Richard Stastny at May 2, 2005 02:43 AM

That’s a nice cup of insight there! :)

Posted by: Ralesk Ne’vennoyx at May 1, 2005 09:41 PM

Post a comment




Remember Me?