Slashdot Mirror

Hmm, it is even easier since drafts for the H.323 protocol (This was the first link I googled which gave a date...) Since H.323 is based on previous standards I'm sure drafts describing this functionality before that date can be found. H.323 was the first open VoIP protocol suite to gain wide use.

the filing date was "February 25, 2000". How many of those were around circa 2000? Skype only began around 2002-2003

Packetizer

I have AT&T. Bad: Rare drop-outs of a few seconds; dropped calls; sometimes one-way audio; too often the Locate Me feature doesn't ring the other phone; can't call Canada for free (unlike Packet8); may be causing 2-second loss of traffic through the TA hub exactly every 60 seconds. Good: Very low delay, great voice quality; TA directly faces network so great local QoS (voice has higher priority than other port, e.g., PC).

Colleague has Packet8. Bad: Frequent audio distortion; slightly noticeable delay; no QoS on TA (behind firewall) so, e.g., PC downloads cause packet loss. Good: Stable calls; inexpensive; can call Canada for free.

Oops! SIP is of course a replacement for H.323.
a comparison of the two.

Ahhh. Just imagine that.
Funny.. Sounds alot like H.323 to me. (The primary VoIP protocol in use today)
If you want to know more about VoIP have a read at www.packetizer.com

There are plenty of hardware and software H323 and SIP (A competing but less powerfull VoIP protocol) based phones out there. Voice quality comes down to a combination of available bandwidth and compression codec in use.
As the article implies a well designed VoIP network can infact give better voice quality than PSTN.
H323 already support seemless routing of calls on and off of PSTN networks, and publishing of your H323 gatekeeper/gateway's ip address in DNS for seemless interdomain routing without any prior configuration between the calling and called parties.
H323 DOES already do everything you want, it is an open protocol, and you can already buy hardware phones that are basically plug in and call.

Since I work for a company that has products using both protocols, any bias I have is hopefully personal rather than commercial. Here are the differences between the protocol that really matter to me:

• H.323 is definately telecom oriented, being a product of the ITU. If you just want to get a phone call across the Internet, this is the protocol to use.
• The packetizer.com comparison mentions that H.323 is built on the ASN.1 notation, but it fails to mention that there are multiple options for encoding. Most ASN.1 protocols, such as SNMP, use the Basic Encoding Rules (BER), which are relatively easy to understand and implement. However, H.323 is the only protocol family I know of that uses Packed Encoding Rules (PER). The PER spec is HUGE and hard to understand. As of a couple of years ago, just about all H.323 apps were built on three or four commercial toolkits which were quite expensive because there were very few people who wanted to devote a year of their life to figure out PER. Now that OpenH.323 is out, I suppose the situation is better, but this used to be a huge barrier against getting into the H.323 space.
• SIP is more Internet oriented, and its design takes advantage of capabilities on the Interent that you don't have in a telephony network. On the Internet, if you want to send queries to a dozen different servers to see which one I'm connected to, its not that big of a deal, but the equivalent search in a telephony network would tie up a lot of resources. Therefore, telephony data tends to be centralized, but Internet data can easily be distributed. This differences shows up in that H.323 favors a central Registration/Authentication Service (RAS), where SIP offers a distributed search mechanism.
• Since SIP is a text base protocol, a Perl hacker can experiment with SIP services without a toolkit.

Ick. H.323 is a dog to operate through NAT. If both parties are using NAT, you have problems getting one side to call the other. I've been able to call out from behind a NAT router to a modem user, but not accept calls coming the other way. It looks like SIP also needs a proxy of some sort.

For a useful comparison check out this H323 vs SIPcomparison. Looks like SIP is a lot simpler but less interoperable with things like PSTN.

Really, these days there's no excuse for protocols that hide IP information in the packet data (that's FTP, H.323, and a ton of others).

Jon

You might want to look at H.323 - http://www.openh323.org/. It's got support for Linux, plus H.323 is used by NetMeeting on Win32, now all you need is a Mac client. Also look at http://www.packetizer.com/h323link.html. You may even be able to do a porting from the openh323 linux code to OSX?? (not sure on this one)

Any kind of plan involving a single identifier to reach a person anywhere would be accomplished>through creative use of DNS, and could involve actual names and words. The accounts themselves might possibly use a email-like name@provider kind of system. But I dream.

Well, it looks someone already tought about that... It's called SIP (Session Initiation Protocol) (RFC 2543) I know it allows IP phone addresses like sip://user@host and support call forwarding and other nice stuff... and its much much simpler than H.323

LinuxTelephony
openphone.org
Packetizer.com
SpeakFreely.org
Voxilla.org
___