August 23rd, 2005

APAN Day Zero

» , , , ,

Someone left a comment asking me what’s APAN or Asia Pacific Advance Network. It is a meeting where AP network researchers get together to discussing advance networking issues and also a place where Advance Research and Education Network (AREN) (and also the GRID lately) people gather to discuss network collobration. SingAREN is the representative for Singapore in this area which is one of the reasons I am here. In fact, SingAREN is going to host APAN next year in Singapore so we have quite a big team here this time.

But like APRICOT, a lot of side meetings are also held concurrently. Just yesterday alone, I have to switch between APSTAR, IPv6 Summit and also chairing APEET and JET-Internationalized Email Address.

Incidently, for those who is interested in Internet statistics in Taiwan, you should look at the presentation Ching Chiao did at APSTAR. 14.6M Internet users, E-Commerce NTD 35b (~1b USD), 2.2M Skype, 400k blogs and 108% Mobile penetration. The last one 108% is interesting because it means many people are holding two mobile plans or more.

(Speaking of werid statistics, do you know that Japan actually has a higher broadband penetration then PC ownership? That’s a story for another day)
Read the rest of this entry »

August 6th, 2005

IETF Day Five

» ,

Today is the last day of IETF. And it is also one of the most important reason why I am attending IETF this time – ENUM or specifically Carrier ENUM.

The (User) ENUM part of the meeting went quite well easily but I think I surprised quite a few people when I stood up and objected to Shin’s mobileweb registration (basically, a ENUMservice for mobile web). Now, Shin is a close friend but I think it is a bad idea to start putting session negiotation information into ENUM. Its a slippery slope to go near there – DNS should remains as DNS – you throw something at it and it give you back something. The capability negiotation should be done within the session setup, and especially HTTP already provides User-Agent negiotation that serves the need.

I can understand why mobileweb would make sense – It is very helpful for a registry/registrar who can start to market a new product (“register your mobile web address now!”) but lets not taint the protocol.

The Carrier ENUM portion of the meeting is far more exciting. Two ideas was thrown around (1) use of non-terminal NAPTR record and (2) use of some defined carrier label delegation under the tree to tie Carrier ENUM tree under After some interesting discussion, the rough consensus seem to be leaning towards Michael’s proposal using carrier label delegation. I still dont like it but ah, rough consensus is rough consensus and I lost :P.

Most surprising is that the room also have rough consensus that Carrier ENUM as part of the ENUM working group work. This is quite different from the last time we have such discussion where people objects quite strongly against it. In this regard, everyone wins :-)

On different note, SPF/SenderID seem to be dying.

August 4th, 2005

IETF Day Four

» , ,

Nakayama-sensei from Tokyo University shared a very pleasant story with me this morning.

They runs a popular site called Live Eclipse that keeps track of eclipse schedule. The site also have an Japanese IDN 日食中継.jp which they publise concurrently.

On the last eclipse, 9th April, they have over 2m hits in a single day. The interesting part is 15.8%, or over 400k of the hits comes from the IDN name 日食中継.jp. This is very surprising because the data we have in the past shows very little (less then 1%) utilization of IDN. This maybe the coming of age for IDN :-)

(Please bear in mind that 15.8% resolution is despite the fact that IE still don’t have IDN support – Michel told me it will be in IE 6 beta 2 ie in IE 7 release.)
Read the rest of this entry »

August 4th, 2005

IETF Day Three

» ,

Woke up very early this morning to do a presentation for an internal event in Singapore. Yep, I am still in Paris so the presentation was conducted using Skype with video4im. Both sound and video is amazingly good although I look like a dork in the terminal room talking into the computer as if I am giving a speech (I am!).

The highlight of the day must be the VoIP Peering and Interconnection BoF. The room for the BoF is filled (one of the largest BoF i’ve seen) and 2 hours was dedicated to the discussion (wow). Presenters includes people from AT&T, Comcast, Telio and most of them talks about (a) the need to have a standard mechanism for seamless (and cheaper) VoIP peering (b) the use of (carrier) ENUM in faciliating (a).

The discussion thereafter was even livelier. Layer 3 or 5 peering joke aside, VoIP peering is serious especially as cable & dsl ISP starts offering IP Telephony. Rough consensus seem to be that we don’t really want to create yet another protocol (yep yep) but to use existing protocols to do a BCP (Best Current Practice) document – the only problem none of the solution we have is considered best practice yet :)

Ah, Friday Carrier ENUM discussion will be exciting….

btw, APEET did our VoIP interconnection using private ENUM with wildcard for our trial in Kyoto. Yep it works but I am not too sure it will scale.

Update: Slides from the VOIPEER BoF

August 2nd, 2005

IETF Day Two

» ,

Spend the morning in Emergency Context Resolution with Internet Technologies (ecrit) WG meeting. Basically, the working group is looking to handle emergency services (police, fire, ambulance etc) over the Internet. Very appropriate too considering US has now mandated emergency service and in the long term, we do need an pure IP-based emergency services when we phase-out POTS (our evil long term plan! :-).

There are only two problem I can see:

1. lots of talks about how caller can authenticate the emergency response centers but not much talks about how emergency response center going to authenticate the callers. Considering how many prank calls typical emergency response center gets, I wont be surprised if their priority is to identify the caller first and less about your house on fire.

2. the group assumed that the emergency response centers will receive the emergency “call” over the Internet. Any bet how long it will take response centers to install a SIP server into their system? Especially a system which will allow anyone from anywhere in the world to call them anonymously?

Regardless, its an important piece of work to watch out for.

After lunch, spend an hour with some people to discuss how we should proceed with Internationalized eMail Address. Good meeting and now it’s time to get to work.

Now sitting at the Remote UI BoF. Yea, sound like glorified VNC or RDP but (1) they are doing this on widgets level not framebuffer and (2) the people behind this works for Nokia. In fact, the slides they are showing have Nokia phone as a remote terminal. Hmm…

August 1st, 2005

IETF Day One

» ,

I missed a couple of IETF and oh man, so many things to catch up! Anyway, here are some interesting happenings, in no particular order, in the last 24 hours:

– Did a presentation on Internationalized eMail Address earlier at the AppArea meeting and got my 5min of fame (literally haha). Then spend the next 30mins along the corridor debating with John Klensin whether we should modify the SMTP or not.

Dave Crocker couldn’t stop talking about dkim. Basically, it is the DomainKey proposal by Yahoo! and it will be discussed at the Mail Authentication Signature Service BoF on Thrusday.

Richard Stastny was grinding me on merits of combing user & carrier enum in at the opening reception last night. Ok, Ok, I agree why we absolutely need it but I still think it is extremely ugly. We probably going to continue the debate at the ENUM meeting on Friday.

Christian Huitema gave an interesting presentation at the Application Area meeting basically along the line Password (and even Challenge-Response) is dead. Yes, read that again : Password is dead.

Very controversial obviously but he raised some very good points. Based on a 0.10cent for a zombie PC per week, a 30bit strong password can be broken for less then 1cent, a 40bit passphrase for less then 20cent, 7 random char password for $50 and 8 random char for $5000. Fundamentally, “any password generated by the user or can be memorized by the user can be cracked”.

– Had lunch with Michael Suignard who is here to give a talk on UTR 36 (Security Considerations of Unicode). He is also here for an IAB discussion on Internationalization of Hostname, or something along that line.

Geoff Hutson gave the presentation on “IPv6 Multi-Addressing, Locators and Paths” to lead the age-old IETF debate “Is IP Address an Identifier, Locator or Routing Object?”. Very lively discussion at the Internet Area session.

– Lark-Kwon Choi from Korea Telecom presented his I-D on the requirements for Data Broadcasting Service over IPTV at MMUSIC. Seem like they are very clear how they want to do IPTV :-)

– Henry Sinreich & Jon Peterson have an interesting discussion on L3 & Voice peering yesterday evening. Unfortunately, I was distracted by a friend so I didnt caught much of the discussion.

– France Telecom/Orange has a “IPv6 IMS over IPv4 GRPS/UMTS” demostration at the corner of the terminal room. Apparently, they got a ISATAP client running on Window Mobile and also Nokia Symbian Series 60. Hmm…

– Looks like there is a lot of support for ICE (Interactive Connectivity Establishment), a mechanism for NAT traversal using existing STUN, TURN etc. Great..hope we can get rid of all those confusing uPnP, STUN, TURN settings we have on our SIP phone soon!

– Will be meeting some Japanese who is working on some high-speed wireless (4G) backhaul later this evening. That would be interesting: Imaging high-speed Wifi Internet access on MRT :-) Actually, the Japanese already have a trial on their Shinkansen several years ago. But having it on our MRT? Ah, that’s still only a dream…

April 15th, 2004

RFC 3743


Today mark the publication of RFC 3743. aka JET IDN Guideline. A work I started in the Joint Engineering Team (JET) after realizing that it is not possible to handle traditional and simplified ideographs within the IDN protocol in 2001. To certain extend, ICANN’s Guideline for IDN is based on this work. been a long journey (my god..more then 2 years!) and I am glad I can put this behind. This work would not be done without the assistant of John Klensin. Thanks all!

ps: This is not Standard Track. It is an Informational RFC.

March 3rd, 2004

Time flies

» ,

Don’t asked me what happened the last few days…Working Group meetings, breakfast meetings, lunch meetings, tea meetings and swoosh, I am now back in Singapore Airline lounge waiting to go back Singapore. Hmm, what really happened? I wonder how am I going to write my trip report…

February 28th, 2004

IETF, Seoul

» ,

I am now sitting in the bar in the Hotel Lotte with fellow IETFers in Seoul. Beer and WiFi..what more could we ask?

ps: Oh, dda come over my place for dinner. Yes, I have the Zaurus sync charger cable for you :-)

September 27th, 2003

IETF Education


Spend last night trying to put together a site for IETF Education Team using movabletype. Yes, I know MT isnt designed for CMS but hey, it works :-)

Take a look at how I twist MT at