Ajit Jaokar is the founder of the London based publishing and research company futuretext (
How long should we wait for a standard to emerge and what is the cost of doing so?
Recently, I have been thinking that We need to innovate standards but not standardize innovation
Over the last few weeks, I have been busy both with the Web and the Telecoms side – with publishing, conferences etc; and often you see the same issue of standardization reform across different domains.
In a nutshell, I am saying that it’s easy to standardise simple communications protocols(and that’s where standards efforts should be focussed) but it is very hard to standardise more complex elements globally within a decent timeframe.
Here are two examples: The question is – How long should we wait for a standard to emerge and what is the cost of doing so?
HTML5
HTML5 is the next major revision of HTML and it covers important areas like offline browsing. Work on HTML5 started in 2003 and it was expected to be completed in 2012 (which is bad enough) but now it is expected to be completed in 2022(19 years!)
Some more comments HERE and HERE
and also some strong language in this last link HERE.. but still worth a read including it’s 105 comments!
VOLGA(not the river!)
This second example comes from the domain of mobile telecoms. As the world of mobility comes to an all IP world through LTE , we realise that the handling of phone calls over IP (especially in the mobile domain) is non trivial. Normally, call handling is session based. With IP, that’s not easily achieved(or at least needs more management in the mobile domain). Martin Sauter explains this issue in detail in his blog
There is a (non) standardised solution called VOLGA (Voice over LTE via Generic Access) which was a topic of discussion at the LTE world summit in Berlin last week where I was speaking.
However, here, the standardization body (3gpp ) seems to be taking a short sighted view (in my opinion). A problem exists for voice in an all IP world. Everyone acknowledges that. The solution is cs-fallback (clunky) (i.e. going to circuit switched for voice and using a packet switched network for data) or fully IMS IMS (is distant). VOLGA is a more pragmatic solution – but 3gpp explicitly commented that don’t endorse VOLGA!
And what’s the standardised solution?
ALL Telecom Operators should go to IMS(i.e. support IP) AND that IMS must seamlessly interconnect AND furthur more, devices supporting IMS must exist. ONLY THEN can a solution be created for voice over IP. Note that voice on mobile devices is certainly a bearer aware application – so no matter how you look at it we need a proper network layer IP solution for voice
So, the options are between a pragmatic immediate solution(VOLGA) to a distant standardized solution
In a recession, we have to be pragmatic – waiting for a standard to emerge cannot be indefinate.
There are no easy answers to these questions, but certainly a discussion is necessary ..
Image source: http://3.bp.blogspot.com/_jKscR9AiAls/SGu0fdZMSkI/AAAAAAAAATE/cT1eopGa0MQ/s400/waiting.gif
You’ve made the same unfortunate misunderstanding regarding HTML 5 that Jeff Croft made in his famous rant… Ian *never* said that HTML 5 would be “done” in 2022. In fact, the HTML 5 spec is nearly done, and it is headed to “Last Call” status in a few months. What Ian specifically said is that HTML 5 would not be considered a full W3C recommendation until 2022, which is dependent on *two fully compliant implementations*. Seeing as you can’t even find one fully compliant implementation of HTML 4 out there, saying “2022″ for that criteria to be met is no surprise. HTML 5 will be “done” in late 2009 or early 2010, if I had a guess, based on the activity I have been seeing on the HTML mailing list. In fact, it’s been published as a “working draft” for some time.
Is the stadnard process perfect? Far from it. Probably 2/3rds of the messages on the HTML 5 list of political nonsense regarding things that most people find irrelevant, like matters of process. But at the same time, HTML 5 is not nearly as broken as many people have interpreted based on the “2022″ number.
J.Ja
Comment by Justin James — May 26, 2009 @ 5:36 pm