Jul 09
29
P2P -v- Akamai.
In 1994, AS CEO of Australia’s
fastest growing ISP, I insisted that we install W3c cache servers at each dial-in
pop around Australia.
And we cached the “J” server from Palo
Alto to our Portland Oregon
offices and then replicated that in Sydney.
All DNS queries on the Ausnet network stayed in Australia
saving overseas bandwidth.
Every morning (2:00 am),
we analyzed what overseas content the users read yesterday and cached today’s
version of that content to make sure that our users had the maximum possible
enjoyable online experience.
We were charging by the minute/hour so in effect we were (apparently)
working against our own interests by making sure that users had rapid access to
their content. So would argue the Telco’s. Nevertheless, in reality, the word
of our “network” speed got out to Australians and we were deluged with daily
sign-ups. After all APC magazine voted Ausnet as Australia’s
best value and fastest ISP several months running.
For a long time – at least until Skeeve Stevens, Ausnet developed
the trust of its users.
When I built OGN, the meme continued with the development and
construction of the world’s first Terabyte internet cache. We constructed the
AUIX and allowed other ISP’s to access the Terabyte cache through the Sydney
Internet AUIX facility.
However, the Koltai caching concept was only targeted at
decreasing internet “lag”, packet loss and increasing the user online
experience.
I had not considered the concept of Web 2.0 or Google to
increase the user experience by scrapping cached content to deliver it in a new
format.
This idea was the business model of the Newspapers,
Television and radio news.
i.e.: Grab the content from Reuters/AP – add a couple of opinions
apply the current editors’ editorial guidelines and publish – charging for the
result.
In other words, News organisations were in the aggregation and
arbitrage opportunity business.
The internet caching “meme” continued to expand through the
nineties until a firm called Akamai utilized young Adrian Chadd’s Squid
software and rolled out a worldwide implementation.
Suddenly all ISP’s had the benefits of Akamai caching.
Unfortunately, for Akamai, P2P started to become popular about the time they
were rolling out.
Akamai’s IPO in November 1999 at $174 per share must have
put a smile on quite a few stag investors over the following six months as the share
priced grew to nearly $300.00.
As P2P utilisation increases Akamai shares decrease in value
to where in the last twelve months the Akamai share price has been trading in
the $12.29 – $23.58 range.
Why is that?
Akamai was the big white hope only a few years ago. It would
“decongest” the internet – particularly in a country like Australia
where the lag time of the pacific fibre transit added an inordinately
unacceptable delay to the delivery of each packet of content.
But the Akamai model is based on the concept of server client
distribution.
If that model is nullified by:
a) every
ISP installing their own CDN for content distribution (e.g.: TPG Internet –
IPTV, Internode TIVO content)
b) P2P
illegal file sharing.
What then is the winning combination?
Well, one winning combination is the concept of the content
cache as the delivery model – whoever owns Akamai also owns the world largest
Internet Filter and automatically the largest amount of content. (No, I don’t own any Akamai shares.)
Is uncle Rupert looking at a triple play with Yahoo and
Akamai with a couple of ISP’s in every country thrown in for good measure?
Well possibly not, but if I were he – then that is what I
would be looking it.
But, like always, I left out the best part.
The above triple play would only come into it’s own as a
Goggle killer with the inclusion of P2P.
After all, today the media lacks consumer trust.
P2P has consumer trust plus it’s edgy, a little bit naughty
and currently free.
Gee if you added ubiquitous caching (with appropriate
editorial control), added the worlds dominant financial services search engine to the
front end and delivered it via P2P, who wouldn’t love it?
Ummmm Google?