Ever
since Microsoft blamed
netbooks for some of its troubles,
there has been no shortage
of opinions on this new device's disruptive effect on the PC, software and
semiconductor businesses.
A recent article
"It turns out that about 95 percent of what I do on a computer can now
be accomplished through a browser."
Without proper research it is hard for me to say how much of
the regular activities of how many people can be accomplished using a browser,
but anecdotal evidence clearly indicates that a browser does a lot of things
for a lot of people. The question is not
whether the browser is indeed the killer application, but whether it belongs to
the netbook. The problem with the browser as we know it - running on a client
machine - is that it carries a lot of user state data (your personal profile).
That means I need my netbook to use my browser.
Just any browser is not good enough for me any more. My
browser has become a very personal thing to me: it is intimately personalized
with my bookmarks, cookies, RSS feeds, usernames and passwords, form auto-completes,
browsing history, extensions, plugins, and more. I have two netbooks and five
PCs, but only one of them is really "mine" because only one has all of my personalized
"state."
Are you familiar with that sinking feeling when your precious
netbook, laptop, smartphone is dropped to the floor - "did I just lose my
latest everything?!" Imagine the horror of leaving it on the airplane or at a
Starbucks, with all the passwords and browsing history for anyone to see? What
makes it precious to you is not so much the device itself, but the personal
digital environment that was created among the applications and settings.
Let us now consider a new approach to computing. Why not
move the browser itself into the cloud? After all, we trust the cloud with our
email, calendar, contacts and files. Why
not the browser with all its personalization? Just think of the market
opportunities:
- Internet
service providers can include value-added Virtual PC or Virtual Browser
services to their broadband bundles
- Software
vendors can get off the declining curve of OEM shipments and embrace broadband
service providers as their new channel
- Processor
vendors can ship tens of millions of server chipsets to run billions of Virtual
PC/Browser user accounts
From a business and consumer perspective, Virtual PC and
Virtual Browser services will liberate users from the invisible chains they are
tethered to, such as the bloated desktop or laptop. Having all your computing services in the
cloud offers freedom, convenience and flexibility.
According to Gary Krakow's article in The Street.com,
Ballmer knows that making his
company's software available on the Web (similar to a "live" version
of Microsoft Office that can be accessed on any computer, anywhere) will be one
of the "next big things."
We know that using a local
browser to access things is neither easy (need your bookmarks, cookies,
passwords, etc.) nor safe (keeping your state on the client makes you
vulnerable). According to Washington Times columnist Mark Kellner,
there is a better
way.
Netbooks are
clearly a disruptive threat to the PC industry as we know it. Something tells
me that netbooks are just the first step in breaking away from the "fat client
with a browser" model. They need to get cheaper, simpler to own, last longer on
one battery charge, and also weigh less. The way to get there is not by adding
CPU, storage and memory (these belong in the cloud!), but rather increasing the
usability of the netbook by giving it more connectivity options, a nicer
keyboard, a fancier multi-touch display and turning them into sexy, stateless
"zero-touch" clients. And then they will really take the world by storm,
improving the quality of experience for existing personal computer users, and
bringing the digital lifestyle to billions.
interesting, but... how do I connect these zero-touch clients to my different cloud services?
Good question. The whole point is that a true zero-touch client (unlike good old thin client) does not connect to any services. Services themselves connect to it instead, under the control of the universal cloud services platform.
As a user of a zero-touch stateless client all you do is turn it on. Once powered up, it connects to the network, and thus makes itself visible to the cloud-based services platform. The platform asks for user credentials, and in response displays the selection of available services. The service selected is told by the cloud-based platform to connect to the client.
The client is too dumb (read simple, reliable, cheap) to do more than that. It just knows how to draw pixels on the screen. The smarts are in the Cloud. IMHO, that's the only way to scale it really big.
You make some good points. Browsers have become bigger and more complex than desktop operating systems of 10 years ago. The key to seamless service delivery is to marry user state data with the specific applications and the chosen device, at any location and anytime. This must be done in a secure way with the right trust relationships.
This is not possible when personal information and preferences are spread across a number of devices with different vintages of the data. This situation also compromises security of personal information.
So in order to have my information and preferences persist between say, using my laptop and using my phone, I would need my browser to be virtualized and accessed over the network from a very thin client. At the same time, this should allow me to access non browser-based applications from any device too.
The only question is whether networks have enough capacity and low enough latency to make this feasible. My impression is we are definitely there on wired networks today and are getting close with wireless.