Matthew Ingram at the Globe and Mail takes Wired to task over a recent article that implies Big Search will save the world and change the way we solve problems. I think Matthew’s right, and that the way we can now mine vast amounts of data isn’t a substitute for science — merely an accelerant.
I think this for two reasons. First, Google can’t solve the problems with machines; and second, correlation is not causality.
Cloudcamp San Francisco, the first in a series of events centered on cloud computing, took place on June 24. If the roughly 300 attendees at this informal industry meet-up are any indication, cloud computing is a popular and rapidly growing subject.
“Based on attendance at recent Amazon cloud conferences I was expecting maybe 100 people”, said CloudCamp instigator and Enomalism co-founder Reuven Cohen, “this is an amazing turnout after only 3 weeks of planning.”
GigaOm’s Structure08 event is a wrap. It was an amazing turnout of people in the next-generation infrastructure world, and a packed day of panels and discussions.
I had my computer and phone off most of the time, since I was announcing speakers and moderating panels for much of the day. So I’m now scouring the net to see how we did.
Readwriteweb wrote a great piece on cloud operations.
Dan Farber, who covered Structure for CNET, wrote several pieces on the sessions.
I was incredibly lucky to have great panelists for panels on cloud platforms. While most of the discussions were fairly pragmatic, we did of course invent some new terms:
“Bare metal” clouds: Clouds that aspire to nothing more than giving you root more economically.
“Little Fluffy Clouds” (thanks to The Orb) after Tony Lucas referred to “Loving clouds” as those clouds which really want to do no evil.
A cloud user’s bill of rights, which would outline portability and so on.
“Cloudbursting”, the idea that an enterprise private cloud might burst into the public cloud temporarily.
I wish I’d known that Facebook was hiccuping while I was talking with the scaling panel, particularly since we had Jonathan Heiliger with us.
Then a dozen of us headed to an excellent pub on Haight and stayed up until far too late. Great end to a great couple of weeks in the Bay Area. Now, as the Spirit of the West say, it’s home for a rest. But I leave you with this.
Some people are born beautiful. Others become beautiful. The rest visit cosmetic surgeons as they wage a holy war against Father Time and their own genetic limitations. And when those people eventually choose the wrong surgeon, they’re disfigured for life, turned into freakish mutants who are a strange and distracting parody of beauty.
I just finished watching a presentation on Eucalyptus, an open source layer for cloud computing that emulates Amazon’s EC2.
The Eucalyptus team replicated nearly all of EC2. They know this because they pulled down Amazon’s Web Services Description Language (WSDL), which describes the various function calls someone can make to Amazon, and made sure Eucalyptus could do the same thing. It’s not a secret; in fact, you can check it out here.
This raises an interesting point. For a traditional desktop developer, if two interfaces are identical, then writing code to one means it will work on the other just fine. But there are two other things to consider if you’re choosing a cloud platform: Operational reliability, and network effects.
The first one’s pretty basic: Don’t use someone who can’t keep their cloud running.
The second one is less obvious: The value of a cloud service isn’t just what it does; it’s also how many people use it.
Someone building a CRM application on Salesforce has a lot more contacts and partners to work with than someone using Sugar
A communications application built atop Webex Connect can automatically invite more people to a meeting than someone using Acrobat.com
In other words, when considering a cloud’s services, we can’t just look at the richness of the APIs it offers. We have to also consider the network effects it enjoys.
I’m in San Francisco this week for Velocity, CloudCamp and Structure’08 and the valley buzz du’jour is definitely Cloud Computing. Tomorrow evening is the inaugural CloudCamp and I’m bracing myself for more argument over the rules of membership in the Clouderati.
It strikes me that for the business minded, the question is not who’s in and who’s out, rather, who’s going to be able to stay in. Cloud is simply an IT delivery model. It comes with high customer expectations that few companies are tooled up to meet. 24×7 availability, self-service provisioning, pay-per-use billing and internet scale, all for $1.78 a month.
I’m hoping for good debate about Cloud adopters, their expectations, and how we can implement Clouds to meet them… without going broke. See you there.
Web2Expo New York is happening in September. We’ll be moderating a panel on Big Cloud platforms.
From the session description:
Cloud computing is self-serve outsourcing for web companies. Clouds give even the smallest startup access to world-class infrastructure that can grow as needed. And developers build apps faster, because they start with the building blocks of online applications: Authentication, storage, messaging, and the social graph.
Can’t get into specifics on panelists yet, but they’re good; April’s session on databases of the future brought together senior folks from Microsoft, MySQL, and Amazon SimpleDB.
Cloudcamp SFO is the inaugural event in a series on cloud computing. Launched by the irrepressible Reuven Cohen, it’s a free mashup of smart people from all over the cloud computing landscape.
It’s free to attend, and if you’re in town for Structure 08, it’s a great way to get in the cloud mood.
If you’re in the Bay Area and run a website, I’ll be speaking at two lunch events in both Sunnyvale and San Francisco next week.
Coradiant started running these lunches regularly a couple of years ago, and they’ve always been a great chance to talk with web operators about the challenges they face running websites. Enrolment is free but it’s filling up fast.
Bitcurrent is part blog, part analyst firm, and part resource site for web operations. We're a loose federation of pundits and entrepreneurs with experience in networking and technology.