Cloud Hosting

One webmaker: Zainab. Image: Courtesy of Mozilla

The service currently supports hubs for about 50 cities and towns. In theory it can support thousands. Will it get there? That depends only on my ability to build the right service, show the benefits of the loosely coupled syndication model Im evangelizing, and convince key stakeholders in cities and towns to adopt it. It depends not at all on my ability to deploy and manage servers, operating systems, or networks, for which I am deeply grateful.

Thats because Ive been there and done that. For my first web project, which I documented in a monthly column for BYTE starting in 1995, I wore all those hats. At first it was fun, and I guess for some folks it still is fun, but for me the novelty wore off long ago. The infrastructure is just a means to an end. I want to focus on the destination, not the vehicle.

Its true that, in one sense, todays infrastructure feels less personal. My first server, a DEC Alpha machine called BYTEWEB, sat under my desk where I could hear how busy it was. Now my servers are virtual machines with names like RD00155D31705A, sitting in data centers I will probably never visit.

In another way, though, the modern arrangement is more personal. I have an idea. Its expression requires me to write and deploy software. Writing the code was always hard, still is, and maybe always will be. I think thats the nature of writing, whether youre writing code or prose.

Deploying the code was always hard too, but it never should have been, and now it isnt. If I can think it, and write it, I can put it where you can use it without having to sweat the infrastructure details. In his now-classic talk Inventing on Principle, Bret Victor argues that creators need a direct connection to what they create. If youre a developer of web apps and services, cloud infrastructure helps you forge that direct connection.

Most of us, of course, arent developers. But all of us, nowadays, can find ways to feel that direct connection. The elmcity project, in fact, is partly about that. Im providing a cloud-based calendar syndication engine. But its useless without cloud-based calendars to syndicate. Thats a service that you can provide, and more frictionlessly than you might think.

Nowadays I talk to a lot of folks who publish calendars on their websites. The vast majority of these calendars, as Ive mentioned before, dont support the internet standard for the exchange of calendar data, and dont provide online data feeds that are open to syndication. When I explain this to people in schools, libraries, businesses, chambers of commerce, and arts councils, they invariably want to refer me to their webmaster, IT person, or resident geek. Because providing a machine-readable data feed on the web just seems like it ought to require somebody like that.

Well it used to, but no more. Now I show people how they can simply create a new calendar, in a cloud-based app like Google Calendar or Hotmail Calendar, that can serve both as a webpage for visitors to their site and as a data feed for syndication. If they can think it, and write it, they can put it where any person (or any computer) can use it. In so doing they become creators, not just consumers, of cloud-based services. Its wonderfully empowering.

This sort of personal empowerment is available in many other ways. When you adopt a consistent naming and tagging scheme for a set of web resources you control, for example, you become more than a producer of content you become the creator of a service that enables people (and machines) to use your stuff in ways you might not have anticipated, and to combine your stuff with other stuff you might not know about. Cloud infrastructure is a godsend for those of us who make software, and equally for all who aspire to be webmakers.

Read the rest here:
The Personal Cloud's Third Dimension: Webmakers