Wednesday, May 15, 2013; 11:14:23 AM Eastern
Working without servers
- One of our fetishes at Small Picture is getting cool stuff working with no servers.
- You can get a lot done. For example -- Little Outliner. There's only one function it relies on a server for, the importing of OPML. And if you think about it, it's ridiculous that that function has to be done through a server. The code is running in a browser, which is perfectly suited to get a file over the Internet.
- Many of the limits of the 2013 web are accidents of history. Someone at some time thought an operation was too powerful, that it could be abused, so they made it illegal. But they didn't close off the ability to call a server to do it for us, so we effectively can do anything we need to do.
- A great example of that is the way Dropbox allows Fargo to save to the local file system.
- This is something we're not supposed to be able to do. But we're doing it anyway. And nothing is breaking as far as I can tell. We're inside a very simple easy to understand sandbox, a sub-folder of the Dropbox folder.
- And in addition to being able to write to the local file system, we get cross-device synchronization for free. That's a great deal. (Understatement.)
- All this means things are shifting pretty radically. Who is the operating system vendor if my files are equally accessible on Windows, Mac, Android, Linux, IOS, etc? Dropbox is, that's who.
- A couple of years ago I asked the guys at Automattic to add a feature to their API that would allow me to store a small XML file along with a blog post. Had they done this, and if we had followed where it would logically have led, today they would have the equivalent of Dropbox, along with content management. You'd be able to publish just by saving a file.
- Today you have to figure Dropbox sees this opportunity. And that they will, eventually, navigate to roughly where WordPress is.
- The server is moving. A fair amount of what used to be "up there" is now in JavaScript, in the browser. The network services are doing what only they can do, provide global access to published information. And unfortunately due to history, they are also acting as pointless proxies for things a browser app should be able to do for itself.