I’d kind of gotten out of the habit of doing tab sweeps, largely because my Twitter feed is such a seductive place to drop interesting links. But as of now there are around 30 tabs open on my browser, each representing something I thought was important enough to think about and maybe write about. Some are over a month old. Some of them have been well-covered elsewhere. All I assert is that after I read each one of these, I didn’t want to hit command-W to make that window go away. Unifying theme? Surely you jest.
Databases · Jeff Atwood’s Maybe Normalizing Isn't Normal unifies a lot of the pointage that’s going to the (many and interesting) developments in the storage space; and along with the links there’s stuff that’s worth reading.
Also on the database front, I got email from the CouchDB guys wondering if I might be able to help them get access to a Niagara machine. Since the work on Wide Finder 2 is slowing down, they now have CouchDB running on that machine. That sucker has decent I/O and runs Erlang well, so I expect great things.
Erlang · Speaking of which, two of my tabs belong to Joe Armstrong: first, Itching my programming nerve, which plugs an Erlang-based Wikipedia clone. Since Wikipedia is already insanely efficient in my opinion, if they can do better that’s a news story. And I totally enjoyed UBF and VM opcocde design; literally laughing out loud, which is rarely provoked by opcode design discussions.
Further on the subject of cloning things in Erlang, Rabbiter - Open Federated Pubsub Server claims to be the Next Big Thing in federated Microblogging. No matter how much you like Twitter, the notion that there’s going to be One Big Centralized Microblogging service is just not compatible with the Internet, so federation is interesting.
Oh, and the new Delicious has some Erlang too. It’s popping up all over. Hey, the Seventh ACM SIGPLAN Erlang Workshop is right here in Vancouver in a few weeks. Wonder if I can sneak away from the family on a Saturday?
REST and the Web · Even if AtomPub turns out to be as big as I think it’ll be, it’s strongly document-optimized, and there are lots of Web Resources you’d like to do CRUD on that aren’t documents at all. Joe Gregorio started wondering out loud what the AtomPub analogue for data might be, if by data you mean JSON; see RESTful JSON and RESTful JSON Followup and Mailing List. I smell low-hanging fruit.
Ruby & Friends · This is really only Ruby-related because it happened at a Ruby Conference, but the second half of Chris Wanstrath’s keynote, from the Ruby Hoedown conference, is worth a listen. He tells you to start a side project and read lots, but to stop reading blogs like the one you’re reading now.
I also enjoyed Matt Aimonetti’s Ruby developers don't scale, if only because you could have run it a dozen years ago, globally changing all instances of “Ruby” to “Java” and “Java” to “C”. When there’s a hot new technology, there’s always going to be a developer shortage. For a while.
Next, there’s Zero to Production in 15 Minutes from Charles Nutter. He points out how easy it is to pull together a complete application staging environment with JRuby and GlassFish and so on. But I see his stanza of eleven shell commands and I’m thinking “One-click installer time”. This is just a step along the way. Obviously, something similar is in order for native Ruby; quite likely Phusion is at work on it.
When I link to Charles I think about the JVM and then I think about John Rose, who recently published Happy International Invokedynamic Day!, which makes me personally very happy. I think back to December 2004, one of the first steps along this road.