Mark Roseman's Blog

JotSpot Got Bot, err, Bought.

Well, the Google giant has bought JotSpot. Good (presumably) for the JotSpot gang - congrats! As for Google, you always wonder what they're thinking with these acquisitions. Acquiring some good engineers? Maybe. Acquiring a service so they can host generic wikis (like Blogger for generic blogs)? Possible, but l'd guess less likely (though wikis are a natural for content network ads). Acquiring a platform that's been focused on allowing users to develop their own form/workflow etc. based web applications in the hopes that Google can use that to somehow throw enough darts to hit on that next killer app to drive increasing end user generated content? That'd be my guess.

De-rail-ed.

With better than expected progress on 5.2 and some other things I needed to do, had a bit more time this week to really start into my for fun project, which I'm using as an excuse to play around more in depth with Ruby on Rails (modulo the various interruptions that always seem to occur).

It's been a while since I've done stuff within a large framework (as opposed to with libraries, or more sparse frameworks), so I'd pretty much forgotten what its like when you first start working in that kind of environment. Follow manual, search documentation, browse files etc. for a long time, and then add one line of code, which after-the-fact makes perfect sense. But like good frameworks, it's not long before you know what should probably go where, and things are just zipping along. More like wiring up a bunch of pieces than "real" development. Because this app is fairly generic and along similar lines to other things I know have been done in Rails, I'm guessing most of the things I'll need will probably be available already.

Some of Ruby's idiom's work particularly well for code-in-web-template things. Also noticing that the DRY philosophy is somewhat infectious, along with the general quest for the most concise way to express things. Not all wonderful, but the basics of incrementally building out the application seem to be coming along pretty smoothly. And it's a nice change of pace.

Definitely not the NHL.

Went to a Guelph Storm (local OHL team) hockey game yesterday... fairly entertaining, and while you'll never mix them up with NHL'ers, they weren't bad. Plus, being a five minute walk from the arena, and with the most expensive tickets being $20 (we were sixth row, center ice... I can imagine how much that would cost at the ACC if it actually were available!), it was a good time, and I'm sure we'll go a bunch of times this season.

Tcl/Tk 8.5.

Starting to put together a quick guide to Tcl/Tk 8.5, which is still in late alpha stage. While the whole core team structure has I think been slowing things down, it seems about one or two decisions away from feature-freeze, after too long in development.

I Love Flying.

After the 2am phone call alerting me that my flight getting here was cancelled, I was just thrilled to get a call today saying my return flight was cancelled too. With any luck maybe I'll get home by 2am. Grumble.

Lifestyle Issues.

Not conference related, except that the main reason I come to this is really the range of interesting people who attend, including several who are both spectacularly successful at what they do as well as spectacularly understated and humble (and also Larry). There's a reasonably good number of entrepreneurs and others involved with small businesses of various sizes, so its especially nice to talk through a lot of the issues that can come up in that context, particularly when it comes to balancing work and related activities alongside lifestyle, bigger picture motivations and so on. Thanks Steve.

Paper Summaries.

For the non-Tcl folks curious to follow (for genuine interest, gallows humor, or anywhere in between), Michael Cleverly is blogging summaries of each one. Thanks!

Press Release Leads I'd Like to See.

Larry McVoy of BitMover and Mike Doyle of Eolas, both well respected and universally revered by the global free software community, today announced at the annual Tcl language conference their new browser-based operating system. This remarkable innovation was inadvertently spurred by a small clause inserted into the new version of the GPL just before it was ratified, combined with Eolas' well-known browser patent.

Tk in a Box.

Usual discussion about the sad state of Tk (well, actually the sad default state, as compared with the surprisingly good improvements in what is possible, if you do enough digging). Usual people involved on each side. Because of the improvements that actually have going on behind the scenes, and the frustration among some of us increasing as the steps needed to make it actually work out of the box getting smaller... well, the incentive and motivation to just get it fixed in some form (fork a new distribution most likely) have taken a big jump.

With an 8.5 version coming together, and a second edition of Ousterhout's book actually going to be coming out (John's going to be somewhat involved, but mainly Ken Jones, with help from Brian Griffin and Dave Welton amongst others), and a couple of other things that will be happening soon... well, might be an interesting mini-opportunity for all.

Geeks and Money.

Don't have a ton to say about this year's Tcl conference... it's been good so far, a few particularly interesting things, and nice to see many of the people here again.

Richard Hipp's opening keynote was nice, about SQLite, which as most people know is in use in tons of different places, open source and commercial. Interesting new thing was the new addition of a full text search thing, which he did with the help of four engineers from Google. Looked like quite a nice addition (I'm sure Jean-Claude will have stolen most of it for Metakit by the time I'm finished this post).

Larry McVoy and I spent an hour last evening with Richard, trying to persuade him that he should be making a whole lot more money than he is from SQLite (particularly on the support side). Not that it was necessarily for the sake of more money, but we all could see the possibilities if that freed him up to do even more things. It's hard for a lot of tech people to understand value-based approaches to pricing, why people buy support, and that sort of thing, and Richard has probably the most pathological notion of "fair" I've seen. I hope he comes around, as he's such a lightning rod for great ideas, and doing that would help him help both himself and everyone else even more.