Thursday, February 26, 2009

Project team blogs

I'm thinking about setting up a Bazaar group blog, separate from As of Friday the 27th, a site is up, in the sense that you can read it, but not yet announced (beyond this article). It's still a bit of an experiment.

Within one person's personal syndicated chronological publishing (ie "blogging", broadly), there are different strains. The tension towards those different strains may be one reason why people have tended to go quiet, or to feel a sudden agoraphobia at how widely their person thoughts are read or personal photos reproduced.

It's more subtle than a binary private/public switch, and a simple password or even openid is not enough. It's more than technical.

At the moment there's a proliferation of different web-based tools in use: twitter,, facebook, flickr, personally-run blogs, dopplr, planets. It's not just that they're just technically imperfect that's causing the fragmentation (though repeatedly getting semi-spammed invites is tedious), but also that they provide genuinely different forums. There are some things that are not secret but personal and more appropriately shared with people you know; some that are personal opinions but that you're happy to share with anyone; some that are about projects like Bazaar that are personal but that are also bigger than just me.

A project team blog seems too to becoming one of the channels that people expect to have.

People tend to raise the question of whether this will just dilute the same amount of writing across multiple channels. It might, and there does seem to be a critical level of activity for a blog beyond which it's not alive. On the other hand, now that there's more syndication that level may be lower: infrequent posts will still pop up. But I also suspect that creating a place where a particular type of content feels really at home will create positive feedback.

For instance, Gary van der Merwe just made a nice improvement to the revision selector control in qbzr, using the layout originally invented by Scott for bzr-gtk. I like this, and I'd like to express that approbation in public but I don't want my sourcefrog blog mostly occupied by neat bzr features because I have other things to say.

Sunday, February 15, 2009

recent readings

Hal Abelson and Jerry Sussman with Julie Sussman, Structure and Interpretation of Computer Programs chapter 4, with Mary and friends. Like The Little Schemer, deeper than it seems.

Alan Greenspan, The Age of Turbulence: Adventures in a New World. Not exactly deathless prose, but a pretty interesting perspective on some aspects of recent history. He's certainly been around a lot of it.

Arbinger, Leadership and Self Deception. Corny style, but an interesting point that people put them selves to a lot of emotional and actual trouble to avoid the cognitive dissonance of not living up to their conscience.

Armstrong, Programming Erlang. It's good; it deserves a longer post. I'm kicking myself that I didn't twig to what was good about Erlang when Luke raved about it probably ten years ago.

Jose Saramago, Blindness. It seems interesting but slow, and I stopped about half way. I'll try again later.

Mark Z. Danielewski, House of Leaves. Gripping, reawakened a particular type of book-appetite. Requires a fair amount of extended concentration - not the sort of book that you can pick up and read just a couple of pages.

Edleson, Value Averaging. A good research paper padded out to an only OK book, by adding several extended walkthroughs of different situations. Maybe I'm just grumpy because Kinokuniya sold it shrink-wrapped, I half suspect so that you couldn't see discover this before purchasing.

David Lebedoff, The Same Man: George Orwell and Evelyn Waugh in Love and War. A lovely dual biography starting from the conceit that Orwell and Waugh while very different were also much the same. It's inspired me to read much more from both of them this year, starting with Down and Out in Paris and London which was itself excellent.

Chris Okasaki, Purely Functional Data Structures. Just started, looking good.

J.R.R. Tolkein, The Lord of the Rings. Re-reading it for the fourth time or so, and seeing some new parts on each visit.

Tim Keller, The Reason for God. Not bad, and less ranty than The God Delusion, but in the first half showing an unworthy predilection for straw men. We get "Stalin and Hitler were atheists!" on page 5 (!!) but it's mostly uphill from there.

On a Sony PRS-505, using only free content and (non-embedded) software:

Adam Smith, An Enquiry into the Nature and Causes of the Wealth of Nations - very readable and interesting; the only caveat being that there are sometimes a few pages of text when (to the modern reader) a table or graph would be better.

Anthony Trollope, The Warden. Easy to read and enjoyable; I'll definitely read more of his work.

Mark Twain, Christian Science. Picked pretty much at random from PG as one I hadn't read or heard of. Not particularly noteworthy or recommended, except that reading a minor work gives a broader perspective of the author.

I tempted pitti into buying one of these and then packaging the Calibre software for maintaining the e-Book reader. It makes an excellent interface for reading free content such as Gutenberg texts that would otherwise have to be printed out.

To be continued...

Friday, February 06, 2009

mailing lists and hot tubs

Via Mary:

Email listservs often parallel in person group growth patterns and grow very fast, too fast. Sometimes this will lead to a situation where pleas to the list have no effect and the list is in danger of degrading into flames and lots of useless noise.

Here's a proven way... to get a list back on its feet and back to its core mission and people.