Off to Portland

I'm heading off to the O'Reilly Open Source convention, where I give a half-day tutorial as well as a regular session, and a book-signing (which the publication schedule just missed last year). Sporadic updates here depending on availability of time and connection. -m

Developer Declaration of Independence

Read it, sign it. http://www.opengroup.org/declaration/declaration.htm -m

XHTML 2.0 draft released

The sixth Working Draft of XHTML is out. Come and get it while it's hot:http://www.w3.org/TR/2004/WD-xhtml2-20040722/ -m

Colloquy

I'm trying out a nifty new IRC client for OS X. Strong emphasis on good UI and following OS X conventions. And a .info domain. What's not to like?http://www.colloquy.info -m

Zoot turns 4.0

Huge congratulations to Zoot software, where their namesake flagship product turned 4.0 last week. http://www.zootsoftware.com/ -m

Too bad to read

Once I pick up a book, I always try to read it all the way through, even if I don't care for it much. It keeps me from retreating fully into a tiny comfort-zone of reading materials. But this book crossed the line.

The cover looks like this--tiny print at the top: "From the #1 New York Times Bestselling Phenomenon". Top third of cover: "Tom. Clancy's". Middle of cover, medium font: "Net Force" Then a golden Mercator-projection map over some blocky pixels, then a medium font again at the bottom: "CyberNation". At the bottom, in the miniscule font again: "Created by Tom Clancy and Steve Pieczenik" "Written by Steve Perry".

All the warning signs were there. At least I only paid $.50 for it at a garage sale. -m

Dramatica

I'm experimenting with a piece of software called Dramatica. It's a "story engine" based on the work of Melanie Anne Phillips & Chris Huntley to develop a framework for what makes a complete, satisfying story.

The 300-page book describing the theory is freely available online at http://www.dramatica.com/theory/theory_book/dtb.html, as well as in a typically-awkward PDF format.

Many writers rail against these kinds of systems, calling them too constraining. Those who say that are probably right for them. I, however, am a freak for metrics. Any reasonable way to measure a story is helpful. Even if I don't agree with all of the underlying theory, I can at least stretch out a story over the framework and see what lines up and what doesn't.

A great feature is a built-in database of many common books, plays, and movies. As you look at individual pieces, you can compare how familiar stories line up.

A free demo version (save and export disabled) is at http://www.screenplay.com/support/demos/index.html. Also, a lighter version called Writer's Dreamkit is available. It has 80% of the functionality for around $35. Hit the Amazon link on the web page for more. -m

Daddy's Back...

Powerbook back from the shop; brand spankin new hard drive. After a massive reinstall-fest, I'm back in operation. If you're reading this, that's proof everything's working. :) -m

Hard drive failure

For future googlers, here's some signs of hard drive failure on OS X, on a PowerBook.

Early Signs: (start backing up at least weekly) Intermittent rainbow disks, freezing most, but not all system activity, especially for disk intensive apps like iTunes. Mouse works. Cmd-tab works, little else. S.M.A.R.T. status reads OK. If you notice this, try to keep Utilities -> Console running most of the time. Try to read what it prints during the freezes. A verbose mode boot, (hold Cmd+V( might reveal repeating error messages, like (disk 0s5: 0x8 (UNDEFINED)) Reformatting drive might relieve symptoms for a month or two.

Intermediate signs: (start backing up at least daily) All the above symptoms, but worse. Possibly lengthy boot times. Tools like DiskWarrior might fail to rebuild your directories. Reformatting might buy you another month or two. Console log contains messages like this:

PowerBook kernel: IOHIDSystem: postEvent LLEventQueue overflow

PowerBook last message repeated xx times

PowerBook kernel: IOHIDSystem: postEvent LLEvw

Danger Signs: (start backing up at least hourly) Freezes with increasing frequency, even without any significant running apps, requiring hard power-off. Console messages contain this:

PowerBook kernel: IOATAController device blocking bus

PowerBook last message repeated xx times

PowerBook disk0s10: 0x3 (UNDEFINED)

At this level, S.M.A.R.T. status in Disk Utility or many other tools will say "failing".

This is all from actual documented experience. I've had Early Signs since January 24. The Apple Store Guy reformatted the drive and gave it back. I've had Intermediate Signs since March 27, except no good console captures. The Apple Store Guy reformatted my drive and gave it back. I've had some Danger Signs since May 22, but only a S.M.A.R.T. failure code since July 4.

From what I've read about S.M.A.R.T., it keeps all kinds of detailed information on your drive, varying by manufacturer, but including things like spin-up time, head height, number of remapped sectors, error counts, recalibration counts, and so on. The catch is, you don't get to actually see this information. All you see is filtered past some internal threshold, down to a single bit: "OK" or "failing".

Maybe this theshold is set a bit high, at least in consumer drives? Wouldn't it be nice to se the underlying data.

Anyway, back to the Apple Store Guy I go later today. -m

Chasing down a PowerBook lockup (repost)

This seems to describe it: http://www.neato.org/~page/virtmem/archives/000140.html Even like the May 21 comment, it only happens for me when iTunes is running. It'll play music for 4-8 seconds, then pause for 10-20 seconds. If I wait long enough, usually until the end of a song, the system comes back to life.

This is a TiBook 800, running 10.3.4, with nothing plugged into any port but power, and running no non-apple kernel extensions (as reported by kextstat | grep -v apple).

More detail here: http://www.macfixitforums.com/php/showflat.php?Cat=&Board=Forum35&Number=552714&page=2&view=collapsed&sb=5&o=&fpart=1#Post555561 where the problem is discussed in some detail. At least I'm not the only one.

My ominous theory: a failing hard drive is timing out IO requests, which is jamming up the event queue. How can I test this theory? -m

Contact

mdubinko@yahoo.com

Terms of use

For external use only. I doubt the enforcability of click-through licenses anyway. Copyright 2004 Micah Dubinko. All rights reserved.

Archives

Older stuff here