« November 2005 | Main | January 2006 »

December 31, 2005

I'm a news junky (but I don't read the paper)

I tied Phil on the Salt Lake Tribune Trivia test. I don't read the paper except Sunday, so shows how effectively RSS replaces the newspaper.

Posted by pete at 12:32 AM

December 30, 2005

Goal for 2006: Get(ting) things Done

I first heard about Getting Things Done (called GTD by it's fans) when Mark and Mike kept mentioning it in their podcasts on Manager Tools. I didn't catch it the first couple of times (I thought they were talking about "getting things done"--as in, doing stuff). It was obvious it was such a natural part of the way they managed (they seemed to assume that any good manager knew about it), they didn't mention the book specifically until the 5th or 6th podcast. I picked it up immediately, I had to see what it was all about.

Everyone's had some (probably lots) "time-management" training. I started with a Day-Timer in college, then switched to Franklin after a few jobs (because that's what the company bought, plus everyone used the Franklin system), and of course I read all about Covey's 7 Habits. Every one of these inspired a flurry of self-assessment and self-organization, followed shortly by a realization that my (dis)organizational reality somehow did not fit the one I had just learned. And I was back to lists in my head, in Notepad.exe, and on scraps of papers.

GTD starts out with some unusual (at least Covey and Franklin and those types would think) perspectives on time-management (which GTD calls action-management):

I've just finished reading through the book once, and I have a pretty good feel for the system. My next reading will be as I implement the system, using the book as a reference manual.

If you're still struggling to organize yourself, or you feel your current personal-management system is not comprehensive or ineffective, or you just want to refine the way you manage things, or even better, you want to experience stress-free productivity, I recommend you get this book.

For other people's opinions, check out A Venture Forth, 43folders (1) and (2), OfficeZealot, Matt Vance, GTD with your Mac, Wikipedia, Shu-Ha-Ri and GTD. (And there's a whole lot more).

Posted by pete at 11:02 PM | Comments (1)

last.fm takes a dive

Looks like the folks over at last.fm are having a pretty bad day (about to be two bad days):

We are in fact the least lucky web site on the internet.
We've suffered some database issues;
please bear with us while we try to piece everything back together. Again.

Was well on my way to 300 songs when they went down. Fortunately iScrobbler (the plug-in for iTunes) will queue up my listening activity until they get things put back together.

Posted by pete at 10:50 PM

December 29, 2005

C programmers aren't relics (anymore)

I have often felt like some kind of relic because I enjoy coding in C. For various reasons, I never got excited about Java. It just didn't feel ... hacker enough (plus by that point I wasn't doing programming so much anymore).

Robert Scoble blogs about Joel Spolsky's essay The Perils of JavaSchools, then writes about the increasing demand for (and sparse supply of) C/C++ programmers:

I've heard the same kind of thing repeated around halls at Microsoft. Almost every team I interview with my camcorder says they can't find enough C or C++ programmers to get their stuff done. Some on very exciting teams with hundreds of millions of users. Some that, gasp, actually have budget to hire real programmers. And, this isn't just a US problem. The problem exists at our offices around the world. Every team I talk with says they wish they could hire more hard-core programmers.

This is probably driven (at least in part) by the increasing difficulty (= cost) of adding more CPU's to data centers (wish I could remember where I saw a blog post just recently about a company struggling to scale beyond the city-mandated maximum for power to their data center building). The economic model of Java (and many other programming-efficient/CPU-intensive languages) assumes programmer time is more scarce/expensive than processing (computer) resources ("you can always add another processor"). But that assumption is becoming less-true as the cost of programmers decreases (Thanks! India/China/Russia/Phillipines) and data-center costs increase (power, HVAC, physical management, etc).

It must be compelling to hire a C/C++ programmer to squeeze another 20-50% performance out of an existing farm of 10,000 Java-based server instead of adding another 2,000-5,000 servers. When your minimum starting data-center is 100,000 sq. ft and you've got many around the world, an optimizing programmer (team) is an easy investment to justify.

Looks like I could have a second career in programming, going back to my roots. I've always loved optimizing (of any kind). And none more than some tight C code with all sorts of nasty pointer fiddling.

Posted by pete at 5:47 PM

December 28, 2005

Creating a renaissance in enterprise software in 2006

David Heinemeier Hansson predicts that in 2006

Enterprise will follow legacy to become a common insult among software creators and users.

Enterprise software vendors' costs will continue to rise while the quality of their software continues to drop. There will be a revolt by the people who use the software (they want simple, slim, easy to use tools) against the people who buy the software (they want a fat feature list that's dressed to impress).

In a way this is a self-serving prediction, as David and 37signals are certainly doing their part to make it happen. But who'd argue with Peter Drucker?

Why not expand this same renaissance beyond the 37signal products (though I'm sure they'd love every enterprise to dump BigCo software for Basecamp or Backpack), by adopting the 37signals philosophy of product development and teamwork.

I'd love to create my own "Getting Real/Building of Basecamp" experience where I work. I think I'll make that one of my goals for 2006.

Posted by pete at 3:38 PM

December 27, 2005

Last.fm out-Apples iTunes recommendations

One of the reasons I started using iTunes--besides to support my iPod--was because of the Amazon-like recommendations ("people who liked this also bought"). It was a good start, but really hasn't gone where it could have (the Just for You feature in iTunes 6 is a lame front-end for the recommendations that were already there).

The iTunes recommendations are weak for a couple of reasons. They are based only/primarily on what I've bought instead of what I listen to, and only on purchases made through iTMS. They seem more motivated by solving Apples problem (they want me to buy something, anything) than my problem, which is to find something I want to buy.

What I could never figure out is why Apple doesn't integrate all of the data iTunes collects (including from my iPod) on what I do. It is even more incredulous since Apple builds all of these pieces (iPod, iTunes and iTMS, and even the underlying glue of AppleScript and Mac OS X), but hasn't made the comparitively trivial step to hook them together to give me music recommendations based on my listening habits and ratings.

Last.fm (thanks Alex and Ed) has made the link, and added some unexpected features. To get started, you download software (Windows or Mac) that plugs in to iTunes and integrates your iTunes data to your last.fm account. This data is (optionally) shared to come up with recommendations based on what other last.fm users listen to. Once you listen to 300 songs, last.fm will start generating personalized recommendations. Pretty cool.

But it gets better. You also download a last.fm streaming client, which lets you listen to your own personalized recommendations like a radio station. Plus, you can explore and listen to any music, again based on finding music similar to your own tastes.

I haven't made it to 300 songs yet, so I haven't seen the full feature-set last.fm offers. But what I've seen so far is a huge improvement over iTunes recommendations. I hope last.fm will be successful, but would also be happy if Apple incorporated some of these features into iTunes.

Posted by pete at 3:21 PM

December 26, 2005

More free airport wi-fi than you think

Robert Scoble laments the lack of free wi-fi in airports, and points to this list of (mostly second-tier) airports with free wi-fi.

Not to split hairs, but in my experience quite a few airports (and hotels and coffee shops) have somewhat free wireless. Yes you have to pay for TCP and UDP wireless, but plenty generously offer free ICMP service (I assume they figure that once you get hooked on pinging things, you'll want to upgrade to more-useful for-pay UDP/TCP services). If only there was a way to tap the unused potential of ICMP...

Turns out not only is that entirely possible, it's already been done. Daniel Stødle (btw, is Scoble actually spelled Scøble?) found a way to fully exploit ICMP service wherever it is offered. PingTunnel will let you tunnel TCP over ICMP using an intermediate proxy (which looks to the wireless network like you're pinging the heck out of your proxy). Takes a bit of setting-up, but once you get it going, it's surprisingly functional. Turns out that free ICMP service is quite useful after all.

So next time you find yourself in an airport, check first whether they offer free ICMP wireless service. If that doesn't work, then you can start grumbling (or head to an airport hotel to test their wireless service).

Posted by pete at 7:42 PM

golf tips