Lost

I'm trying to decide if the persistent telegraphing of the "twist endings" on Lost is intentional. I think it has to be. I don't read spoilers and don't obsessively rewind and freeze frame with the TiVo, but I still rarely find myself surprised by the revelations. Wasn't it evident about a minute into "Walkabout"'s first flashback that Locke was handicapped? Every shot was framed to accentuate Locke's powerlessness, not to mention that he was never standing up. Was there any doubt that Sawyer was going to kill the wrong man in "Outlaws"? If he kills the right man, there's no emotional turmoil, and no surprise, and Lost is never without both.

Kate's doomed childhood friend... Locke's kidney stealing father... Desmond in the hatch... all obvious well before the "payoff".

Part of it must be Hitchcock's "Bomb Theory": if the audience knows what to expect, and the characters don't, the result is suspense. But that doesn't really explain spending so much time building up to the reveal of Locke's handicap. Maybe it was more subtle than I give it credit for, but it appears that Lost could be accidentally underestimating its audience's intelligence, or deliberately taking advantage of the typical viewer's desire to feel clever.

Of course, Lost could be brilliantly getting the best of both worlds: casual viewers are surprised, and regular viewers feel self-satisfyingly clever. The omnipresent "Easter eggs" of pre-island character connections (Hurley on Jin's TV, Locke's box company), and the inclusion of some "surprises", like Jack's future wife's miraculous recovery, that aren't surprises if you've been watching regularly, support this interpretation.

It's a tough balancing act: keep the twists from seeming like they come out of nowhere, but don't make them so obvious that the audience spends the whole episode anticipating the ending, ruining the suspense. Frequently (particularly in the case of the flashbacks), I think they err too far towards the latter.

This is all by way of praising with faint damnation. Lost is terrifically entertaining. The premiere was surprisingly satisfying, especially given the tendency of many shows to "wimp out" after building up to a suspenseful finale, premiering with:

• a flashback episode - Battlestar Galactica almost went this route for season 2, and much to showrunner Ron Moore's credit, they recognized it was a terrible idea.


• a flashforward episode - The West Wing began heading toward the shark jump in season 3, when the premiere skipped ahead to months after the MS revelation, which something like the last 12 episodes of season 2 had intimated would be a huge ordeal. I think ultimately the payoff was one congressional hearing that focused on Leo.


• a left-field episode - South Park becomes Terrance and Phillip...
  

Lost opened the hatch in the finale, and showed us what was in it in the premiere - and better still it wasn't another hatch, an empty room, or some other clichéd metaphor. Even though all of television history suggests the writers/producers don't have any idea where they're going (paging Chris Carter), so far everything has fit together so well that I have some faith there's a story being told here. Maybe I'll theorize in a later post.

Good advice for Microsoft

As it struggles to compete for mindshare and marketshare with Google and Linux, Microsoft would do well to take this anonymous minimsft poster's advice:

Microsoft is way too focused on building the next billion-dollar business; there is no way for a team to start something that will be a great 50 million dollar business even with great profit margins. Why not create 100 teams like that? Some of them (with no way to predict which) WILL turn into billion dollar businesses. But if you don't let them start they never will.

Read the whole thing.

There's actually a plethora of insightful comments (and the expected sour grapes) on the anonymous-MS-employee minimsft blog.

I actually interned for Microsoft in the Windows product group in Summer 2001 (working up to the XP release). The top-down direction criticized by the poster was very much in evidence. It baffles me why an organization which prides itself on hiring amazingly smart people would put a management structure into place that actively discourages small, exciting projects. Instead, the review-goals-uber-alles mindset means that if you're not polishing a spec for a dialog box or oiling a cog in a compatibility database, you're hurting yourself professionally.

At Google, it appears, a great idea can go from a glimmer in an employee's eye to a beta system used by millions in a matter of weeks. There's no reason Microsoft couldn't have the equivalent of Google Labs. If they don't start encouraging individual innovation, they're going to see those most interested in innovation go elsewhere.

Python

I'm finally starting to do some real work in Python. There are some small annoyances, like the weird nodelimiterlowercase built-in names, but generators are (a carryover from Icon, I believe) are terrifically useful. I'm surprised something like them didn't show up in other scripting languages earlier, since they're naturally suited to the sort of source/filter/sink piping you see often in scripts.

In this particular case, I'm using a generator to produce specification objects one at a time from Perforce, which conveniently gives you an option to get your command results in the form of marshalled Python objects. Generators let me open up one connection to Perforce for a whole set of these, and return them one at a time as requested. The generator "owns" the network connection, and keeps it up while waiting for future requests. Poor man's lazy evaluation!

I can't believe...

... that I missed Grilled Cheese Monday on Slashfood!

At least now I know what I want for Christmas. Ought to be perfect for making the delectable mozzerella, fig jam and prosciutto sandwich.