An evening of careful thought and successful execution, or weekend debugging in the physical world

*Not* bridge, though that was very fun, and we played for about 12 hours, and I had some successful bidding play and some successful hand play (though never at once), and one 11-card fit, and a few jumps to game. And in the end Naath and I beat Peter and Ian by about a thousand points due to Naath's incredible skill with a dealing sheet. And I now feel (despite not being 100% with simple bidding) that I've got the hang of when one would like to slam search, and would be pleased to play splinter/cue bid and RKB conventions next time.

I was in fact referring to my coat zip. This experiences a discontinuity in working in Pearl Close (don't ask). Specifically, the lower zip and upper zip were both zipped, and with effort could be moved up and down, but would not meet, and hence the upper zip could not be detached and the coat not (annoyingly) put on or (more annoyingly) taken off.

The zipping between the zips was fragile and easily dissociated, though both zips did zip the sides together or apart when moving up and down, just that the newly zipped area could be unravelled. The lower zip could be returned to the bottom and unhitched. Hence one could zip the top zip to the top (closing the collar), the lower zip to the bottom and detaching it, and manually unzipping the length in-between, temporarily allowing the coat to be donned or un-donned a la a jumper.

This got me to a warm inside place where I could experiment properly.

The problem was the zips were out of alignment. When they were brought together, there were more zip-links between them on one the right side of the zip than the left, and so they couldn't meet. How could that come to be?

With some experimentation I realised that it could be undone only the way it must have been done in the first place, that the bottom of the left side of the zip when being attached to the lower zip could be pulled through further or less far before the zip engaged, and thus the position of the zip on the left, relative to that on the right, tuned or broken. I had been instinctively aligning the lower zip correctly, but must originally have aligned it a bit wrong, putting the upper zip in the corresponding configuration I had not yet brought the lower zip to meet.

That gave the solution immediately. Pull the upper zip to nearly at the lower zip. Pull the left-side of the zip through the lower zip until the distance between the zips on both zip-lines was the same. Zip the lower zip to the upper zip, or vice versa, until they meet, bring them both to the bottom, and then detacth the left side of the zip from both. Tada. Time for bridge.

In retrospect, I might have been able to bring the zips together at the bottom (bringing the top zip down both zip lines) with the lower zip unzipped, and then been able to detach the lower left side from the top zip directly, I'm not sure.

However, it just amused me that exactly the same sort of "experiment, experiment, experiment, think, think, think, it works" that's so satisfying in debugging works here too.

Cambridge hugs

I still can't get used to people spontaneously coming up to you and hugging you for saying something unbearably pedantic and drawn out. I can't even remember what I said now. It's just so lovely and so very Cambridge.
Thank you, friends.

*hugs*

Matrix: Second order Tensor (some spoilers)

I wonder, maybe "revolutions" is a reference to tensors? Nah, too much to hope for.

Anyway, I rewatched Matrix II (I really can't remember what it was called) and decided that on the whole it *was* pretty good after all. It had some weaknesses and didn't live up to the first, but wasn't *bad*.

Things that are just cool:

* Morpheus, Neo and Trinity walking about in their black leather looking ominous
* Agent Smith
* Neo fighting fifty agent Smiths
* Neo flying
* The virus twins
* The Merrovingian, both the idea of a powerful rogue program, and the petulant French incarnation.
* Neo blithly fighting a dozen Merrovingian minions.
* Neo being repeatedly baffled by doors which close and open onto a mountain temple. Apart from making the absolute most out of Keanu's best acting ability -- going "Whoah!" -- it's a perfect impediment. Neo has a lot of skill and can brute force most problems. But M has lots of experience, so it makes sense he can do something which Neo doesn't know how to defeat, but can circumvent by flying really fast and/or being bad-ass.
* The idea that the freeway is suicide. Of course it's scary -- you're always less than two meters away from a potential agent with a two-ton blunt object under his control¹.
* Some of the freeway fight is pretty cool.
* The idea that Neo must find a well-but-not-perfectly-protected door to the machine mainframe, and the three crews can just achieve this with a sophisticated combination of hacking and motorcycles.
* The ending. Despite being stupid in resolution, and being too ambiguous with still being in the matrix, Neo suddenly freezing a lot of sentinels is emotionally impressive and miracle-like.

( 'Things that aren't stupid despite people saying they are' and 'Things that aren't very good, or are kind of gratuitous or drawn out' )

Finally, the philosophy. This is another "Thing that isn't very good, or is kind of gratuitous or drawn out", but sufficiently interesting I put it by itself.

The first film was cool. It took the philosophical concept of a cartesian demon (roughly, the question "how do we know we aren't living in a matrix?") posed by Descartes, and made it easily accessible to everyone. It did it so well most people can't believe it needed explaining, can't believe it ever wasn't obvious to them.

(Note, I'm not saying it explored all the possible ramifications of that. But it did something that needed doing, and did it very very well.)

The second I think tried to do the same for predestination. Well, it gets credit for trying. I think I'm biased because I think I think the point it's trying to make is stupid, but I don't think it did it very well, either. AFAICT their world has no reason to be more predestined than hours (making the Oracle a sort of cheat). The first film worked because the whole idea explained the concept, but here it was tacked on, and their "insights" into predestination didn't have anything to back them up, so were just vapid.

[1] Where *are* the female agents?

Splinter bids

A brief explanation of splinter bids. First, the problem to which they are the answer.

If we agree a good trump fit in the first or second bid, and someone knows we have more than enough strength for game, can they indicate the potential for slam?

Eg. If I have five losers and you have seven, and I open 1S, and you have a fit, I would hope you jump to game. We have a fit, we have 14 losers, no more communication can help. Then, because I'm stronger, I could call Blackwood or whatever.

However, if you open 1S, then I know we have a fit, and maybe more. But if I bid 4S you will pass, and if I bid 2S or 3S, you will pass (as they mean I have nine or eight losers, and would ask you to bid game if you were better than 7). How can I say "Game in spades. But investigate slam."

If you (as I until recently) had not experienced this problem a lot, and cannot reliably bid to a makable game, this is a solution to a problem you don't have. Ignore it. Be satisfied to find the game, and let the slam be down to luck. Don't try to cram too much into your head at once, or you'll forget the stuff useful for 80% of the time in favour of that useful much less often.

Of course, it's still useful to be aware of it.

For the details of when to do so and what it means, see a wiki page, I don't reliably know it. But the basic idea is, A double-jump to the four level in a new suit says "Game in what you said, but also I have a singleton or void in this suit."

Partner can then see if his hand looks better or worse knowing you have first or second round control in that suit. If he has A-x-x-x (or x-x-x-x) then it's rosy -- those x's can probably be ruffed immediately. And if the ace isn't there, that's only one loser which was already counted. If he has K-Q-x or worse, it's not so useful -- the honours aren't very useful, and represent the loser count is probably optimistic.

Partner can then either cue-bid (eg. splinter 4c, response 4d saying "And I have Ace/void in dimonds", is looking good), bid blackwood, or sign-off in game of the original suit. Of course, then the splinter-bidder can *then* take control of the auction and bid blackwood if he so chooses, though of course he could have done so immediately instead of splintering if he'd wanted to, and is ignoring his partner's qualms.

Note, please look up when this is appropriate. I would say something like "a new suit either jumped to the four level, or double-jumped," is probably a splinter since it can't mean anything else. However, I wouldn't know if there were interference, or an argument about suit first. Check the convention sheet.