Dear Niantic,

Jul. 22nd, 2017 03:29 pm
green_knight: (Abandoned)
[personal profile] green_knight
We have received reports that Trainers haven't been able to collect their Defender bonus after the Gym update. We’ve investigated many of these reports and have not been able to reproduce any bugs related to this issue.

(as posted here) is not a good conflict resolution.

If you're unable to reproduce the bug, that just points to it being intermittent. Fair enough. Doesn't mean you should stop looking. However, you have the stats: you can compare pokemon activity and gym rewards, and if they don't match what they should be, you can fix.

Trainer of a Pokemon which stayed in the gym for 10 days, got fed a lot of berries, was kicked out this morning, and brought home 0 pokecoins.

It's a wrap!

Jul. 16th, 2017 07:46 pm
green_knight: (teh end)
[personal profile] green_knight
My WIP is no longer a WIP, it has graduated to 'finished first draft' and I am in that strange space where all the characters who have been taking up residence in my head have moved out, swept the floor, painted the walls in a neutral colour and are now looking into every drawer and under the bed to see whether they've forgotten anything.

274,696 words, including scene titles, placeholders, and 'the end', so call it 275K.

It will be either two or three books, though I am tending towards two, since there's a definite change of pace/location in the middle. This thing started as a comedy-of-manners, and was my go-to book for a while when I wanted something light and fluffy.

About the book and the writing of it )

And now it's half a day later and the book is still done: of course it will need a second draft, and I need to sort out the timeline, and I'd love to know how all of the loose threads will work out, and I am holding my breath just a little whether [redacted] will double-cross [redacted] but it's over, the characters have moved out, and while they might visit from time to time, the book. Is. Done.

After spending literally years with the compulsion to write down so many seemingly unimportant events in my protag's life (which all came together in unexpected ways), there is an empty space in my head now, and it feels weird. Other characters will move in - I have a fragment which isn't as complete as I thought it would be, so I'd like to write down the extra bits I know before finishing _something else_, but for now, I am WIP-less, and that's just a weird place to be.

Thanks for sharing your life with me, Firtal. I wish you all the best.

A surfeit of algorithms

Jul. 15th, 2017 08:55 pm
green_knight: (Eeek!)
[personal profile] green_knight
100 days, 100 algorithms

I'm somewhat in awe of this project - that's more algorithms than I could have listed, even with some research, and the discipline of implementing a new one every day for months is impressive.

However, it also shows a systemic weaknesses of programming: understanding the problem domain. I can't say much about the other 99 - some of them are algorithms I've never heard about, and at the very least I would need to make an effort to understand the python code and read it carefully, if not implementing the same thing in Swift, but this one, I spotted immediately:

Day 18; Monopoly

Some time ago a friend of mine asked me to help him with this problem.
Playing Monopoly, what is the probability that you step at position #24 during the first round?

Law of total probability says that the chance to step on certain position is sum of disjoint events of how we could get there. In this case, we get at #24 if we tossed 1 while standing at #23, or we tossed 2 while standing at #22, … or we tossed 6 while standing at #18. This leads to a recursive formula.

Any of my genteel readers who have ever played Monopoly will spot the most obvious problem here: You play Monopoly with two dice, so you can throw 2...12, so this is someone who hasn't done the most fundamental homework.

The less obvious problem is that you're trying to specifically solve _Monopoly_, rather than 'a board with x fields throwing 2D6'. Monopoly has a couple of extra rules: if you throw a double, you get to go again, but if you throw three doubles you go to jail (field 10); once you come out of jail you get another chance to land on field #24; and you have a chance to step on several fields where you may draw a card that moves you to a different field (named or 'three fields back'; IIRC that could even end your round!). In other words, the probability for 'step on field x' is partly determined by the dice, and partly by the game and its very specific rules; if you wanted to give an _precise_ answer, you'd have to calculate by how many routes you can reach each field including the 'go to jail' mechanism which gives all fields after 10 a higher probability and which means that there are ways of reaching #24 from every field between 2 and 35 (double-one, double-one, double 1...6, jail, and try again.) Heck, you could even go to jail several times until you run out of starting money, but if you get _both_ get out of jail free cards....

So, yeah.

This also illustrates why board games are not just the sum of straightforward probabilities: once a system becomes complex enough that you cannot simply do the rough calculations in your head, it becomes much more interesting, surprising, and, at a certain level (and given an appropriate mechanic), that rarest thing of all: a creator of narrative. It's no longer 'then I drew card x and rolled y on the dice' but 'so here I was, going about my business curing sick sheep and setting broken limbs when those pesky elves turned up right in front of me and–' (Terry Pratchett's The Witches. Brilliant short game for 2-4 players.)

Plus ça change

Jul. 13th, 2017 11:37 pm
green_knight: (fragile)
[personal profile] green_knight
mathematical models allow experiments to be run on environmental systems, and generate realistic output which can be used as the basis for rational and informed environmental management policies. That, at least, is the hope. In practice, the irrational side of human nature seems often to Coe to the fore, as in the reluctance of the United States government to accept the reality of global warming as induced by 'greenhouse' gas emissions, despite the repeated warnings uttered by the climate-modellers.

R.J. Hugget (1993): Modelling the Human Impact on Nature. Oxford, OUP. p. 20.

This is very much in line with my recollections of being taught climate change as an accepted scientific consensus in the early 1990s.


shalanna: (Default)

November 2012

25262728 2930 

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jul. 26th, 2017 10:26 am
Powered by Dreamwidth Studios