gluing tinsel to your crown

many are the county road maintenance supervisors vexed by michigan's infamous freeze-thaw cycles; this worm appears to have received the coldest end of this particular stick. but relief is in sight, as the equinox wraps around us.

a week ago sunday i rode the follis to fleming creek, to parker mill that is, to see if all that melting snow had flooded the portion of the hoyt post boardwalk that passes underneath the old michigan central railroad bridge. judging from the creek's stage in the park, the flooding was indeed taking place, but naturally i had to see for myself, so i walked the bicycle along the boardwalk because bicycles are not allowed, but i have walked the bicycle down the boardwalk before, on low-pedestrian-turnout days, and i feel that holds to the spirit of the rule, and anyway i have been coming down there for far too long to be kept away by the parks department. so along one stretch i got up and rolled along, and when i spied a couple of walkers ahead and through the trees i made to stop the bike and get off, but at that very moment came onto a wet patch and the bike slid out from under me, and while i caught myself fairly well with my hands i miscalculated the effect of the 4×4s along the edges of the boardwalk and hit one, squarely, with the lower part of my left rib cage. this was not something my rib cage was intended to do.

i got up right away, to act as though nothing had happened, and fortunately the couple of walkers did not engage me in conversation, because my gasping might have alarmed them. sure enough, the boardwalk was flooded, as this picture attests (note also the 4×4s along the edges):

so for the last week i have been in something of a cloud, intermittently pained. the injury was not serious, with no discoloration or swelling or anything ominous, so i have been simply taking it easy, or as easy as i can, which is never easy enough, but notably i haven't even ridden the bike since last wednesday, because i decided it was slowing down the healing process. i'm felling a lot better now (thank you), and yesterday i had a bit of a breaktrough: i actually sneezed. so i'm thinking thursday, back on the bike.

another aspect of that cloud has involved developing FORTRAN code for the numerical model we are developing to analyze surge issues in a large combined sewage storage tunnel that is being designed for a city somewhere. the model itself was developed at u-m as part of a doctoral thesis by a very hard-working young man from brazil, who is currently teaching at the university in brasilia but is also working with us to further develop his model and apply it to this very visible project. actually it is rather exciting for us to be doing this, because we had to convince the project owner to have us do this innovative thing rather than go with the "experts" who have done all this surge modeling in the past, with models that aren't really vetted and can't necessarily answer all the right questions. it has been something of a personal struggle for me to define a role in the project, but i have hung on and when it came up that the code could be ported from delphi-pascal to fortran, for speedier execution, i spoke up and was allowed to run with it.

so the cloud involved a problem with continuity, with volume disappearing throughout the trial runs. my recoding bascially ran, but somehow the volume of inflow was continuously vanishing even though the velocity and cross sectional areas were there, just going off into the ether or attaching itself to some string with a loop, or some parallel universe or similar science fiction. it was driving me crazy, until yesterday when i finally figured out the problem. i had mistakenly declared a vector of indices as floating point, which produced one of those classic moments in computer programming: give the computer ambiguous instructions, and you can never be sure what the result will be.

so that's fixed, but i am having other issues with stability associated with the flow regime transition, which of course is the hardest part of the problem. but if i can get this licked, we'll all be rich and famous. when my royalty check comes, i think i will buy a mustang. no, that's not what i will do. i think what i will do is buy a corvette. no...