I need VLOOKUP to tell the difference between a zero and a blank because MAX, MIN and AVERAGE count zeros but not blanks. I'm also plotting the same looked up data, and I could do for the graph to leave a gap where there are blank cells too, instead of joining the ends with a line. Simply replacing zeros with "" won't work, because the data contains zeros. At the moment the cells that would be blank have been filled with NA, but AVERAGE won't put up with that either.
I'm plotting data on a graph that has an x axis that's continuous, but the data in the table is discontinuous, with gaps in it. (There are nearly half a million lines in the table, and the graph needs to be plotted using samples selected from it at random, so anything that involves any manual faffing around is out of the question.)
About 13 years ago (so my recollection of the details is a bit hazy) I inflated a cycle tyre quite rapidly after mending a puncture using a Zefal plastic pump similar to this, which has a small threaded brass bush for the hose moulded into the end of the cylinder.
Whilst I don't recall the end of the pump getting any hotter than usual, on this occasion the brass bush got hot enough to soften the plastic until the pressure (~80-90 psi) pushed the bush out of the end of the cyclinder. What intrigues me is why the heat was so localised, if the whole end of the pump had softened to that degree it would have inflated like a balloon and burst. The outer surface of the bush was heavily knurled to give it a key, so it wasn't going to come out lightly without the plastic softening.
Zefal describe the plastic as a 'technopolymer resin', so there's no prospect of looking up the melting point, but some plastics have an upper 'useable' temperature below 100C. It seems possible that the heat was getting funnelled toward the outlet along with the air, and it must be travelling at quite a high speed, so 'wind warming' would be quite significant. Pressing a finger over the outlet, I can burn it in a couple of strokes of the pump without the cyclinder end even getting warm.
If you look closely at this building you will see that the lightning conductor runs horizontally in a ring around the base of the roof just above the gutter. There are numerous strips coming down to it from the top of the roof, but none connecting it to the ground.
I've lost count of the number of times I've seen them playing around with stunts like this, why the hell aren't they fazed about thousands of water droplets floating around in the air and getting into the electronics?
Having seen ships in dry dock from time to time, and been struck by how precarious the supports looked, I'm having difficulty in being surprised at the "capsize" of the Petrel.
The system I recall seeing has the keel, and I presume the majority of the weight, supported on blocks on the floor, with props at the sides holding the ship upright. With the ship balanced on the keel and minimal load on the props, since they were wedged and in compression only it would take relatively little force from a gust of wind to momentarily release the compression at the windward side, at which point all the props fall out.
It looks like there's about 900m² of windage, which would give about 20 tons at 45mph, perhaps that's enough if the props were put in carelessly, especially if there are repeated nudges from repeated gusts over a long period of time.
On Radio 4 this morning someone was saying that in the days of B&W TV only 9% of the population dreamt in colour, but now in an era of colour TV the ratio is reversed. That reminded me of a news story a few years ago in which a blind man who had just been given an operation to restore his sight said that he was now dreaming in colour instead of B&W.
At the time, that set me wondering about whether I dream in colour or B&W, and after much thought I realised it's neither. My dreams are like memory, they occur in terms of salient points: if I think of a rectangle or a circle they're not in in colour or B&W, they just 'are', unless I think of a no enty sign, in which case I see a red circle and a white rectangle.
In the most recent dream I recall, I was talking to someone. I don't know who I was talking to, what they said, or what they looked like, but they were trying to show me something, and I don't know what they were trying to show me either, I just recall telling them that I couldn't see it because I was having an ocular migraine.
Then I woke up and found I was having an ocular migraine. That's all I rememeber/saw, because that's all that was salient.
The pressure to which my BP monitor inflates the cuff varies widely. It would seem reasonable to suppose it would inflate until the pressure exceeds systolic by a sufficient margin, but if you look below, the data is grouped into two distinct clusters pretty much independent of systolic bp.
Is it inflating for a preset time? No. Is it inflating until it exceeds sys bp for a predetermined time? No. Is it inflating until it exceeds sys bp for a predetermined number of pulses? No. Is inflation pressure dependent on how tightly the cuff is wrapped? No.
It occurs to me that if oxygen consumption is directly proportional to energy metabolised, then water produced must also be. Wikipedia just gives typical levels, not the relationship to metabolic rate.
Referring to this page, there are 3782 cyclists at fault for failing to look properly, and 7565 non-cyclists. If we assume that there are 8 times as many non-cycling road users as cyclists, does it follow that:
1. Cyclists are four times more likely to be at fault because they’re producing half as many incidents from 8 times fewer vehicles. 2. Cyclists are 16 times less likely to be at fault because non-cyclists are still producing twice the incidents despite having 8 times fewer potential targets for careless road users to collide with. 3. From the product of 1&2 above, cyclists are four times less likely to be at fault overall. 4. Cyclists are just half as likely to be at fault, because those who haven’t crashed are not relevant. 5. None of the above.
Personally, I’m inclined to think it’s (3), because the probability of an accident must be the probability of finding a careless road user multiplied by the probability of them finding something to hit.
This seems interesting. I'm assuming that a high proportion of African deaths are going unrecorded, and that African countries without the resources to collect death stats also have fewer resources to administer vaccines.