Josh on Mosh

When is science art, and art, science? It is metaphysical or metadata? Josh moshes the meta without harshing the mellow:

A Stephen Mosher post at Watts Up

From Josh at

15 thoughts on “Josh on Mosh

  1. That is just brilliant. Love it. I will not have to do an expose on this cartoonist.

  2. Josh,
    An artist can be viewed as a person projecting their values onto a selective recreation of reality.
    To be an artist is a courageous act. It is through the vehicle of the artistic process that we can see ourselves externally manifested. The risk is one may not like what they see about themselves.

  3. Josh portrays the artist with eyes so narrow he cannot see. I think that’s totally appropriate. (Or are they his eyebrows and he has no eyes at all?)
    Either way, it matters not.

  4. Thus is OT, but it was interesting that a Colorado campaign ad appeared in “Ads by Google” under this piece. The political parties must really be grasping at straws this late in the game.

  5. I like it.
    Also, Mosh, even if our overall measure is reasonably good, what about the fact that thermometers in general are at least plus or minus one degree in accuracy, stability, and repeatability. Doesn’t that obviate the argument regarding the apparent change meaning anything?

  6. Don’t know why anyone in the field should balk at calling himself a “numerical artist”. It was politically incorrect long ago to be an English teacher – had to be a “language arts” teacher. Math and sciences have some serious catching up to do.

  7. Lonnie.
    1. The accuracy number you cite is not exactly correct. But
    2. even if the error of all daily measures had a standard deviation of 1, you are helped of course by the LLN. when computing a monthly average if you take 30 measures
    with a mean of zero and SD of 1….
    J summary(J)
    Min. 1st Qu. Median Mean 3rd Qu. Max.
    -1.42700 -0.65000 -0.04219 0.03755 0.75110 1.92100
    really big errors?
    > J summary(J)
    Min. 1st Qu. Median Mean 3rd Qu. Max.
    -13.8200 -3.6870 -0.4405 1.5550 6.9460 23.3900
    Now take 1000’s of stations and 1000’s of months.
    Basically there is nothing to the argument about “accuracy” of thermometers.
    there is also nothing to the argument about the number of thermometers.
    there is nothing to the argument about the march of thermometers.
    There IS something to the argument about the categorization of stations into rural and urban.
    That drives the whole analysis.
    Can’t say anymore, as my lips are sealed.

  8. arrg Lonnie, some R code dropped out of my post.
    basically generating random variates. with SD =1 and sd =10

Comments are closed.