Back

@dougmerritt

When we moved from mostly analog clocks to mostly digital, I forced myself to still read digital readouts like "1:44" as "quarter of 2". Just because we see precision doesn't mean it's there.

These things that know how many of something in social media seem to be elusive and approximate in a way we need to be aware of. My number of upvotes on Reddit seems to wobble up and down faster than I'm willing to believe anyone is caring about what I write, and I'm convinced it's because I'm hitting different servers with different cache values at different times. The evil twin of 'eventually consistent' is 'give up on hope of consistency for now'.

So, I deferred to the master of disinformation (ChatGPT) and it insisted that there was only one answer to 3^6, but I told it that it needed to use more imagination and with a bit of back and forth it gave me enough ideas that I was able to cobble together a way to help you accommodate the new reality of approximate social media arithmetic.

(loop
with base = 3.00
for exponent from 5.99 below 6.2 by 0.1
do (format t "~&~0F^~0F=~0F~%"
base exponent (expt base exponent)))
3.^6.=721.
3.^6.=805.
3.^6.=898.
NIL

If I varied the base similarly, or either by smaller deltas, I'm sure we could find the numbers you were hitting.

And though this is a display trick, who's to say really that the right understanding of these values is not some kind of stochastic average. The function of math is not just to be some sort of stiff arbiter, but also to provide "identities" that make different ways of computing something be meaningful, so if the real value is 729, let's say, but the social media display is failing, we need some kind of corresponding "identity" that makes the surrounding math back that up. :)

I think the relevant quote here is by Mark Twain on the three kinds of lies. ;)

@dougmerritt

OK, well, that's what happens when one rushes code into service without review. :) So before someone else corrects me:

(loop with base = 3.00 for exponent from 5.99 below 6.02 by 0.01 do (format t "~&~1F^~1F=~0,F~%" base exponent (expt base exponent)))
3.^6.=721.
3.^6.=729.
3.^6.=737.

I could have just edited the other post, but I don't like papering over the idea that people make errors. We all make errors. We need processes that find them in time, and a willingness to own errors and create processes to avoid them. So I'll leave the egg on my face here for the sake of that public service reminder.

The rest of my remarks still apply.

@kentpitman
> I don't like papering over the idea that people make errors. We all make errors.

Strongly agree. It used to be a factoid in the early days that only 1% to 0.1% of social media consumers also posted, which no doubt still has a germ of truth, and I suspect part of the reason is fear of making mistakes in public.

@dougmerritt

> I suspect part of the reason is fear
> of making mistakes in public.

100.1%

@screwtape @kentpitman
Right, wouldn't want to accidentally XOR 3 with 6

@dougmerritt @kentpitman
Oh, neat. Though the pun I had been trying to make was that we often say "I had the realization.." to mean a discovery, though this also has the statistical useage "My particular sample of this random variable was"

@screwtape @kentpitman
Substituting "humor" for "irony" in the Steve Martin quote from Roxanne (1987):

"Oh, ho, ho, humor! Oh, no, no, we don't get that here. See, uh, people ski topless here while smoking dope, so humor's not really a, a high priority. We haven't had any humor here since about, uh, '83, when I was the only practitioner of it. And I stopped because I was tired of being stared at."

Also about statistics, you may have missed my final comment on Lambda Moo:

When my research mathematician friend Arunas Rudvalis (discoverer of the Rudvalis group) switched from group theory to statistics, I asked, isn't that a much more narrow field?

He said "no, you don't get it. Take *all* of mathematics, and make all of the variables random variables. That's statistics."

(No doubt I still don't get it, but that expanded my horizons anyway.)

@dougmerritt
After being reminded, I do remember that.
@kentpitman