Finding the best world or phrase

Paul

unread,

Jan 21, 2020, 9:21:28 AM1/21/20

to

Recently, I was working on a maths problem. Intuitively, before doing
the calculations, I knew that the correct value had to be very slightly
more than 50%. The answer came out to 20 - (square root of 380) which is approx 50.64% and was a perfect match for my intuitions.
I'm looking for a good word choice to express concisely the fact that the
calculation matched my prior intuitions. "Reasonable answer" seems a bit
weak although it is concise. "...which corresponds excellently to one's
prior intuition." sounds too wordy and too pompous.
Any ideas? Maybe "sanity check" would be a good phrase to include.
"Passes a sanity check" sounds a bit weak although maybe that is best.

Thanks,

Paul Epstein

Paul

unread,

Jan 21, 2020, 9:22:35 AM1/21/20

to

"world" in the title should be "word".

occam

unread,

Jan 21, 2020, 9:39:12 AM1/21/20

to

"I knew it!" (It is neither too wordy nor too pompous. It is a bit
know-it-all though.)

Peter Moylan

unread,

Jan 21, 2020, 9:55:33 AM1/21/20

to

Back-of-the-envelope calculation.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW, Australia

Paul

unread,

Jan 21, 2020, 10:10:44 AM1/21/20

to

I mean a more formal context. I'm preparing for a job interview where
this computation will be relevant.

Thanks, though.

Paul

Paul

unread,

Jan 21, 2020, 10:11:44 AM1/21/20

to

Yes, that's a good one. Didn't think of that.

Thanks!

Paul

Janet

unread,

Jan 21, 2020, 10:40:54 AM1/21/20

to

In article <03da3cee-1a49-4efc...@googlegroups.com>,
peps...@gmail.com says...

Guesstimate.

Janet

Athel Cornish-Bowden

unread,

Jan 21, 2020, 11:07:03 AM1/21/20

to

On 2020-01-21 14:39:08 +0000, occam said:

> On 21/01/2020 15:21, Paul wrote:
>> Recently, I was working on a maths problem. Intuitively, before doing
>> the calculations, I knew that the correct value had to be very slightly
>> more than 50%. The answer came out to 20 - (square root of 380) which
>> is approx 50.64% and was a perfect match for my intuitions.

Many years ago (1977, 1979) I spent a week or ten days struggling with
some statistical theory that I knew very badly (struggling so much that
it generated the first migraine that I had had since I was a teenager).
Finally I came to the conclusion that a particular calculation should
have coefficient of variation of about 38%. Then I tried it on a lot of
data from the literature, and guess what: the coefficients of variation
for 8 data sets ranged from 30.9% to 45.7% (JTB65-735.pdf; Table 3 of
JTB76-369.pdf). I was dead chuffed.

(To save anyone the trouble of pointing it out, the method discussed in
these papers is now totally obsolete (though still occasionally cited),
as much more powerful methods of comparing protein and gene sequences
are now available. As one of the referees pointed out at the time, this
was not a method whose time had come; it was a method whose time had
passed. That wasn't entirely fair, as much worse methods were still
being used.)

>> I'm looking for a good word choice to express concisely the fact that the
>> calculation matched my prior intuitions. "Reasonable answer" seems a bit
>> weak although it is concise. "...which corresponds excellently to one's
>> prior intuition." sounds too wordy and too pompous.
>> Any ideas? Maybe "sanity check" would be a good phrase to include.
>> "Passes a sanity check" sounds a bit weak although maybe that is best.
>>
>> Thanks,
>>
> "I knew it!" (It is neither too wordy nor too pompous. It is a bit
> know-it-all though.)

--
athel

Spains Harden

unread,

Jan 21, 2020, 11:14:51 AM1/21/20

to

On Tuesday, January 21, 2020 at 2:21:28 PM UTC, Paul wrote:

"Passes a sanity check" sounds like something out of "A Night at the
Opera", so I wouldn't go down that road. Nor "a back of a fag packet"
calculation.

"A Night at the Opera": <https://www.youtube.com/watch?v=G_Sy6oiJbEk>

Richard Heathfield

unread,

Jan 21, 2020, 11:33:20 AM1/21/20

to

"...which was in line with my first-order approximation..."

--
Richard Heathfield
Email: rjh at cpax dot org dot uk
"Usenet is a strange place" - dmr 29 July 1999
Sig line 4 vacant - apply within

charles

unread,

Jan 21, 2020, 11:43:50 AM1/21/20

to

In article <03da3cee-1a49-4efc...@googlegroups.com>,

Paul <peps...@gmail.com> wrote:
> Recently, I was working on a maths problem. Intuitively, before doing
> the calculations, I knew that the correct value had to be very slightly
> more than 50%. The answer came out to 20 - (square root of 380) which is
> approx 50.64% and was a perfect match for my intuitions.

20 squared is 400. so the square root of 380 is goingb tobe abit less than
20 - 19.5, to a first order.

> I'm looking for a good word choice to express concisely the fact that the
> calculation matched my prior intuitions. "Reasonable answer" seems a bit
> weak although it is concise. "...which corresponds excellently to one's
> prior intuition." sounds too wordy and too pompous.
> Any ideas? Maybe "sanity check" would be a good phrase to include.
> "Passes a sanity check" sounds a bit weak although maybe that is best.

> Thanks,

> Paul Epstein

--
from KT24 in Surrey, England
"I'd rather die of exhaustion than die of boredom" Thomas Carlyle

Mack A. Damia

unread,

Jan 21, 2020, 12:03:36 PM1/21/20

to

On Tue, 21 Jan 2020 06:21:24 -0800 (PST), Paul <peps...@gmail.com>
wrote:

What is wrong with describing yourself as "intuitive" in a sentence?

Other words, too: prescient, predictive, insightful, perceptive.

Jerry Friedman

unread,

Jan 21, 2020, 12:55:06 PM1/21/20

to

If you start by saying, "I started by making a rough estimate" or
"Intuitively we can expect the answer to be slightly greater than
50%," you can say at the end, "which comes out to about 50.64%, as
I expected." Or "in line with my initial estimate" or whatever.

--
Jerry Friedman

Peter Moylan

unread,

Jan 21, 2020, 7:42:43 PM1/21/20

to

On 2020-Jan-22 01:21, Paul wrote:

> Recently, I was working on a maths problem. Intuitively, before
> doing the calculations, I knew that the correct value had to be very
> slightly more than 50%. The answer came out to 20 - (square root of
> 380) which is approx 50.64% and was a perfect match for my
> intuitions.

In the past I tried to teach my students the skill of estimating rough
answers to problems, but few of them saw the point. They had
calculators, so why try to do calculations in their heads? It's possible
that they saw me as a dinosaur who couldn't adapt to new technology.

But, when doing things like calculating a current in an electrical
circuit, I got answers that were within about 10% of the right answer,
while they got answers that were painfully precise but out by a factor
of a million.

(Sorry, Paul, I've snipped your main question because I couldn't think
of a good answer.)

Athel Cornish-Bowden

unread,

Jan 22, 2020, 2:41:08 AM1/22/20

to

On 2020-01-22 00:42:38 +0000, Peter Moylan said:

> On 2020-Jan-22 01:21, Paul wrote:
>
>> Recently, I was working on a maths problem. Intuitively, before
>> doing the calculations, I knew that the correct value had to be very
>> slightly more than 50%. The answer came out to 20 - (square root of
>> 380) which is approx 50.64% and was a perfect match for my
>> intuitions.
>
> In the past I tried to teach my students the skill of estimating rough
> answers to problems, but few of them saw the point. They had
> calculators, so why try to do calculations in their heads?

Yes. That's why you get people finding (for example) a concentration of
16832 molar as the solution to a problem. About 30 years there was an
established researcher I knew who got a calculation wrong by a factor
of about 10^15. In the days of slide rules everyone knew they had to
determine the order of magnitude themselves. Slide rules also had the
advantage that they couldn't give absurdly precise anwers with 10
significnt figures.

> It's possible
> that they saw me as a dinosaur who couldn't adapt to new technology.
>
> But, when doing things like calculating a current in an electrical
> circuit, I got answers that were within about 10% of the right answer,
> while they got answers that were painfully precise but out by a factor
> of a million.
>
> (Sorry, Paul, I've snipped your main question because I couldn't think
> of a good answer.)

--

athel

Sam Plusnet

unread,

Jan 22, 2020, 2:03:20 PM1/22/20

to

On 22-Jan-20 7:41, Athel Cornish-Bowden wrote:
> Slide rules also had the advantage that they couldn't give absurdly
> precise anwers with 10 significnt figures.

Anyone care to offer a design for a slide rule which could offer that
sort of precision?

--
Sam Plusnet

Peter T. Daniels

unread,

Jan 22, 2020, 3:15:24 PM1/22/20

to

It just needs to be very long -- the more room you have for the lines,
the more you can squeeze in while keeping the same upper bound. You
might also want to work on your engraving or etching technique to get
the lines really fine.

You still get to figure out the order of magnitude for yourself.

J. J. Lodder

unread,

Jan 22, 2020, 5:14:12 PM1/22/20

to

No length at all can be mesured with that kind of precision,
for it is the precision to which the meter can be realised

Jan

Richard Heathfield

unread,

Jan 22, 2020, 6:08:03 PM1/22/20

to

I have no doubt you're right, but at least we could 'ave a go, couldn't
we? What about a laser ruler in space (no atmospheric wossname to worry
about)?

I mean, okay, it'd be a bloody big slide rule, but I suspect Sam was
probably expecting something quite big.

Or maybe we could somehow rig up a multi-slide mechanism, where the
heavy lifting is done by one slide, and the finer detail filled in by
another?

Look, we don't get this kind of opportunity very often, so when it does
come along, we don't just *give up*!

Sam: I reckon I might be able to do it for as little as £10,000,000.00
(which, rather pleasingly, is correct to precisely ten significant
figures). If you accept the tender, I am definitely willing and able to
give you a receipt.

Sam Plusnet

unread,

Jan 22, 2020, 9:35:32 PM1/22/20

to

On 22-Jan-20 23:07, Richard Heathfield wrote:
> On 22/01/2020 22:14, J. J. Lodder wrote:
>> Sam Plusnet <n...@home.com> wrote:
>>
>>> On 22-Jan-20 7:41, Athel Cornish-Bowden wrote:
>>>> Slide rules also had the advantage that they couldn't give absurdly
>>>> precise anwers with 10 significnt figures.
>>>
>>> Anyone care to offer a design for a slide rule which could offer that
>>> sort of precision?
>>
>> No length at all can be mesured with that kind of precision,
>> for it is the precision to which the meter can be realised
>
> I have no doubt you're right, but at least we could 'ave a go, couldn't
> we? What about a laser ruler in space (no atmospheric wossname to worry
> about)?
>
> I mean, okay, it'd be a bloody big slide rule, but I suspect Sam was
> probably expecting something quite big.
>
> Or maybe we could somehow rig up a multi-slide mechanism, where the
> heavy lifting is done by one slide, and the finer detail filled in by
> another?
>
> Look, we don't get this kind of opportunity very often, so when it does
> come along, we don't just *give up*!
>
> Sam: I reckon I might be able to do it for as little as £10,000,000.00
> (which, rather pleasingly, is correct to precisely ten significant
> figures). If you accept the tender, I am definitely willing and able to
> give you a receipt.
>

Surely you will be seeking public funding for such an important task?

Something based on a Fuller calculator might be less of a handful.

https://www.hpmuseum.org/srcyl.htm

The model shown is said to be capable of 4 to 5 sig figs.

Perhaps it could double as a prayer wheel?

--
Sam Plusnet

Richard Heathfield

unread,

Jan 22, 2020, 9:44:41 PM1/22/20

to

On 23/01/2020 02:35, Sam Plusnet wrote:
> On 22-Jan-20 23:07, Richard Heathfield wrote:

<snip>

>> Sam: I reckon I might be able to do it for as little as £10,000,000.00
>> (which, rather pleasingly, is correct to precisely ten significant
>> figures). If you accept the tender, I am definitely willing and able
>> to give you a receipt.
>>
> Surely you will be seeking public funding for such an important task?

I'll leave the financing to you.

> Something based on a Fuller calculator might be less of a handful.
>
> https://www.hpmuseum.org/srcyl.htm
>
> The model shown is said to be capable of 4 to 5 sig figs.

If a 41 ft scale gives you 5 sig figs, you'll need a 4100000 ft scale to
give you ten. That's 776 miles, by the way.

> Perhaps it could double as a prayer wheel?

It could indeed. In fact, it could double as that much-needed bridge
between Thessaloniki and Alexandria.

Rich Ulrich

unread,

Jan 23, 2020, 1:47:29 AM1/23/20

to

On Wed, 22 Jan 2020 23:07:53 +0000, Richard Heathfield
<r...@cpax.org.uk> wrote:

>On 22/01/2020 22:14, J. J. Lodder wrote:
>> Sam Plusnet <n...@home.com> wrote:
>>
>>> On 22-Jan-20 7:41, Athel Cornish-Bowden wrote:
>>>> Slide rules also had the advantage that they couldn't give absurdly
>>>> precise anwers with 10 significnt figures.
>>>
>>> Anyone care to offer a design for a slide rule which could offer that
>>> sort of precision?
>>
>> No length at all can be mesured with that kind of precision,
>> for it is the precision to which the meter can be realised
>
>I have no doubt you're right, but at least we could 'ave a go, couldn't
>we? What about a laser ruler in space (no atmospheric wossname to worry
>about)?
>
>I mean, okay, it'd be a bloody big slide rule, but I suspect Sam was
>probably expecting something quite big.
>
>Or maybe we could somehow rig up a multi-slide mechanism, where the
>heavy lifting is done by one slide, and the finer detail filled in by
>another?

Or inside of slide mechanisms, you could use wheels. And
you could use more than two of them. Work this out some
more and I think you can invent the mechanical calculating
machine.

>
>Look, we don't get this kind of opportunity very often, so when it does
>come along, we don't just *give up*!
>
>Sam: I reckon I might be able to do it for as little as £10,000,000.00
>(which, rather pleasingly, is correct to precisely ten significant
>figures). If you accept the tender, I am definitely willing and able to
>give you a receipt.

--

Rich Ulrich

occam

unread,

Jan 23, 2020, 3:04:21 AM1/23/20

to

Err... if your ruler's gradations are the wavelength of (say) laser
light, yes you can! Two parallel laser beams, movable in relation to
each other. The interference pattern should pinpoint the answer to
better than 10^-9 meters. (Details to be worked out on the back of an
envelope.)

J. J. Lodder

unread,

Jan 23, 2020, 5:00:03 AM1/23/20

to

Richard Heathfield <r...@cpax.org.uk> wrote:

> On 22/01/2020 22:14, J. J. Lodder wrote:
> > Sam Plusnet <n...@home.com> wrote:
> >
> >> On 22-Jan-20 7:41, Athel Cornish-Bowden wrote:
> >>> Slide rules also had the advantage that they couldn't give absurdly
> >>> precise anwers with 10 significnt figures.
> >>
> >> Anyone care to offer a design for a slide rule which could offer that
> >> sort of precision?
> >
> > No length at all can be mesured with that kind of precision,
> > for it is the precision to which the meter can be realised
>
> I have no doubt you're right, but at least we could 'ave a go, couldn't
> we? What about a laser ruler in space (no atmospheric wossname to worry
> about)?

Yes, that's it.
The best practical meter standards are stabilised lasers,
and the best you can do is a precision of about 10^-10
Next excercise: can you modulate a log scale onto it?

> I mean, okay, it'd be a bloody big slide rule, but I suspect Sam was
> probably expecting something quite big.
>
> Or maybe we could somehow rig up a multi-slide mechanism, where the
> heavy lifting is done by one slide, and the finer detail filled in by
> another?

The thing to do is to build a ring laser in a tunnel under the equator.
You might be able to do it, for a few zeros more,

Jan

unread,

Jan 23, 2020, 9:49:55 AM1/23/20

to

On 2020-01-23 14:42:22 +0000, Richard Heathfield said:

> On 23/01/2020 14:25, Athel Cornish-Bowden wrote:
>> On 2020-01-23 14:01:40 +0000, Richard Heathfield said:
>>
>>> On 23/01/2020 12:15, Athel Cornish-Bowden wrote:
>>>> On 2020-01-23 10:00:00 +0000, J. J. Lodder said:
>>>>
>>> <snip>
>>>>>
>>>>> The thing to do is to build a ring laser in a tunnel under the equator.
>>>>> You might be able to do it, for a few zeros more,
>>>>
>>>> I suspect he'll need more than 10000000€ to do that.
>>>
>>> Indeed (except that I said £, not €).
>>
>> I thought you probably did, but I hadn't the energy to check. In any
>> case, by the time Boris Johnson has done his stuff £10000000 will be
>> less than 10000000€.
>
> Perhaps it will, or perhaps it won't. You never can tell, with currency
> exchange rates.

I said that to provoke a response. I didn't think that a Brexiter such
as you would agree. Maybe once the transition period has faded into
history £10000000 will be worth 20000000€. We can only guess.

--
athel

Richard Heathfield

unread,

Jan 23, 2020, 10:10:56 AM1/23/20

to

On 23/01/2020 14:49, Athel Cornish-Bowden wrote:
> On 2020-01-23 14:42:22 +0000, Richard Heathfield said:
>
>> On 23/01/2020 14:25, Athel Cornish-Bowden wrote:
>>> On 2020-01-23 14:01:40 +0000, Richard Heathfield said:
>>>
>>>> On 23/01/2020 12:15, Athel Cornish-Bowden wrote:
>>>>> On 2020-01-23 10:00:00 +0000, J. J. Lodder said:
>>>>>
>>>> <snip>
>>>>>>
>>>>>> The thing to do is to build a ring laser in a tunnel under the
>>>>>> equator.
>>>>>> You might be able to do it, for a few zeros more,
>>>>>
>>>>> I suspect he'll need more than 10000000€ to do that.
>>>>
>>>> Indeed (except that I said £, not €).
>>>
>>> I thought you probably did, but I hadn't the energy to check. In any
>>> case, by the time Boris Johnson has done his stuff £10000000 will be
>>> less than 10000000€.
>>
>> Perhaps it will, or perhaps it won't. You never can tell, with
>> currency exchange rates.
>
> I said that to provoke a response.

I know. And I was happy to oblige. But if you were looking for a
partisan reply, I may have disappointed you.

> I didn't think that a Brexiter such
> as you would agree.

I prefer reality to tribalism.

Yes, *of course* the pound might suffer from Brexit. Or it might not.
"Prediction is very difficult, especially about the future" - Niels Bohr
(attrib).

For many people, the economic outcome will be the most important.
Personally, I am of the opinion that, when left to themselves, free
market economies tend to bounce back from temporary hiccups, and so I
consider other aspects of Brexit to be of much greater significance; but
mine is just one opinion among many millions, and I won't go into
nauseating detail because it would be a shame to turn a light-hearted
sub-thread about slide rules into Yet Another Brexit Row.

Like any wholesale political upheaval, Brexit will have many
consequences: some will be good for the UK, and some will be bad, and
opinions on which are good and which are bad will vary widely.

> Maybe once the transition period has faded into
> history £10000000 will be worth 20000000€. We can only guess.

Indeed. And in a hundred years, people will look back and wonder why we
all made such a big fuss about it all. (Or why we didn't make a big
enough fuss.)

--
Richard Heathfield
Email: rjh at cpax dot org dot uk
"Usenet is a strange place" - dmr 29 July 1999

J. J. Lodder

unread,

Jan 23, 2020, 11:06:38 AM1/23/20

to

I have some friends who are quite happy with it.
They sold their overpriced house in France
to some Brit who wanted to convert his pounds
into euro real estate in a hurry, before it happened,

Jan

Sam Plusnet

unread,

Jan 23, 2020, 3:24:29 PM1/23/20

to

On 23-Jan-20 6:47, Rich Ulrich wrote:
> On Wed, 22 Jan 2020 23:07:53 +0000, Richard Heathfield
> <r...@cpax.org.uk> wrote:
>
>> On 22/01/2020 22:14, J. J. Lodder wrote:
>>> Sam Plusnet <n...@home.com> wrote:
>>>
>>>> On 22-Jan-20 7:41, Athel Cornish-Bowden wrote:
>>>>> Slide rules also had the advantage that they couldn't give absurdly
>>>>> precise anwers with 10 significnt figures.
>>>>
>>>> Anyone care to offer a design for a slide rule which could offer that
>>>> sort of precision?
>>>
>>> No length at all can be mesured with that kind of precision,
>>> for it is the precision to which the meter can be realised
>>
>> I have no doubt you're right, but at least we could 'ave a go, couldn't
>> we? What about a laser ruler in space (no atmospheric wossname to worry
>> about)?
>>
>> I mean, okay, it'd be a bloody big slide rule, but I suspect Sam was
>> probably expecting something quite big.
>>
>> Or maybe we could somehow rig up a multi-slide mechanism, where the
>> heavy lifting is done by one slide, and the finer detail filled in by
>> another?
>
> Or inside of slide mechanisms, you could use wheels. And
> you could use more than two of them. Work this out some
> more and I think you can invent the mechanical calculating
> machine.
>

First we must rid ourselves of a lot of excess Babbage.

--
Sam Plusnet

Richard Heathfield

unread,

Jan 23, 2020, 3:27:53 PM1/23/20

to

That was uncalled for, you dirty Pascal.

Kerr-Mudd,John

unread,

Jan 23, 2020, 4:48:12 PM1/23/20

to

On Thu, 23 Jan 2020 20:27:49 GMT, Richard Heathfield <r...@cpax.org.uk>

Please, gentlemen, settle your Differences amigably.

--
Bah, and indeed, Humbug.

Quinn C

unread,

Jan 23, 2020, 5:56:09 PM1/23/20

to

* Spains Harden:

> "Passes a sanity check" sounds like something out of "A Night at the
> Opera", so I wouldn't go down that road.

In software development, the phrase is entirely quotidian.

--
- It's the title search for the Rachel property.
Guess who owns it?
- Tell me it's not that bastard Donald Trump.
-- Gilmore Girls, S02E08 (2001)

Quinn C

unread,

Jan 23, 2020, 5:56:12 PM1/23/20

to

* Kerr-Mudd,John:

And I hope we can Adapt to the idea that this subject isn't for
gentlemen only.

--
Vivo junto con mi amiga - I live with my home computer
-- My Spanish teacher, ca. 1990

Anders D. Nygaard

unread,

Jan 23, 2020, 6:33:57 PM1/23/20

to

Den 21-01-2020 kl. 15:21 skrev Paul:
> Recently, I was working on a maths problem. Intuitively, before doing
> the calculations, I knew that the correct value had to be very slightly
> more than 50%. The answer came out to 20 - (square root of 380) which is approx 50.64% and was a perfect match for my intuitions.

> I'm looking for a good word choice to express concisely the fact that the
> calculation matched my prior intuitions. "Reasonable answer" seems a bit
> weak although it is concise. "...which corresponds excellently to one's
> prior intuition." sounds too wordy and too pompous.
> Any ideas? Maybe "sanity check" would be a good phrase to include.
> "Passes a sanity check" sounds a bit weak although maybe that is best.

Something like "which closely matches my first impression"?

/Anders, Denmark.

Sam Plusnet

unread,

Jan 23, 2020, 10:21:59 PM1/23/20

to

True. We must embrace those who Love lace.

--
Sam Plusnet

Peter Moylan

unread,

Peter Moylan

unread,

Jan 25, 2020, 5:19:12 AM1/25/20

to

Sorry, I didn't mean to misle you. It was a joke, which makes sense only
if you've heard of Linda Lovelace.

Mack A. Damia

unread,

Jan 25, 2020, 11:06:46 AM1/25/20

to

Vacuuming your hard drive, innit?

Anders D. Nygaard

unread,

Jan 27, 2020, 4:58:17 PM1/27/20

to

But, but ... that was Athel's joke.

/Anders, Denmark.