The more I play with it, the more I enjoy it. It even comes close
to the three body problem if you play it right.
Bob
What is the height of the ceiling?
--Mario
Now the really nice version of this is to ask what is the largest
sofa you can turn around the corner. (The sofa is a rigid
2-dimensional object, and you want to get it around the corner
in the hallway without tilting it. What is the largest possible
area the sofa can have?)
--
Bergen,
Per Manne
p...@hamilton.nhh.no
------------
Allen Windhorn
Kato Engineering N. Mankato, MN 56002
> > boc...@ix.netcom.com (Robert DeSoucey) wrote:
> > >I've been flirting with this one for 30 years. Any takers?
> > > You're walking with a ladder down a hallway with a width "a". You
> > >have to turn a corner into a hallway with width "b".
> > > What is the longest ladder you can turn around this corner? Of
> > >course the ladder is a line with no width dimension.
> > >
> > > The more I play with it, the more I enjoy it. It even comes close
> > >to the three body problem if you play it right.
> > >Bob
>
> Now the really nice version of this is to ask what is the largest
> sofa you can turn around the corner. (The sofa is a rigid
> 2-dimensional object, and you want to get it around the corner
> in the hallway without tilting it. What is the largest possible
> area the sofa can have?)
Must my sofa be rectangular? :-)
Ben Tilly
and in article <47oiii$e...@Oak.IC.Mankato.MN.US>, wind...@ic.mankato.mn.us
Allen Windhorn) writes:
|>
|> There was a related problem I read about what the largest sofa was which
|> could be taken around a corner, and what the optimum shape was for it.
|> May have been one of Ian Stewart's books. Anyone remember?
It is often a piano rather than a sofa, which does have a better sense of
rigidity to my mind.
Try section G5 in [UPIG] for references. The Shephard piano is a good try:
E_______F
. | | .
. | , . | .
. |. .| .
A----B C----D
(ObApology for ASCII art) - a rectangle BCFE with quarter circles ABE, CDF
added and a semicircular bite with diameter BC taken out. AB = CD = BE = CF
= 1, the width of the corridors. It's fairly easy to see that this goes round
the corner: think "right angle in a semicircle". The area is maximised if
BC = EF = 4/pi, at 2/pi + pi/2 = 2.2074... The best known solution (as of
the 1st edition of [UPIG], anyway) is 2.2156...
There is also the "Conway car": what is the maximum area rigid 2-D shape that
can reverse in a T-junction, all roads having unit width?
[UPIG] Unsolved Problem in Geometry
H.T. Croft, K.J. Falconer, R.K. Guy
Springer (1991 for 1st edition, but there is a 2nd out)
Chris Thompson
Email: ce...@cam.ac.uk
>In <DHo7C...@dutiws.twi.tudelft.nl>,
> Jos van Kan <j.va...@math.tudelft.nl> writes:
>>>boc...@ix.netcom.com (Robert DeSoucey) wrote:
>>>>I've been flirting with this one for 30 years. Any takers?
>>>> You're walking with a ladder down a hallway with a width "a". You
>>>>have to turn a corner into a hallway with width "b".
>>>> What is the longest ladder you can turn around this corner? Of
>>>>course the ladder is a line with no width dimension.
>> (nice math deleted)
>> L = (a^(2/3) + b^(2/3))^(3/2)
>>
>There was a related problem I read about what the largest sofa was which
>could be taken around a corner, and what the optimum shape was for it.
>May have been one of Ian Stewart's books. Anyone remember?
>------------
>Allen Windhorn
>Kato Engineering N. Mankato, MN 56002
If the height of hall is h and the ladder can be carried any way
suitable the new ladder's length will be equal to L^2 +h^2 where
L is the length of ladder if carried horizontally. And this new
ladder is equal to [(a^2/3 +b^2/3)^3] + h^2
>Benjamin J. Tilly wrote:
>> Per Erik Manne <p...@hamilton.nhh.no> writes:
[snip]
>> > Now the really nice version of this is to ask what is the largest
>> > sofa you can turn around the corner. (The sofa is a rigid
>> > 2-dimensional object, and you want to get it around the corner
>> > in the hallway without tilting it. What is the largest possible
>> > area the sofa can have?)
>>
>> Must my sofa be rectangular? :-)
>>
>> Ben Tilly
>
>Then it wouldn't be too difficult, would it? Let's say the sofa
>is bounded by finitely many smooth curves.
>
>Note: I assume that the problem is unsolved. Does anyone know
>any good upper bounds on the area?
I have heard that the problem was solved a couple of years ago by
Joseph Gerver, but I don't have a reference. His solution involved
trimming a few bits off the shape called the "Shephard Piano" (after
Geoffrey Shephard) so the whole shape could be expanded a bit.
Shephard's work is described in Croft & Guy's "Unsolved Problems
in Geometry"; Gerver's work may have made it into the new edition,
which I haven't seen.
--
Fred W. Helenius <fr...@ix.netcom.com>