Just a note: We know that the inclusion

yields for

. A similar inclusion statement is valid for

-spaces
iff we look at
![L^p[a,b] L^p[a,b]](http://stephan.paukner.cc/syslog/plugin/mimetex.php?q=L%5Ep%5Ba%2Cb%5D)
. Here, for

, the inclusion is the other way round:

. This is not valid for

, as

is not finite.
In Cigler’s script, the fact

had just been written down quickly, and it wasn’t mentioned that this was only valid for bounded intervals. I already wondered why it was always said that the Fourier transform could be expanded to “

” and not simply “

”.