Ciao Alessio,
interesting (and quick) start!

a few comments/questions:

1) I do not agree on the interpretation of table 1. What I think we need
to check is the agreement of the fraction of events of a given kind over
the total. In this sense D*lnu is in perfect agreement (4.6% vs 10.2%)
while Dlnu is slightly off (~25% too little in our MC). Anyhow you are
folding the efficiencies in, so I do not think that table is too
interesting. I would like to see the numbers at generator level (compared
with the PDG - this is better than the DECAY.DEC because there can always
be some tricky effect).

2) as far as the shapes are concerned I am quite confused by the Dlnu and
the D*lnu shapes. Why is there such a big high MX tail? Does it mean we
are grabbing a lot of pions from the reco side? This should be
investigated because we might still be in time to recover tails for the
signal finding out what is going on here

3) As far as table 2 is concerned, please forget about Dslnu. They are
cabibbo suppressed (i.e. ~0.1%). You must have grabbed the Bs stuff and
you need a Y(5S) to produce them ...
You are also confusing things as far as the letons are concerned: by Dlnu
the PDG means the average between Denu and Dmunu and not the sum.

4) Your question (in private e-mail, but I think it is good to discuss it
here as well) what to do with the D**lnu BF which are not measured.

First of all I believe this is a study that should be done on our data and
get these very useful parameters ourselve. I would say that the recoil
can be used to this purpose. Since this is a longer term plan here is a
start to the discussion:

	- AFAIK (starting from Urs's list) the most up-to-date knowledge
on the topic is
http://www.slac.stanford.edu/~ursl/slbfsx/e/ep-2001-050.ps.gz
and it gives the total D**/D*X lnu B.F. and the B.F. for two narrow
resonances. I would suggest to follow this prescription and vary the
unmeasured components generously (keeping the known fractions fixed)
	- I would also suggest to look at the MX distribution for each
component and see if we can cluster them by "similarity" and fit for them.
Actually I though this was something the moments analysis would have done.
In particular I am slightly confused on how the relative fractions of D**
impact the moments measurent.



As a summary, I think we should look at
  a) the table of the fraction of semileptonic events that in our MC at
generator level are Dlnu and  D*lnu, D**lnu, D(npi)lnu, D*(npi)lnu for
each D** resonance present in our MC and for several values of "n".
The fractions should add up to 100% [ok, I guess there will be a 'other'
category with some % of events]
  b) the spectra for each of these (and mean and RMS) in order to see if
we can fit for superblocks
  c) understanding the Dlnu spectrum
  c) implementing a way to vary within errors the D(*,**)lnu known
branching fractions and to generously randomize the other components
keeping the known constraints satisfied

	ciao
	ric


On Thu, 11 Apr 2002, Alessio Sarti wrote:

> Hi all,
> I've ran over b0 cocktail and now I have the first results posted in
> http://www.slac.stanford.edu/~asarti/recoil/sys/index.html
>
> Comments and questions are welcomed.
> Alessio
>
> ______________________________________________________
> Alessio Sarti
>  Universita' & I.N.F.N. Ferrara
>  tel  +39-0532-781928  Ferrara
>
> "Quod non fecerunt barbari, fecerunt Berlusconi"
>
> "Che il bianco sia bianco, e che il nero sia nero
>  che uno e uno fanno due e che la scienza dice il vero....
>  DIPENDE !"
>
>
>