From: Greg Aldering (aldering@panisse.lbl.gov)
Date: Thu May 22 2003 - 10:04:53 PDT
Hi Rob,
I spent a few hours last night looking over SC4, focusing on the areas
where there were changes. Overall, I think this is now looking very good.
Your motivation for the low-extinction sample, and putting it in context
with the Riess prior has really helped that area a lot.
Here are the few issues/questions that I spotted:
I think that the SNe without suitable colors should be removed from Table 4.
Then that table is more self-consistent, being "New Fits to Perlmutter (1999) SNe
used in this paper". I think this is necessary because, e.g., the fit to 1997S
in this paper probably is not superior to the fit given in P99, whereas we
think the fits are better for others with good colors. These fits were not
necessary to *make* the cut on color error, since the fits from P99 were
suitable for that. So, there is no need to introduce them in the table.
Has the issue of 97ai been resolved? It disagrees badly with P99. Are
we just going to ignore that difference? (Note - this could be handled
later.)
p18 : I suggest a slight wording change:
"... differently, by applying a one-sided Bayesian prior to ..." --->
"... differently, as the basis for a one-sided Bayesian prior applied ..."
p20, figure 3: If the SNe without colors are removed from consideration, as
I advocated above, then the dotted tail to negative E(B-V) is
eliminated; that is, the P99 histogram has to be redone. This
has the benefit that the horizontal scale can be improved to
span only -0.75 to 0.75.
p21 : Help me out here: at the start of section 4.1 you describe K-corrected
rest-frame B-band peak magnitudes. But, aren't they also stretch-corrected?
If not, why not?
Other minor glitches:
p18 : "... whose confirmations as ..." -->
"... whose confirmation as ..."
p28 : "... this distance reduced distance to ..." --->
"... this reduced distance to ..."
- Greg
This archive was generated by hypermail 2.1.4 : Thu May 22 2003 - 10:04:55 PDT