Hi Gerry,

   I apologize again for missing the meeting and the discussion last
night.  But I do want to make sure, in the changes that seem to
have been agreed upon for the gluon experiment section, that
we will still be addressing the real questions underlying this whole
exercise.  I was not present at last summer's DOE review, and
hence have been somewhat puzzled about the origin of this
request for a new report on the spin program.  But from talking
to folks who were present, and who heard Dennis Kovar's sotto
voce mumbling, I'd say the primary interest of DOE in this
report is likely in getting answers to the following questions:

1) Why do we need to study so many different channels to
constrain Delta g(x)?
2) Why do we need to measure at both 200 and 500 GeV?
3) Why do we need so many runs at 200 GeV, and in
particular, why bother running before all the relevant
machine hardware changes (new AGS snake, all warm
bore RHIC NEG coating, full # bunches, etc.) are
done?

Obviously, the answers to some of these questions
involve machine aspects (slow buildup in luminosity)
and other aspects of the program (W production), as
well as the material covered in Sec. 2.4.  But I hope the
intended changes to Sec. 2.4 don't fail to address the
above questions.  In particular, I  am a bit concerned
about the decision to show no indication of what we
can get out of measurements along the way to final
integrated luminosity.  Part of the answer to question
(3), at least for me, is that one's emphasis has to shift
as luminosity increases.  The jet asymmetries are best
to measure at relatively low luminosity, since the cross
section is large.  For STAR, at least, we are likely to
collect jet data at a _lower_ rate after 2005 than during
2005, because emphasis in trigger bandwidth will shift
toward triggers that are efficient for the rarer channels
but less efficient for jets, and our jet rate is simply
limited by the fraction of the total trigger bandwidth
we are willing to devote to it.  Luminosity does not
come into the jet yield estimates, except in this way.
So error bars do not, for the abundant channels, scale
as 1/sqrt(integrated L). Of course, they do still scale as
1/P^2 (but that begs the question: why not wait until
new Snake is commissioned?). Somehow, this likely shift in
emphasis as time goes on, together with realistic estimates
of the different dependences of error bars on integrated L
for different channels, need to come through in the text.
I admit this point was not yet made clearly in the text I
supplied, but that was some of my thinking in deciding
to show jet asymmetries for run 5.  We should show what
we think we can get from intermediate runs, if we want
DOE to support run plans that have the intermediate runs.
And we should try hard to present credible experimental
uncertainties for all channels where we choose to show
projected data.

   Whatever we end up showing vis-a-vis xDelta g vs. Delta g/g
has to make very clearly the point about the importance of the
low-x data and the 500 GeV runs for Delta g.  If this point
gets muddled, then we've failed to answer one of the central
questions above (i.e., I don't think we should pin the 500 GeV
running exclusively on the W production).

   Hope the above is what you have in mind...

Steve

Gerry Bunce wrote:

>Hello fellow planners--we had an excellent meeting last night
>and we all have our assignments!  Next meeting Friday Feb. 4 1700 EST.
>
>We stepped thru the version of the plan that is on the web.
>
>1. Abstract--we now have one.
>2. Authors--the spinplan names will be the authors, with any additions
>of people who have worked on it that I may not know about (you should
>email me with their names).  The Acknowledgements will have a sentence
>to the effect that the plan is based on the work of many over many
>years, including acc., theory, expt, and our colleages on Star, Phenix,
>Brahms and pp2pp.
>3. The executive summary is now the one Tom Kirk edited.  I will discuss
>with Tom Ludlam and Satoshi on what they want.
>4. Science section introduction is beautiful.  Happy with 2.1 (pol DIS),
>2.2 (compelling questions), 2.3 (unpol. pp).  Note that we now have
>subprocess decompositions for all cross sections shown.  We will change
>the direct photon to be isolated with the subprocess fractions for this
>(gq is then about 90%).  (Phenix has presented both.)
>5. 2.4 introduction has been changed some with a new Table 1 showing
>Feynman graphs.  I think it is in good shape.
>6. 2.4.1 (mapping helicity preferences) needs work and Werner and I will
>try editing this.  We decided last night to only show sensitivities
>based on the full integrated luminosity (x.7) and P=.7, with no
>intermediate plots (like fig. 9 with L=7 pb^-1 for A_LL(jet)).
>We are using GRSV as the ranges, as we discussed a week ago.  A strange
>looking cusp in Fig. 8,9, etc. comes from the lower bound being set by
>two different assumptions.  Werner will add a dashed line to figure with
>a sentence to the caption to indicate this.  We will generally show
>Delta g/g and not xDelta g, and add a plot for xDelta g to make the
>point about contribution to the integral.  We will show the Dg/g plot
>only once.  For systematic errors for pi and jet we decided to use 10^-3
>for R (=L++/L+-).
>7. We decided to base statistics on running 200 GeV thru mid-2009 (5 weeks
>200, 5 weeks 500 in 2009).  Wolfram will give us max. L for this, and
>we will then use max.x.7 delivered L.
>8. gamma plots will be shown to 15 GeV for 200 and to 30 GeV p_T for
>500, to emphasize the lower p_T region (they start at 5 GeV)
>9. we will add forward A_LL(pi) and correlation plot eta==>x
>10. Naohito is preparing a Delta g/g plot for pi/jet/photon (I believe)
>that will also have SMC measurement and Compass sensitivity.  This (once
>we all see it and agree with it!) will be the final plot for the gluon
>section and go into the executive summmary.
>11. W draft will be ready Thursday morning.
>12. 2.5 (transverse spin) only came out just before the meeting so we
>didn't discuss it.  The version on the web should have the part from p27
>last paragraph thru the bullet on p29 crossed out--that is the old
>version.  We will use the 200 GeV running only for transverse section
>and use L=.25 x L delivered.  (For longitudinal, L=.75 x L delivered.)
>13. 2.6 (what else is going on) We still need 3 paragraphs for this--
>Compass, Hermes, Jlab.  Physics is included in plots of previous 
>sections.  No plots.  Abhay will come up with a draft.
>14. Elastics is changed some.  Fig. 21 will have 2 panels--carbon and p.
>15. 2.8 (Beyond SM) has changed, still some checking to do.
>16. 2.9 (eRHIC) Revisit and edit.
>17. 3. (Accel.) Fig. 23 will show 5 wks 200, 5 wks 500 in 2009. Targets
>will be shown, min. will be continued to cross target line.  Fig. 25
>will be replaced with measured jet pol.
>18. New Phenix section.  I am happy with its level of discussion and
>coverage.  Some edits but fine. (p49 should be crossed out)  Nosecone
>section too long.
>19. Steve will have the STAR section (5.2) Thursday.
>20. 5.3 (other expts)--Some editing for pp2pp, no fig. 35, Sandro gave
>me a paragraph for jet expts.  I will write a paragraph for new detector
>and send it to John/Thomas.
>21. 6 (schedule)--some edits, would still like to do table on getting
>from delivered L to L used for plots.  We will see if I get the time!
>Tables 5, 6 to be completed.
>
>We plan to have our next telephone meeting Friday, 1700 EST.
>
>Goal--get draft to BNL management and experiments before weekend, 
>request feedback early next week.  We will see.
>
>					Gerry
>  
>  
>