<html>
<head>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">Curtis,<br>
<br>
Ok, let me propose the following. Next offline meeting I will
present the state of the conversion to the distilled SEST format
(Simulated Event STorage format). At that point we can decide
whether it is completed and tested sufficiently well to go into
production or not, or whether to proceed with the simulation hddm
data model in its present state.<br>
<br>
SEST will be what I call the simulated event storage format. SEST
becomes REST as soon as it has passed through dana and the hits
information and simulation-truth elements are suppressed. It will
contain the simulated hitView structure (properly factorized into
pure hits and pure truth tags) plus a hook for appending the REST
structure (for people who want both hits and reconstructed results
for studies) plus a small handful of other extended truth tags
that have been introduced by David and Simon over the years, such
as mcTrajectory, etc. A tool will be provided to strip arbitrary
SEST files down to bare hits. Such "bare hits" files will also be
valid SEST files, but missing the information contained in the
extra structures. The danahddm plugin will be able to read any
SEST file and properly handle situations where various types of
information are available or not, according to David's factory
schema.<br>
<br>
-Richard J.<br>
<br>
<br>
On 10/10/2012 12:48 PM, Curtis A. Meyer wrote:<br>
</div>
<blockquote cite="mid:5567D71A-8292-43F3-87EF-EA86874DD402@cmu.edu" type="cite">
<div>Hi Richard -</div>
<div><br>
</div>
<div> if saving the intermediate storage requirements helps on
the grid production, then there is a case for waiting. </div>
<div>However, there is also a good case to be made for running the
exact same version of the code at both</div>
<div>Jefferson Lab and on the OSG for this data challenge, and I
think that the need for the large sample</div>
<div>is high enough, that I would feel better NOT waiting too long
to start things. </div>
<div><br>
</div>
<div> That said, it felt like the changes that you need to make
are substantial enough that they probably</div>
<div>cannot be implemented and tested on a very short time scale.
If the time scale is ~week, then I would</div>
<div>say go for it, however, if is ~month, I would prefer to move
forward ASAP.</div>
<div><br>
</div>
<div> My two cents --- Curtis </div>
<br>
<div>
<div>
<div>-------</div>
<div>Prof. Curtis A. Meyer Department of
Physics</div>
<div>Phone: (412) 268-2745 Carnegie Mellon
University</div>
<div>Fax: (412) 681-0648
Pittsburgh, PA 15213</div>
<div><a href="mailto:curtis.meyer@cmu.edu">curtis.meyer@cmu.edu</a>
<a href="http://www.curtismeyer.com/">http://www.curtismeyer.com/</a></div>
<div><br>
</div>
<div><br>
</div>
</div>
<br class="Apple-interchange-newline">
<br class="Apple-interchange-newline">
</div>
<br>
<div>
<div>On Oct 10, 2012, at 12:29 PM, Richard Jones wrote:</div>
<br class="Apple-interchange-newline">
<blockquote type="cite">
<div>
<div class="moz-cite-prefix">Hello Paul,<br>
<br>
I saw this proposal in the minutes Mark circulated. I am
satisfied with it, and it means that production can
proceed before I get all of these changes checked in and
validated. I am still hoping to get the changes in before
I begin the grid production, since I am saving
considerable volumes of the intermediate simulation
output. Do you agree with this?<br>
<br>
-Richard J.<br>
<br>
<br>
On 10/9/2012 5:06 PM, Paul Mattione wrote:<br>
</div>
<blockquote cite="mid:B4F8A882-A1BF-44D2-9599-6D9A83FB3297@jlab.org" type="cite"> Hey Richard, I went ahead and modified the
rest.xml file with the TOF uncertainties.
<div><br>
</div>
<div>At this stage are REST and HDDM ready for generating
the large pythia sample? I know that you also wanted to
make some other changes to HDDM, such as separating the
truth hits from the reconstructed ones. But these
should only affect the regular HDDM format, and not the
REST one, right? Although I suppose that if we wanted
to save some % of the full data we'd want these other
upgrades. Should we make a tagged release and start
without it anyway? </div>
<div><br>
</div>
<div> - Paul</div>
<div>
<div>
<div><br>
<div>Begin forwarded message:</div>
<br class="Apple-interchange-newline">
<blockquote type="cite">
<div><b>From: </b><a href="mailto:Hall-D.SVN.Repository@jlab.org">Hall-D.SVN.Repository@jlab.org</a><br>
</div>
<div><b>Date: </b>October 9,
2012 4:57:33 PM EDT<br>
</div>
<div><b>To: </b><a href="mailto:davidl@jlab.org">davidl@jlab.org</a>,
<a href="mailto:brash@pcs.cnu.edu">brash@pcs.cnu.edu</a>,
<a href="mailto:wolin@jlab.org">wolin@jlab.org</a>,
<a href="mailto:zisis@uregina.ca">zisis@uregina.ca</a>,
<a href="mailto:mashephe@indiana.edu">mashephe@indiana.edu</a>,
<a href="mailto:remitche@indiana.edu">remitche@indiana.edu</a>,
<a href="mailto:zihlmann@jlab.org">zihlmann@jlab.org</a>,
<a href="mailto:somov@jlab.org">somov@jlab.org</a>,
<a href="mailto:staylor@jlab.org">staylor@jlab.org</a>,
<a href="mailto:kmoriya@indiana.edu">kmoriya@indiana.edu</a>,
<a href="mailto:pmatt@jlab.org">pmatt@jlab.org</a>,
<a href="mailto:leckey@jlab.org">leckey@jlab.org</a><br>
</div>
<div><b>Subject: </b><b>r9757
- trunk/sim-recon/src/libraries/HDDM</b><br>
</div>
<br>
<div>Author: pmatt<br>
Date: 2012-10-09 16:57:33 -0400 (Tue, 09 Oct
2012)<br>
New Revision: 9757<br>
<br>
Modified:<br>
trunk/sim-recon/src/libraries/HDDM/rest.xml<br>
Log:<br>
Added uncertainties to tofPoint, and added units
to tofPoint and startHit. <br>
<br>
<br>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</blockquote>
<br>
</div>
</blockquote>
</div>
<br>
</blockquote>
<br>
</body>
</html>