<div dir="ltr">Hi Andy,<div><br></div><div>thanks for the heads-up. Looking forward to playing around with the new features!</div><div><br></div><div>I think we should go straight to 2.5 to be honest, whether or not there are a couple more routines coming out beforehand. Once they’re in the contrib area they’re publicly available anyway, plus the sooner 2.5 comes out the sooner we can start submitting the routines in a 2.5-compliant fashion, which should make your lives easier, too.</div><div><br></div><div>Cheers,</div><div>Chris</div></div><div class="gmail_extra"><br><div class="gmail_quote">On 7 April 2016 at 17:41, Andy Buckley <span dir="ltr"><<a href="mailto:andy.buckley@cern.ch" target="_blank">andy.buckley@cern.ch</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi Xavier, Chris, and Alex,<br>
<br>
I just wanted to let you know the near-future plans for Rivet. The next major version will be 2.5.0, in which the major developments are (at last!) a move to compulsory C++11 and dropping of the Boost library dependency, and adding some machinery to include detector effects.<br>
<br>
The first of these will be welcome, I hope: we took the arrival of experiment-contributed analyses in the C++11 dialect as a clear sign that it was finally time to make that move. And it really cleans up the code when used judiciously. It's nice to finally get rid of Boost, too!<br>
<br>
We hope that the second development will make your BSM groups happy. As you know, we have long rejected calls for detector-sim integration, on the grounds that SM measurement is *definitely* best provided in unfolded form. But for BSM purposes this is unrealistic, so hopefully the arrival of particle and jet efficiency/smearing tools (both standard functions and pluggability of user-defined ones) will pave the way for Rivet analysis preservation (and re-casting) from those groups as well as the established Standard Model uses.<br>
<br>
I was particularly happy to find that Rivet's projection model suits fast-simulation very well: analyses can apply the exactly appropriate smearing functions rather than relying on one-size-fits-all objects from monolithic fast-sims, and if the same definition is used more than once, projection caching automatically kicks it to avoid wasting CPU. We'll have plenty more to say about this...<br>
<br>
We'll provide a 2.5.0 beta for you to test in the near future, but in the meantime I wanted to let you know what is coming... so you can pass the message on to those who might be interested. And to ask if you have any more imminent analyses in the pipeline? We can issue a 2.4.2 release before the final 2.5.0 if there is demand, but at present there's only one analysis in our integration queue so we'd be tempted to go straight to 2.5 if no more are on their way. And there's another plus point for that strategy: going straight to 2.5 allows you to use C++11 features! Let us know, please.<br>
<br>
Best wishes,<br>
Andy<span class="HOEnZb"><font color="#888888"><br>
<br>
-- <br>
Dr Andy Buckley, Lecturer / Royal Society University Research Fellow<br>
Particle Physics Expt Group, University of Glasgow<br>
</font></span></blockquote></div><br></div>