F2F Meeting April 27-May 3, 2014, Cardiff

Table of Contents

• Plan

Review Meetings Page

Plan

We discussed how to organise the review next week and what aspects of the pipeline to test so that we can make the best use of the available time. The schedule for each day (except Sunday and Tuesday) is the following:

10 am-1 pm: Understanding and reviewing the algorithm used gstlal inspiral 1 pm-3 pm: Lunch 3 pm-6 pm: Technical review of the pipeline elements (we can stay longer if necessary)

On Sunday we will meet at 11 am and go on until 5 pm with 1 hour break for lunch. On Tuesday we will meet at 11 am but otherwise the programme will remain the same.

Here are some details on what the committee would like to see during the review but we do realise that we might not accomplish all of this during the f2f:

1. The algorithm implemented in gstlal: the review committee will come prepared reading the relevant papers you have posted but we still need clarification on what has been implemented.

1. Does the pipeline implement exactly what is in the quoted papers, if not what are differences.

1. How stable is the code? In particular, what aspects of the pipeline are stable and what aspects are still being developed? What is the timeline for completion of all aspects?

1. Make injections in fake Gaussian coloured noise and study the output: Make ~ 10^3 injections of exactly the same signal in Gaussian noise and study the distribution of recovered masses, times, and other parameters. We would be interested in understanding if the SNR and other parameters obey the expected distribution. If you have already done this study then please point to the results, if not just you might want to start some jobs. (In fact, it would be helpful if you could walk us through setting up such a run).

1. How are triggers ranked, how is the background computed and how does that compare theoretical expectations?

1. gstlal uses the same template bank for all detectors and data. How robust this?

1. During the afternoon sessions we will walk through the codes but it is up to you how you would like to organise this.

1. Status of technical documentation