Vladimir Bashkardin

From Madagascar
Revision as of 15:11, 23 September 2021 by Sfomel (talk | contribs) (→‎When did you first hear about Madagascar?)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Please tell us about yourself

I am currently an HPC Software Engineer at BP America supporting large-scale R&D for seismic data processing and imaging. From 2007 to 2013 I was a PhD student at UT Austin working on computational seismology and seismic imaging problems in Sergey's group, which later became known as TCCS

When did you first hear about Madagascar?

I learnt about Madagascar for the first time in 2006 when I was looking for better alternatives to SEPlib and Seismic Unix.

What was the most difficult part in learning Madagascar?

The basics of building processing workflows in Madagascar and writing new applications for it were fairly easy. What turned out to be less trivial for me was reaching a deeper understanding of the importance of reproducibility and how the concepts of reproducible papers were implemented in Madagascar.

From your contributions to Madagascar, which one is your favorite?

I did a number of improvements to the plotting/visualization system and the parallel data handling systems. OpenGL pen (oglpen) is probably the most famous and widely used of those additions. However, my feeling has always been that my most important contribution was providing basic user support to the Madagascar community. From 2008 up until early 2013 or so, I was very active on the user mailing list answering questions about compiling, running, and troubleshooting Madagascar, its applications, and workflows on top of it. I also provided critical help to my fellow students and postdocs on many occasions allowing them to run their Madagascar-based computational experiments a little more efficiently.

What advice would you give to those who wish to learn Madagascar or to contribute to it?

I think that a lot of people, who get introduced to Madagascar, fairly quickly learn that it has its own limitations in the area of basic processing and productivity tools. The common reaction seems to be to find a quick (and often inconvenient) workaround and then spend all available time writing another batch of user-specific computational applications. I would recommend not to shy away from spending some time on expanding and improving the base toolkit too.