Category Archives: Conferences

The use of social networks and blogs during conferences/workshops

Last week I attended the Python in Astronomy workshop in Leiden, Netherlands. If you have been reading this blog or you follow me on twitter, this should not be big news since I was very active on both. Along with my tweets/blog, loads of other colleagues attending the conference also tweeted under the has-tag #pyastro15. There are also two other blogs (Pauline Barmby
and Abigail Stevens) with summaries regarding the workshop. This has been the first time when both, my blog and twitter account have been boosted massively due to the workshop. Thus, I wanted to share here my experience and a few stats from my accounts that may reflect the impact of my activity during the workshop.

This is not the first time I try to share what is going on during a conference with people that cannot attend it, but probably it was the first time that I took it more seriously.  Since the number of participants was quite limited, we were encouraged to try to share as much as possible what it was happening at the workshop while it was happening. This was great fun and a really efficient way of involving people not in the conference. At the same time, a lot of people in the workshop were also tweeting which kicked off several conversations/discussions online during talks and more importantly after talks during lunch breaks or as unconference sessions in the afternoon.

There are a few examples that can show the impact of tweeting during a conference/workshop, but probably the best one would be the big discussion created from one of Pauline’s tweets. The question/comment was: is software a primary product of science? Loads of people (not in the workshop) engaged on twitter in a long discussion with people in favour and against this idea. I don’t really want to go now into giving my view on this topic (may be in a future post), but as an example of  the impact of that the discussion reached, it even moved from twitter to a blog post by C. Titus Brown where the discussion continued in the comments field.

This is just one example, but not the only one. There were loads of questions/comments directed to people in the workshop. For example, someone pointed out the existence of a software package similar to one presented in the workshop. This led to a quick comparison of both with the result of one (the one presented at the workshop) being 45 times faster than the other code. With this, we all learned two things, 1.- that another software existed and 2.- we gave potential user a useful comparison of both.

My personal view of using twitter was very positive and I will recommend/encourage it for any conference/workshop. It is true that we had the issue of using sometimes jargon only common for those at the workshop. Once people pointed this out to us we tried to be more communicative by explaining less common terms. Lesson learned. But besides this problem, I think twitter not only helped people around the world to feel closer to us, but it also helped us see what things were happening simultaneously at different parallel sessions. And it helped us sharing documents faster and effortlessly. In summary, it made us more communicative with people that for one reason or another weren’t there with us.

In terms of numbers, I saw my twitter account getting much more hits than usually. With twitter analytics it is not possible to see from where these new hits come from, but the numbers were so large that they couldn’t be created just by people in the workshop.

My blog also shows similar statistics, with the advantage of knowing the country of origin. A large portion of views came from the Netherlands (20% of the total), but the country from which most of the views came from was USA (33%). In total, I got views from 21 countries around the world. In my opinion, these results reflect the interest that the workshop had in many countries and the importance that sharing our daily life at the workshop with other astronomers/astrophysicists really had.

As an addition to these numbers, I can also see how many times a link has been clicked. This is quite useful since during the workshop I shared many links to packages mentioned. It is an approximated way of knowing how much interest a package has had. In total, links to these packages were clicked 115 times. Considering that at the workshop we all had access to most of these links and that we also tweeted them, I suspect that most (definitively some) of these clicks come from people not present at the workshop. And thus, people really interested on what it was being presented and potential new users of those packages.

Thus, my final conclusion after all these is, that being actively online with all my workshop colleagues had a really positive impact on the workshop and on every aspect of it, specially spreading the word about software packages that maybe would not have been discovered otherwise.

Advertisements

Python in Astronomy – day 5

Today we said goodbye to the #pyastro15 workshop. A full week has come and go. A wonderful week I should say, full of incredible moments and of course amazing software for astronomy/astrophysics written in Python.

The morning began with a talk by Emille Ishida on COIN (http://cointoolbox.github.io) an interesting package that tries to make a bridge between statistics in several languages and data analysis. COIN is a clear example of an excellent package built thanks to the effort and good will of a small group of people without any real funding.

Matthew Craig showed us astropy tools for undergraduates including his package reducer (https://github.com/mwcraig/reducer) which makes optical data reduction a simpler and user friendly job. Keep an eye to the talks once they are in youtube to get from Matthew loads of interesting ideas on using ipython notebooks for educational purposes.

Joe Zuntz talked about CosmoSIS (https://bitbucket.org/joezuntz/cosmosis/wiki/Home) a cosmological parameter estimator tool. A really engaging talk that led into a small debate on which tools can/should be used in different situations to merge/embed existent C/C++ code with Python. My conclusion from the discussion was, use Cython is you want to convert your Python code into C/C++ and ctypes (or the numpy version) when the C/C++ code already exists.

We finished the talk sessions with a keynote from Perry Greenfield on contributions from STScl to open source packages in Python. I was really impressed with the amount of time and effort spent by STScl on developing such useful pytools. As an example maybe I could mention Imexam by Megan Sosey (http://imexam.readthedocs.org/imexam/index.html). We also learnt that all the software developed for the James Webb telescope will be 100% open source. Incredible!

Before going to lunch and kick off the last unconference sessions of the workshop, we have our last lightning talk session. We heard about the Astrophysics Source Code Library (http://ascl.net) a place to collect all the available astro packages. Erik Tollerud introduced us to Authorea (https://www.authorea.com), a collaborative writing tool that has a very elegant design. David Perez-Suarez showed a very practical tool called recipy (https://github.com/recipy/recipy) for those cases when you want to know how you created a plot or a table from an old script. Finally I was very impressed with PythonTEX presented by Stuart Mumford (https://github.com/gpoore/pythontex) which let you add python code to your LaTeX file. Really useful to add long complex tables or to update figures based on final comments from co-authors. A must that I will start using very soon.

The unconference sessions today were more to make sure that everybody knew what to do after the workshop was over. There was also a lot of sprinting time to finish code that started during the week. Loads of new issues have been opened to improve packages presented in the workshop. New collaborations that we all hope will bring amazing new features in the near future.

As an example of the dedication and hard work from the people here, the conference was suppose to finish at 4pm. At that time an unconference session was starting. Not because we were running late, but because we still wanted to do/learn/discuss more. In fact, productive discussions run until almost 7pm! I guess hunger in the end played an important role on actually finishing the workshop.

I would like to thank personally to all the participants for making this workshop an awesome experience. You all were incredible, always dedicating time to others helping develop new features/tools. And for those following us via twitter or our blogs and engaging us with questions/comments, thanks for being there all these days. Of course, I want to acknowledge-thank here the amazing job that the organisers (Thomas, Pauline, Magnus, Stuart and many others that helped) for creating such a unique experience. Now it is time to get things done. To put out there those codes that were born or matured during this week. Let’s keep the momentum going. It is time to do more and better science with modern and efficient tools.

Python in Astronomy – day 4

The week is coming to an end and unfortunately so is #pyastro15. If you were here, you could see that the atmosphere at the Lorentz Centre has changed over the past days. It is kind of obvious that we see the end of the workshop and we want to get things done. That means more hands-on sessions. More coding less talk. That does not mean that we did not have our daily share of talks, lightning talks, tutorials and discussions. There were loads of that, but everybody here wants to go home knowing that we definitively contributed to improve some packages and that we can move forward towards a more powerful and organise community open to astronomer/astrophysicist that uses Python.

The day started with Steve Crawford on reducing optical observations with Python. I would consider his talk challenging from the point of view that it was the first one after a conference dinner. I think we all understand what that means. The talk was fantastic, and two Python packages were introduced to us, ccdproc (https://github.com/astropy/ccdproc) and a #pyastro15 baby package called specreduce (https://github.com/crawfordsm/specreduce). As pointed out by Steve, IRAF is great and many of us have used it on our PhD thesis. We should be critical with IRAF but not attack it. Once that is said, this does not mean that we need to stop with IRAF if we think we can get more using new tools. This is what is intended with these two packages.

The second talk of the day was by Anne Archibald on testing statistical tests. As a user more than a pure coder, Anne talked about the importance of knowing your statistical test if you want to interpret your data right. You can read more about some topics of Anne’s talk on at http://lighthouseinthesky.blogspot.nl. For me it was great to see how relevant PAST (my software for timing analysis), could be to Anne’s analysis in particular, and also the possibility of adding part of it into astropy.stats. So there it goes more motivation to finish it ASAP.

We have two more talks today, one by Nadia Dencheva on GWCS which it is a very relevant work on  WCS using Python (sorry, I can’t find the link to the software) and SunPy (http://sunpy.org) by Stuart Mumford. SunPy makes the life of a solar physicist much easier by doing an enormous job on simplifying currently used code and bringing additional tools. I was very impressed on how well optimised is and if Solar Physics is your are I would really encourage you to start working with it right now.

As always, the lightning talks were fantastic, very dynamic and really useful to discover new things. For those not here, the idea is that in one hour, more or less, we have about 10 talks. So you can image how difficult it is to keep track with everything specially after 3 days of lightning talks. However, at the same time, there is no time to get bored or sleepy. Thus, please note I will be missing a few of them in my summary, and they will be biased towards relevance to my own interests.

Matt Craig showed us how to work with ipython widgets (an example can be found at (http://nbviewer.ipython.org/gist/mwcraig/2888d0a3790a81bd3391). That was really interesting and I will definitively start using them for my lectures. Becky Smethrust talked about the starpy package to study star formation history (https://github.com/zooniverse/starpy). She also gave us an update on the next two .Astronomy conferences (http://dotastronomy.com) that will be hosted in Sidney (November 2015) and Oxford (June 2016). I’ll try to go to the Oxford one considering how close it will be to Dublin. We then learned about CosmoABC by Emille Ishida (http://cosmoabc.readthedocs.org/en/latest/), a cosmology tool using Bayesian statistics. We also heard about Toyz by Fred Moolekamp (http://fred3m.github.io/toyz/) a really interesting web tool for “big data” where scientist can interact with their data in many different ways. If you didn’t know about this package, go and play with it, you will love it.

Curtis McCully introduced lacosmicx (https://github.com/cmccully/lacosmicx) a tool to clean optical images from cosmic rays. This software is really fast compared to previous tools available in other languages and as Curtis and I (mostly him actually, but I was there) tested later around 45 times faster than similar tools available in Python. Following on wrappers to create nicer plots from the previous session, Thomas Robitaille, a.k.a. astrofrog) showed us the capabilities of APLpy (https://aplpy.github.io). I will be playing with this package in the next few days so I might actually talk about it in future post. I think I am going to like it a lot.  Finally there were a couple of “rants”, one on education by Caroline Villforth and another one in “deluge of data” by Geert Barentsen (https://speakerdeck.com/barentsen/a-3-minute-rant-on-the-deluge-of-data-in-astronomy).

As I said, the lightning talks may be short, but they do give you loads of information.

In the afternoon we had much more unconference sessions than in previous days. People gathered in smaller groups and try to work on different things. There were talks about astropy.model, extragalactic software… Among the things done I would like to mention the tutorials on packaging (continuation from the previous day) and on git branching (http://t.co/ozZnq2Is4t) by Erik Bray, which were fantastic. Also, the great job done by Kelle Cruz and Pauline Barmy putting together resources for astro-computing education (http://www.astrobetter.com/wiki/astro-comp-ed). I would like to share also the tutorial on classes that Erik Bray kindly shared via twitter in case we could not have a session for it http://nbviewer.ipython.org/github/pylunch/tutorials/blob/master/embray/classes.ipynb.

This is likely the longest of the blog posts about the conference and it was probably one of the busiest days too. I go to the last day with mix feelings wanting the workshop to continue a few more days. I am very happy to be here and to have taken part on this amazing workshop with this incredible group of people.

Python in Astronomy – day 3

It is a bit late, but hopefully I can still make sense of my notes about our day 3 at the #pyastro15. We had the conference dinner this last afternoon which explains why I am writing this so late.

It is very interesting to see how the dynamic of the conference is slightly changing as we get to know each other more and more. More interactions, more ideas, suggestions… In conclusion, more progress and hopefully more results in the near future.

We started with a talk about Astroquery by Adam Ginsburg (http://astroquery.readthedocs.org/en/latest/), a really cool tool that can make querying for data almost painless. Really neat software, I have to say. Personally, I have started working with some other people here to add a few additional features. Hopefully we will get something done by the end of the workshop.

That talked was followed by a really engaging talk about LEMON, a photometry package (https://github.com/vterron/lemon) by Victor Terron. Apparently the installation is not easy at the moment, but hey, they are working on it and the software seems to work pretty well. One thing I learn in that talk was that you can do really cool presentations with reveal-js (http://lab.hakim.se/reveal-js/#/). It is a free package if you don’t need to work offline. After that, we heard about AMUSE by Ajern van Elteren (http://amusecode.org), a multipurpose software that tries to gather simulations from several areas. Quite ambitious!

Before the lightning talks, the morning had one final talk by Britton Smith about the yt-project (http://yt-project.org). This is a very cool visualisation tool for simulations that left me wanting to work on an area where I could actively use it. Really neat and as demonstrated with a cool animation by Adam Ginsburg, very easy to use (the whole animation was done with just 2 lines of code).

The lightning talks were as usually quite engaging and very useful. We learnt about the Trillian (http://trillianverse.org) project which tries to connect models-data-science in a painless way for the scientist. Essentially a way of doing more in less time and thus, giving us the opportunity of thinking out of the box and move further. Three other cool projects that caught my attention were Barak (https://github.com/nhmc/Barak) for spectral analysis, which is something that we are all very eager to get done in Python; Prefig (https://github.com/rjsmethurst/prefig), a plotting tool to make nice plots for posters and talks; and Ginga (https://github.com/ejeschke/ginga/), a 2-d visualisation tool for data.

With all these new projects my list of todo’s is getting quite large. It is incredible the amount of cool software being currently developed with Python. It seems that we are going in a very good direction to become more and more effective on our analysis. And by sharing the code with the whole community, the sky will be the limit. To make sure that we all get the right acknowledgement, there is a github project created by @astrofrog that you can find at http://astrofrog.github.io/acknowledgment-generator. It is really important to acknowledge those that help us doing the science that we want!

I read on twitter that we seemed a bit more quiet today. The reason for this is called unconference sessions. We are moving more into sessions where we do stuff rather than just talk about it. I am finding this very efficient and engaging. Of course, since our hands were busy writing code or tutorials, we could not tweet about what it was happening. But we also had a few sessions where we learnt new stuff. Few examples: how to create a setup.py and packaging (see notes at bit.ly/pyastro-packaging), and how to test your code and make it faster (profiling it). Let me add here another awesome project by @astrofrog for doing these kind of tests, called psrecord (https://github.com/astrofrog/psrecord). Tutorials on pyCharm and yt were also quite popular. There was also a session talking about photometry packages written in python regarding to their current status and what needs to be done in the near future. And probably I am forgetting some other unconference sessions, sorry!

As you can see, we don’t have time to get bored. These are really busy days at the #pyastro15 workshop, but I can probably speak for everyone if I say that time flies and we are loving each minute here chatting, working, collaborating… you name it!

I’ll see you in my summary of day 4 and if you cannot wait until then, follow my twitter feed or the #pyastro15 has-tag.

Python in Astronomy – day 2

So day 2 of the Python in Astronomy workshop is now over. Time to prepare ourselves for day 3, probably as good as the previous 2 days have been.

Day 2 just confirmed what day 1 hinted, this workshop is like no other one I have ever been and it keeps getting better and better. But it is not only the topics covered in the conference what makes this workshop worthy. The people are really friendly and there is an amazing atmosphere to bring new ideas and comments to the table and discuss them freely without worrying about being judged.

I think Pauline Barmby (@PBarmby) has done an awesome job summarising the day with links to google docks and more. So I don’t think it is worth repeating everything here when she has done such a nice work. Thus, I encourage you to go and check her blog for a detail description of each of the sections (see it here pbarmby.github.io/2015/04/21/Python-in-Astronomy-Day-2.html).

For me, it was a really positive day. I got to give my “Lightning talk”, a 5min talk, about my python software called PAST for timing analysis. People seemed to welcome my talk and I had the opportunity to chat with few people about it later. It seems that there is here a lot of potential for new collaborations at different levels, new potential users and probably some integration with other packages. I guess I will need to give it a final boost.

Better documentation seemed like the theme in the morning. A very valid point that I should start using more and more even if it takes a bit more time to have my scripts completely ready.

From the talks in the morning, I was very impressed with Althea’s one on Meteor showers (meteor.uwo.ca/research/allsky/overview.html), and Edward’s on telescope networks (lcogt.net). Really interesting works where python seem to play a huge role. It was great to hear a bit more about arXiver. I knew about this site, but never really got into using it. After listening to Vanessa, I think I should give it a try. It sounds really interesting.

Of course, the talk by Erik on Astropy was very important. Astropy as a community is playing a very important role on bringing python to our community and astropy as a software package is getting more and more important (apparently over 100 citations already and more than 120 contributors).

The unconference sessions were very useful. The astropy.stats one ended up being an incredible brainstorming session where everybody said what they would like to see included in the module and why. Also it helped as an initial step towards what it end up being another uncoference session on astropy tutorials. Don’t be surprise if the astropy site has many more tutorials that it currently has after this week. We also set the barebones of a new site to gather educational resources to help learning how to code, specially in python. That was really interesting, too!

Day 2 finished and we are getting almost half way there. It is going to be sad when all this ends, but I can see that we are building a stronger python community and that the goals are very clear. And that the group of people here will be quite active promoting astropython or helping building new packages for a long time.

There and back again — day 1 Python in Astronomy workshop

After a huge pause of a year, I finally have the time (more or less) and energy (definitively) to come back to my blog and give it a big boost. I guess this is what happens when one has his/her first child, time flies!  Anyway, I think there is no better occasion to start blogging again (hopefully more continuously) than to use my presence at the “Python in Astronomy” workshop to share with you all the great things that are happening this week here in Leiden.

We are about to start day 2, so before we get into all the busy schedule it may be time to summarised what happened in day 1. After a short introduction from the people at Lorentz Centre and the organisers of the conference, we started with the “flash introductions”. These are 1min talks where everyone said who they were, where they were coming from, what they do and what they expected to get from the workshop. We are around 55 participants, so we were listening to people trying to summarise their experiences for almost 1 hour. You may think that it could be quite tedious or boring, but reality could be more different. It was dynamic, fun, exciting and as someone tweeted later, it was incredible that the buzzer indicating that the time was out did not go off for anyone. The result was incredible. After one hour of workshop we already kind of know everyone and we already had a reason to talk to some of the other participants based on common interests.

The morning then continued with a really good talk from Kelle Cruz from Astrobetter on how to communicate and engage new people on using Python. One of the conclusions of her talk was in the form of a question: why should we keep trying to convinced those stubborn while we have a brand new group of people ready to learn something new and modern? I could not agree more with her. And a second very good point was: do stuff or organise groups to do stuff, do not just wait there until someone magically does it for you.

After that, we had lunch and the afternoon was for unconference sessions. These are sessions that are planned within the day based on what we, the participants, need or want to do. You may think that this could be quite a chaos, but the organisers know what they are doing. The feeling of the whole workshop is that it is incredibly well organised. Probably one of the best conferences/workshops in terms of organisation I have ever been. So how it works is that people go to a whiteboard and write things that they would like discussing or help others with. As examples, there were 2 tutorials on GIT (fantastic both, by the way), a discussion about schedulers and observation planners for telescopes, discussion about software to do spectral analysis, Monte-Carlo sampling… and many more. They were all so good that almost everyone I know wished we could have clones to go to all the parallel sessions. Luckily some people managed to share notes or links about things that were mentioned. Another possibility was using the break to try to find some of the people that run the other sessions and ask them about some of the points covered while they were fresh in their minds.

I cannot wait for today’s second session, specially to know what kind of things we will be doing in the afternoon. Besides the already known talks, we will have a few “lightning talks” which are 5 min talks.  Keep your eyes on twitter and the hash-tag #pyastro15 to follow in real time what is going on here.