Independent open-developed projects have been growing rapidly in astronomy and related domains - e.g., projects like Astropy, Sunpy, or Scipy. Large astronomy organizations or missions regularly make use of the software products of these projects, and at least in some cases, contribute back. This has the obvious benefit of being able to do more for the scientific community with less effort per mission. However, there are some challenges in this process - inflexible or agressive deadlines by funded organizations may conflict with the timelines of open source projects, science vs engineering cultural conflicts may make contribution more difficult, mismatch between the needs of the general community and a specific mission, etc. This BoF session is aimed at discussing exactly these tensions (or others that are brought up in the discussion). The goal is to draft a set of suggestions that interested parties can take to mission/institution leadership to help resolve these tensions, or at least be able to enter into them with eyes open.
Lightning Talk Slide: B1-lt.pdf
What's happening with FITS? What data formats are being used for new instruments? How can we move the new formats toward widespread readability? Together, both data producers and data ingesters will discuss how our solutions can become more compatible.
Lightning Talk Slide: B2-lt.pdf
Machine learning has become a key tool to analyze and process complexly structured large datasets. This BoF will be different then the usual BoFs, that aim for discussing a specific topic. Due to the large request in understanding and learning machine learning techniques at the last ADASS, this 90 minutes will be used to introduce basic concepts of machine learning. Based on a few examples, different machine learning models will be introduced and their application will be shown. In contrast to a tutorial session, there will be space for questions and discussions of typical challenges and problems.
At the end of the BoF the participants will have a basic understanding what machine learning is about. To enable the participants to further learn about machine learning and to allow for a broader overview, a list of good online-sources will be provided.
For teaching material, download ADASS2018_bg2mlBoF.tar.gzAs data citation and data preservation have gained increased prominence in recent years, we feel that it is important to continue the conversation that was started during the DOI BoF at ADASS XXVI and expand its scope. To be more specific, some of us, as data editors of the AAS Journals, see steady growth in authors using both high-level science and/or multi-purpose generic repositories for storing archival, publication related data & other materials. We also anticipate an increase in policies that encourage/require authors to make such data persistently available at publication. All of us notice many hiccups in the workflows for data archiving by authors with extant tools and are interested in how science platforms and archives plan to support the long term preservation and citation of data produced by their users. The specific questions this BoF might discuss include: how are astronomy archives supporting author data citation today; how could data citation via a multi-purpose science platform expose and support the concept of transitive credit (Katz 2014) to ensure that the software used on a platform is properly attributed and/or cited; what are the use cases for data citation -- what things get saved by whom and when -- regardless whether that data were produced on a platform or housed in monolithic space mission database; how do we ensure the discoverability of preserved data objects through the creation and indexing of uniform rich metadata. The goals of this BoF are then to surface use cases, share implementation ideas, and move us towards policies that would better enable the citation and reuse of the data/software developed as part of the research process. Katz (2014): https://doi.org/10.5334/jors.be
Lightning Talk Slide: B4-lt.pdf
Astronomical software development is typically done in teams, especially when associated with large instrument and data archive projects or projects like AstroPy. There are many kinds of teams, small, medium or large, single- or multi-institutional, formal or informal. How well do these teams function? How can we even measure this? This BoF session will examine different types of teams and ways that we can make our teams work more effectively. We will lead with three presentations from different teams, followed by a moderated group discussion.
Lightning Talk Slide: B5-lt.pdf
The recent multi-messenger observations of gravitational-wave and high-energy neutrino sources together with electromagnetic signatures have brought completely new ways for observing the Universe. These are promising a future where advancing physics and astronomy will be enabled by combining observations and data from all across the electromagnetic spectrum, gravitational-waves and neutrinos. We consider the challenges the field is facing in fully utilizing data for multi-messenger astrophysics. Such data come from heterogeneous detector networks and standards and their analysis is often time-critical to guide further observations. In this area, science capabilities depend on the interplay among observation, theory and computational/modeling work. Advances in data science and computing present additional opportunities and considerations in analyzing such data. We invite all ADASS participants in this Birds of a Feather session to engage in an informal discussion on the challenges and opportunities in data analysis for multi-messenger astrophysics.
Lightning Talk Slide: B6-lt.pdfBoF Slide: B6.pdf
Topics for this Unconference Birds of a Feather session will be chosen from those identified/requested by attendees before the session. Small-group conversation at round tables will allow discussion of numerous topics by interested parties; the tables have signs identifying what people at the tables want to talk about. The session will open with one slide showing the selected topics and will close with one-minute reports from each table.
Lightning Talk Slide: B7-lt.pdf