Norman Gray, University of Glasgow (norman@astro.gla.ac.uk) and Bob Mann, University of Edinburgh (rgm@roe.ac.uk)
The ADASS conferences provide a regular focus for those engaged in astronomical computing. ADASS proceedings provide a very valuable record of each conference, but imperfectly record the activities of the ADASS community, for a number of reasons: (i) appearing after a year, they often present out-of-date snapshots of rapidly-developing projects; (ii) being unrefereed, there is no quality threshold, nor are authors pushed to justify and elaborate where needed to provide the best account of their material; (iii) being tied to the annual conference cycle, projects are reported upon when the opportunity arises, not when they have reached appropriate milestones; and (iv) having restricted page lengths, topics receive only brief coverage.
This matters for at least two reasons. Firstly, material from most ADASS conference papers will be published nowhere else, so valuable technical lessons risk being lost. Secondly, as more people pursue a career in astronomical computing, it becomes more important that they have a means of recording their attainment and a track record of refereed journal papers, with associated citation statistics, is what potential employers will most readily understand.
Open source publishing systems make the establishment of a community-driven astronomical computing journal possible, but is it necessary? Would mainstream astronomy journals publish more technical papers if they were submitted? Are there computing journals that would welcome these papers? What are the benefits of refereeing to this community and would it devote the time needed to referee? Does the lack of refereed publications hinder the career progress of its members?
This BoF will address these and related questions. Most time will be devoted to contributions from the floor, rather than prepared talks. A few contributions will be sought to initiate and shape debate, and those interested in making one are invited to contact the organisers.
A copy of the introductory presentation from the BoF is available here, and notes from the discussion is available here.
Mainstream astronomy journals publish few papers on the sort of astronomical computing projects that form the bread and butter of ADASS conferences. In order to see whether this is a direct result of the publication policies of the journals, or whether members of the ADASS community simply don't try to publish in them because they think their papers will be rejected, we sent abstracts from five papers from the 2009 ADASS Proceedings to the editors of eight journals and asked them to comment on whether each would be considered within the scope of the journal.
The eight journals were:
(N.B. the editors of PASP and AN were contacted after the ADASS conference.)
The abstracts they were sent were as follows:
With the recent advent of time domain astronomy through various surveys several approaches at classification of transients are being tried. Choosing relatively interesting and rarer transients for follow-up is important since following all transients being detected per night is not possible given the limited resources available. In addition, the classification needs to be carried out using minimal number of observations available in order to catch some of the more interesting objects. We present details on two such classification methods: (1) using Bayesian networks with colors and contextual information, and (2) using Gaussian Process Regression and lightcurves. Both can be carried out in real-time and from a very small number of epochs. In order to improve classification i.e. narrow down number of competing classes, it is important to combine as many different classifiers as possible. We show how this can be incorporated in a higher order fusion network and tied with optimal follow-up.
WorldWide Telescope has grown from a standalone visualization platform to a rich set of components that can utilized by portals, data providers or research projects to allow rich data access to both catalog and image data. The WWT Client also provides SAMP enabled interoperability allow a suite of data services across client, web and server.
The SIMPLE Imaging and Mosaicking PipeLinE (SIMPLE) is an IDL based data reduction environment designed for processing optical and near-IR data obtained from wide-field mosaic cameras. It has standard functions for flat fielding, sky subtraction, distortion correction, and photometric and astrometric calibrations. One of the key features of SIMPLE is the ability to correct for image distortion from a set of dithered exposures, without relying on any external information (e.g., distortion function of the optics, or an external astrometric catalog). This is achieved by deriving the first-order derivatives of the distortion function directly out of the dithered images. This greatly help to produce high accuracy on astrometry as well as preserve image sharpness in the mosaicked/stacked image. Despite being designed toward a general reduction environment, the current distribution of SIMPLE has two highly optimized packages, one for the Wide-field InfraRed Camera on the Canada-France-Hawaii Telescope and the other for the Multi-Object InfraRed Camera and Spectrograph on the Subaru Telescope. SIMPLE has produced excellent (photometrically and astrometrically) wide-field images from both cameras. Users and the author of SIMPLE are also developing optimized SIMPLE pipelines for other mosaic cameras such as the Subaru Prime Focus Camera.
In recent years Java has matured to a stable easy-to-use language with the flexibility of an interpreter (for reflection etc.) but the performance and type checking of a compiled language. When we started using Java for astronomical applications around 1999 they were the first of their kind in Astronomy. Now a great deal of Astronomy software is written in Java as are many Business applications. We discuss the current environment and trends concerning the language and present an actual example of scientific use of Java for high-performance computing: ESA's mission Gaia. The Gaia scanning satellite will perform a galactic census of about 1000 million objects in our galaxy. The Gaia community has chosen to write its processing software in Java. We explore the manifold reasons for choosing Java for this large science collaboration including recent sucess using the Amazon Cloud for AGIS.
The last few years have seen the emergence of a new feature in several astronomical visualization tools : the interactive sky browser supporting immediate panning and zooming. World Wind, Google Sky, World Wide Telescope, Wikisky, Virgo and now Aladin, all these tools have in common a view of the sky based on a hierarchical multi-resolution sky tessellation. The aim is to load and draw the good "pieces" of the sky at the good resolution as fast as possible, according to the current user sky view. The goal is the same but sky indexing solutions differ significantly and do not offer the same capabilities in term of performances, underlying data base complexity, available projections, projection distortion, pixel value access, graphical overlays, etc. Actually, most of the tools offer false-colour skies with a unique simple projection. But this new feature can be used not only for providing a sky background, but also for accessing and analyzing pixel data in the same way that astronomers commonly use FITS images for doing science. In this talk, we will present how Aladin is using an HEALPix sky tessellation for building a powerful sky data base. We will present the arguments in favor of HEALPix, notably: - The intrinsic qualities of HEALPix for implementing fast pixel algorithms such as convolutions, Fourier analysis, wavelet decomposition, nearest neighbor searches, topological analyses... - The hierarchical structure of the sky directly mapped in a simple directory tree, allowing immediate usage for local data; - The projection methods for reducing as much as possible the distortions notably at poles and at the "sky borders"; - The available libraries, and especially the Java package supporting deep sky resolution; - Last but not least, the direct usage for current mission data such as Planck; - etc. We will also discuss about the compatibility/interoperability between all these tools and how we could avoid to duplicate these data bases and implement efficient collaboration. This might open the door to a future VO standard describing this new way to explore the sky.
Responses were received from the editors of six of the eight journals (all except Experimental Astronomy and Astronomische Nachrichten), as follows:
and are summarised in the following table:
|
|
ApJ |
MN |
A&A |
ESIn |
DSJ |
PASP |
|
Bayesian methods to classify transients |
Y? |
N |
Y? |
Y |
Y |
Y |
|
WWT progress report |
N? |
N |
N? |
Y |
Y |
Y |
|
IDL pipeline for image processing |
Y? |
N |
? |
N? |
Y |
N? |
|
Use of Java & HPC for Gaia |
N |
N |
N |
Y |
Y |
Y |
|
HEALPix in Aladin |
Y? |
N |
N? |
Y |
Y |
Y |
N.B. These results do, of course, reflect the scope of the journals, and not the quality of the papers concerned.