Context & Scale

Here’s the ongoing 10QViz community CONVERSATION about Context & Scale

glue & the Webb Telescope ❤ a match made for the heavens!

How was this great match made?

On April 23, 2012 in an email, then-Space Telescope Science Institute Director Matt Mountain asked Harvard Professor Alyssa Goodman: 

“Presume through Alberto [Conti] you will touch base with the other JWST folks looking at IFU data visualization like Tracy Beck, Massimo (our Acting Head of the JWST[?].”  

Three days later, on a trip to Baltimore from Boston, Goodman was in Mountain’s office at STScI, where he showed her his copy of  her “Principles of High-Dimensional Data Visualization in Astronomy.”  Conti, then a NASA “Innovation Scientist,”  had shared Goodman’s draft with Mountain, knowing how relevant the “principles” in the paper could be for JWST data in the future.   To Goodman’s complete surprise, about 5 minutes into the conversation, Mountain offered Goodman (who was not actively seeking funding at the time) “something like a million dollars” to make the “glue” software described in the draft “real enough to use.” 

Note: Astronomers often still call the “Webb” or “James Webb” space telescope by its NASA acronym, “JWST.”

Why were these astronomers so interested in glue + JWST? 

The Webb telescope doesn’t just take images.  It can take a spectrum, breaking up light into constituent colors, at many many positions within an image at once, using a device called an “Integral Field Unit.”  The resulting data format, which has “x-y” positions on the sky, plus a “z” axis that corresponds to wavelength, is called a “spectral line image cube.”  Astronomers trained to use radio telescopes, including Goodman, have used such cubes for decades.   Goodman and her colleagues designed glue to exploit both high-dimensional data (e.g. cubes) and the principles of “exploratory data analysis” shown in glue’s logo.   (The red-highlighted points and regions in the glue logo are all coordinated, in that salient values selected in any open display of data are also selected, live, in others.)

What’s glue done in a decade?

Now, ten years later, the glue software environment is a robust open-source ecosystem that underlies all of Jdaviz, the web-based analysis tools being provided to scientists as the way to analyze JWST data. Thanks to initial and ongoing support from the NASA-JWST program, as well as from the National Science Foundation and the Moore Foundation, the glue exploratory data analysis tools are now are now used in many astronomical investigations, in genomics, and in many other contexts.   

Recent astronomy-related discoveries made using glue include the discovery of the Radcliffe Wave and the Perseus-Taurus Supershell, and the star-forming significance of the Local Bubble around the Sun.  glue has also been used to produce the first augmented reality figures published in a major astronomy journal.

glue is also used to teach data science.  At the high-school/community college/college level, it’s a key element of the infrastructure powering the “Cosmic Data Stories” project of NASA’s Science Activation Program.  And, for more advanced data scientists, glue is being used to train data scientists in astronomy, for example in the “Seeing More of the Universe” YouTube series created by Alyssa Goodman for NSFs Rubin Data Science Fellows program. 

About Jdaviz

Jdaviz offers four special packages intended for different specific purposes.  All of the packages use JupyterNotebook, JupyterLab, glue, and many Astropy functions to accomplish their goals. The “Glupyter Framework Overview” page on the Jdaviz website gives a good summary of how glue-jupyter (also called “glupyter”) is used, and can be extended, within the jdaviz environment. 

The four packages that comprise Jdaviz are called “Imviz,” “Cubeviz,” “Mosviz,” and “Specviz,” and super-short descriptions of each, from the Jdaviz website, are shown below.   For power users’ reference, glue outside of Jdaviz can integrate functionality across all of the specific tasks accomplished in these four tools, simultaneously.  See the glue website or these online demo and training videos for more on how to use glue in its most flexible forms.

Imviz is a tool for visualization and analysis of 2D astronomical images. It incorporates visualization tools with analysis capabilities, such as Astropy regions and photutils packages.

Cubeviz is a visualization and analysis toolbox for data cubes from integral field units (IFUs). It is built as part of the Glue visualization tool. Cubeviz is designed to work with data cubes from the NIRSpec and MIRI instruments on JWST, and will work with IFU data cubes. It uses the specutils package from Astropy.

Mosviz is a quick-look analysis and visualization tool for multi-object spectroscopy (MOS). It is designed to work with pipeline output: spectra and associated images, or just with spectra.

Specviz is a tool for visualization and quick-look analysis of 1D astronomical spectra. It incorporates visualization tools with analysis capabilities, such as Astropy regions and specutils packages. Specviz … supports flexible spectral unit conversions, custom plotting attributes, interactive selections, multiple plots, and other features. Specviz notably includes a measurement tool for spectral lines which enables the user, with a few mouse actions, to perform and record measurements. It has a model fitting capability that enables the user to create simple (e.g., single Gaussian) or multi-component models (e.g., multiple Gaussians for emission and absorption lines in addition to regions of flat continua).

“Save the pies for dessert”?

Included in Stephen Few’s very interesting visualization blog (perceptual edge) is the provocatively titled “Save the Pies for Dessert” post.   Pie charts are notoriously bad for perceptually judging magnitude. Here is an annotated excerpt from Few’s post, giving just one example of how hard it can be to judge scale using pie charts…

Not everyone hates pie charts, though…for example, here is a blog post from “Junk Charts” on the downside of discouraging pie charts.

Bonus: An amusing pie chart which shows the shadow illusion featured in the Categories question is this fascinating little image. Once you see the pyramid, you cannot unsee it:


Pyramid pie chart art.

I did not manage to identify the original maker of this -sort of- meme at this point, if you know, please tell me in the comments. Image above is copied from Rebecca Barter here:

Maybe a treemap would be better?

Consider this infographic about imprisonment, from this article on the American Legislative Exchange Council blog.   Most people would look at it and find it very engaging and attractive, which it is.  But, as a visualization expert, one wonders if the odd coloring variations in the outer ring of the main figure and in the “Juvenile” block at right, which just show how the larger wedges (categories) divide up more finely  (into sub-categories) wouldn’t be better shown in a Tree Map, using the ideas about showing hierarchical categories proposed by Ben Shneiderman in the 199os.   A Tree Map version of these data would almost certainly show the area of sub-categories and categories relative to each other (context) better than the snazzy graphic shown here.

Why (and what is) a “3D PDF”?

For hundreds of years, scientists have published their results in scientific journals that were printed on paper.  Today, though, most journals have gone entirely online.  Articles less and less frequently printed out and read on paper, so why should they still look and funciton exactly the way they did in the 1600s?

Josh Peek and I, and our colleagues wrote a fully online paper presenting The ‘Paper’ of the Future back in 2014, which highlights (with embedded demonstrations) many of the technologies available to scientists publishing today, and in the near future.   One particularly important technology–“3DPDF”– discussed in that paper of the “future” was actually first deployed in a Nature article by my “Astronomical Medicine” collaborators and me, way back in 2009.

Our challenge was to show the difference between two “segmentation” techniques used to define salient structures inside of star-forming regions.  The science isn’t important here (sorry).  What’s important is that we wanted to offer the “reader” multiple, interactive, views of high-dimensional data, inside of a journal article.

To see the PDF in action, take a look at this video, or download the “nature_demo” file and open it, on any Mac or PC, with an Adobe PDF viewer of any kind (not Preview).

Other authors (e.g Peek 2012) have since published methods for creating these 3D PDFs using free software, and a (perhaps too small!) number of authors have now embedded these 3D images inside of the scholarly articles.   Even though interactive images are clearly seen to add value to articles, they are not (yet) widely used.  3D PDF as a format may be short-lived, as articles move more and more to a fully online environment, where other (e.g. javascript-based) technologies can offer superior options.  BUT, the general idea of embedding data and interactive views of it, be they “3D” or not, is extremely valuable, and we will return to it in future posts–for now go have a look at The ‘Paper’ of the Future (Goodman at al. 2014).