Katie Bouman and the Black Hole

All of this from the secret chatroom:

First:

These pictures posit an incorrect interpretation.

He uploaded a lot of data files, which github counts as code. The project is under 100k lines of code.

The commit messages are suspect, but if they say she developed the algorithm then I am inclined to believe it because there’s no way to counter that.

She has been involved with the project for 12 years (since she was 17) but only recently started contributing. No good assumptions can be made in that age range.

Also, Andrew Chael is gay and runs an internal LGBTQ support program or something.


(Pruned) Reddit thread:

Second:


$ gitinspector 
Statistical information for the repository 'eht-imaging' was gathered.
The following historical commit information, by author, was found in the repository:

Author                     Commits    Insertions      Deletions    % of changes
Andrew Chael                   477         89748          82408           85.19
Chi-kwan Chan                  159          2135           1562            1.83
Hotaka Shiokawa                  4          1004              4            0.50
Joseph Farah                    26           630             55            0.34
Katherine Bouman                79          1453            497            0.96
Katie Bouman                    27          3134            411            1.75
Lindy Blackburn                  5            26             12            0.02
Maciek Wielgus                  34          1539            133            0.83
Michael D Johnson                3            46             10            0.03
Michael Johnson                267         11762           1997            6.81
danielpalumbo                    8           144             55            0.10
dpesce                           1            13              0            0.01
klbouman                        88          2218           1062            1.62
palumbophysics                   4            10             10            0.01
Below are the number of rows from each author that have survived and are still intact in the current revision:

Author                     Rows      Stability          Age       % in comments
Andrew Chael              25365           28.3         12.3               13.17
Chi-kwan Chan               898           42.1         17.3                3.79
Hotaka Shiokawa              93            9.3         19.5               10.75
Joseph Farah                200           31.7          7.3                9.00
Katherine Bouman           2316          159.4         18.2                7.94
Lindy Blackburn               1            3.8         17.8                0.00
Maciek Wielgus              692           45.0          6.7                8.24
Michael D Johnson            32           69.6         18.7                6.25
Michael Johnson            5467           46.5         13.0               11.19
danielpalumbo                80           55.6          9.0               11.25
dpesce                       13          100.0          5.4                0.00
klbouman                   1506           67.9          5.7                6.91
palumbophysics                5           50.0         14.6                0.00

The stats on github are auto-generated and incorrect, Katie Bouman has 225 commits, not 90.

$ git shortlog -nse
   761  Andrew Chael 
   314  Michael Johnson 
   206  Chi-kwan Chan 
   105  klbouman 
    52  Katherine Bouman 
    38  Maciek Wielgus 
    33  Katie Bouman 
    27  Joseph Farah 
    11  Katherine Bouman 
     9  Hotaka Shiokawa 
     9  danielpalumbo 
     6  Lindy Blackburn 
     5  Katherine Bouman 
     4  Katherine Bouman 
     4  Katherine Bouman 
     4  Katherine Bouman 
     4  Michael D Johnson 
     4  palumbophysics 
     2  Katherine Bouman 
     2  Katherine Bouman 
     1  Katherine Bouman 
     1  Katherine Bouman 
     1  dpesce 
     1  klbouman 


Bouman’s work was used in section 2.2.2, where it was one of the three considered for the same.

Section 5.1:

Prior to the 2017 observations, we organized a series of "imaging challenges" that used synthetic data to assess how conventional and newly developed imaging algorithms would perform for the EHT (Bouman 2017).
To compare EHT 2017 results among teams while keeping submissions blind, we built a website that allowed users to independently upload images and automatically compare them to the ground truth images and submissions from other users (Bouman 2017).
And she was on one of the teams using CLEAN, hence the software contributions. That is all she has done.

Excluding the image comparison, Andrew Chael was also engaged in the same roles.

And he actually had more input than Bouman on the imaging part, and also did more work.

For the other team:

smili.git$ git shortlog -nse
   114  Kazunori Akiyama 
    19  moriyamakotaro 
     4  Fumie Tazaki 
     2  Kazu Akiyama 
     2  ftazaki 

It seems like all the participating women have a really nasty habit of not using their own computers.

Kazunori Akiyama is more experienced and has been working on projects like this since before 2009. Not only has he contributed heavily to one of the imaging groups, he was also significantly responsible for the direct reads from the EHT.

Although the significance of this “image” is debatable, Bouman does not deserve all the credit, and certainly does not deserve even the amount she has received.

There are other people who deserve much more, and one blurry photograph (of the team) released by Bouman does not mend anything.


Fox:

“Public figures from Washington to Hollywood learned her name. And some advocates, familiar with how history can write over the contributions of women, quickly moved to make sure she received the recognition she deserved. In their eagerness to celebrate her, however, many nonscientists on social media overstated her role in what was a group effort by hundreds of people, creating an exaggerated impression as the photo was shared and reshared.” [nyt]

A good example how an ideological modus operandi, and a systemic campaign to create false impressions is being framed as a completely unintended accident.

Of course, the fact that these “advocates” weren’t no-names but people writing articles for major media outlets (or even the ones deciding what gets written) is swept under the rug.

The nyt article does appear like damage control.