[ad_1]
Last week, fans of cool astronomical phenomena (read: almost everyone) rejoiced when an international team of scientists released the very first image of a black hole. For astrophysicists, software engineers, philosophers and mathematicians who worked on the Event Horizon telescope that captured the image, the announcement was an unprecedented milestone.
Katie Bouman, a computer scientist from the Event Horizon Telescope team, hid her radiant smile from her hands as she watched the monumental rendering. Bouman had many reasons to smile: the image was created using petabytes of data assembled using CHIRP, an algorithm that Bouman worked on. And Bouman had long served as a public face to the computer imaging aspect of the Horizon Event Telescope, delivering a TED talk about the project in 2016.
But shortly after the announcement, online stalkers created fake Instagram accounts for Bouman, started creating furious links on Reddit and Hacker News, claiming that she had not done so much to help the project that she had been and that she had produced long tirades on YouTube. , all with the aim of discrediting his contributions to the project.
In computer science, the alleged inferiority of women is unfortunately a fairly widespread belief, sometimes expressed by software engineers of large technology companies. But here, rather than using the psychology of the crackpot, these trolls have used bad faith interpretations of Github's public repositories, which programmers use to host source code for collaboration and contribution purposes, in order to to legitimize their content.
The main argument that unites these attacks was apparently based on data: according to a Github repository related to the project, astrophysicist Andrew Chael would have written: "850,000 of the 900,000 lines of code … in the # 39, historical algorithm of black hole images. A ridiculously simplistic evaluation for a software library built with open source licenses, which allows all users in the world to use and adapt their code without worrying too much about the number of shared lines of code during of the process.
Microsoft software engineer Safia Abdalla, who is responsible for the nteract open source interactive computer library, told me via Twitter DM that one of the main tenants of open source is a community-based code approach.
"Many interesting innovations come from open source because innovative and interesting technical ideas are developing in open spaces," she said. "Any open space must have a healthy community to foster collaboration and execution on ideas."
"We could spend six hours together solving a very difficult problem … and at the end of the day, there could be five lines of code modified by one person"
Abdalla noted that open source programming feeds on many contributions that never appear in a GitHub repository or in the code itself, such as publishing plans, reviewing collaborators' work, and writing documents. . In fact, the lack of collaborative participation is one of the biggest threats to many valuable open source libraries. She said the criticism of "lines of code" was reductive: "A person can add 20 lines of code to a project that adds relatively little value, but another person could add a single line of code to solve a bug. major. You really have to look at open source with more nuance than lines of code. "
For programmers even having a basic knowledge of software development, it should seem obvious that the number of lines that one engages on Github is barely correlated to success as an engineer. In principle, this type of evaluation is completely detached from the reality of software development.
Judging the skills of someone based on their code volume violates fundamental principles of software development, such as "DRY" or "Do not repeat yourself," a philosophy that values reusable software over code redundant. But beyond the technical criticisms of "good" and "bad" codes, such judgments simplify years of collaborative reflection of a few minutes on Github. For those who are most intimately involved in the Event Horizon telescope, it could not be further from the truth.
Andrew Chael, whom the stalkers of Bouman wanted to portray as the "real" hero of the black hole image, quickly took up Bouman's defense on Twitter, reinforcing the level of teamwork inherent in the project. "Although I wrote a lot of the code for one of these pipelines, Katie has contributed immensely to the software. it would never have worked without his contributions, "he said. "It was a team effort involving many beginning scientists, including many women scientists. Together we improve each other. the number of commits does not tell the whole story of who was indispensable. "
Astrophysicist Michael D. Johnson, who worked with Chael and Bouman on the Event Horizon telescope, also found that these individualized credit assignment attempts were terribly far removed from the reality of collaboration on the Event telescope Horizon.
"Of course, we have never compared the contributions of everyone by lines of code," he said on the phone. "We could spend six hours together solving a very difficult problem … and at the end of the day, there could be five lines of code modified by one person."
"All this work is so closely linked. Trying to evaluate it according to the code here or there is so wrong, "he added.
Seeking to credit a person for developing an open source technology is not only extremely simplistic from a technical point of view, it is also contrary to the goals and ambitions of the open source community.
And the development of the Event Horizon telescope was not solely related to brain power. Like many code bases, the "eht-imaging" repository at the center of this debate is indebted to open source technologies, starting with Python, the language of the project. Johnson said the use of open source technologies such as Git and Python "simply opens up many scientific opportunities that would never be possible in a language like C, where it would take years of professional developer development to do the same kind of things ".
Without Python modules such as numpy and matplotlib, community libraries providing tools such as computer science and tracing, workdays would have taken "months or even years of effort," said Johnson.
Of course, these scientists have used these technologies to create proprietary algorithms for solving dense and complex problems related to computer imaging of black holes; The use of open source libraries certainly does not discredit years of research and development. But looking to credit a person for the development of an open source technology is not only extremely simplistic from a technical point of view (otherwise we count the millions of lines of code written in the only numpy library), but it is contrary to the objectives of the open source community. ambitions.
In scientific fields that rely on data, sexism can be considered legitimate when it is expressed in cold, insensitive language and in percentage points. But anyone with a basic knowledge of modern computing should quickly realize how dangerous and trolled these trolls are when they arm metadata from public Github repositories.
Although most members of the scientific community quickly rejected these criticisms and acknowledged the monumental scientific achievements of the Event Horizon Telescope team, thousands of people gathered on YouTube, Reddit and HackerNews to develop this false story. When contributions are reduced to individual lines of code, it threatens to derail the ethics of open source, a community that at its best is a symbol of the free exchange of ideas and information that the Internet is committed to achieve.
[ad_2]
Source link