Werden wir Helden für einen Tag

Home | About | Archive

OKR #9

Posted on Jun 18, 2023 by Chung-hong Chan

Previously on this blog: my objectives / adjustment; actual OKR #1, #2, #3, #4, #5, #6, #7, #8.


The badge I never wore at the conference. This photo was taken before I dumped it somewhere in Toronto.

I don’t want to write about the world and myself. Let’s go directly to the OKR.

Think out loud

I wrote several blog posts in the first half of 2023. And most of them are about R (Docker, Bioconductor, research compendium, and two posts about R’s C interface). Three are about tiling window managers. One is about Rust. I enjoyed writing these posts. Somehow, they are more spontaneous than the ones I wrote in the second half of 2022 (most of them are about emacs, because of Advent of emacs). And they documented the technical things I have learned.

I have some new drafts in my silo, which are melancholy ones talking about my recent thinking. Of course, there are still some older unfinished drafts, as old as a post about the ICA conference in 2020. I never know if these drafts would one day be finished. Probably forever will they be in the silo.

Contribute to open source

Let’s talk about Projekt 71. It is progressing and stagnant at the same time.

It is progressing because write_ods is 100x faster. This is a great performance gain, but that’s not my contribution. Dr Detlef Steuer made that contribution.

It is stagnant because lower-level language rewrite of read_ods has very little no progress in substantive terms. However, I am equipping myself with lower-level language knowledge. My first choice was Rust because of the ready-made crate calamine. But after reading about how CRAN treats Rust (and of course, the hacky nature of making Rust in an R package running on Windows), I think it’s better to invest my time in the trusty C/C++.

Another reason for the stagnancy is that I am too native to assume I can put the development of other R packages under my wing on hold. For my job, I still need to develop R packages. In the first half of 2023, I released grafzahl and rang on CRAN. I am also a developer of rtoot, which the maintainer is my team lead David. Even for the released package such as oolong, I am still need to maintain it to make sure it still works.

I am quite pessimistic about whether I can finish Projekt 71 in the second half. Timewise, I can foresee the coming six months would be quite busy. But I am now quite knowledgeable in The R API. Let’s hope that I can at least get something done substantively.

Publish top papers

I published seven papers in the last six months. Yes, sometimes you publish one, sometimes you publish seven.

(No, I will not talk about papers for ICA 2023. To me, it’s still kind of embarrassing to talk about them.)

I believe they are top papers: they are in some of the best journals of the field. I am either the first or second author of these papers. And I don’t have any PhD student and I really the one who did the work. Three of these papers are many years of work. For example, the research of the IJPP paper was conceived immediately after the Christchurch attack in 2019 and it took many round of internal reviews, data collections, data recollections, data analyses, data reanalyses, external reviews, rejections, and revisions. In the end, this paper took almost 4 years of my life. It’s also the last paper from the Germany team of the Responsible Terrorism Coverage I project. I was hired and moved to Germany for this project. The IJPP paper is the perfect ending of the project.

The Bayesian paper was conceived in 2020 at the beginning of the pandemic. It was meant for a special issue. But well… After several rejections by several journals, it’s now published in International Journal of Communication (probably my favorite open access journal). The GitHub history shows that the entire process took 27 months of the my life. I like this paper a lot. But I am not sure whether the industry would take this paper seriously. The things that I think can potentially change the field usually won’t happen. For example, I thought the industry would take validation of topic models and dictionaries seriously. But unfortunately, it doesn’t happen.

The Political Communication paper was the only output of the research project Erfahrungen von Alltagsrassismus und medienvermittelter Rassismus in der (politischen) Öffentlichkeit. Similarly, the research project was conceived during the beginning of the pandemic. The project was a multi-team project. We are the communication science team. There are two other teams (political science and sociology). The PhD student of the political science team is now a professor. But the postdoc of the communication science team is still a postdoc. In the first half of the 2023, I have also cleaned up the code and released the reproducible material.

In comparison, the rest of the papers were developed in a more agile manner. The development of grafzahl was started with the rented GPU laptop last summer. I still fondly remember demoing the first working prototype during my visit to the University of Mainz Computational Communication Research Group last August. It was super difficult to get it published. (I don’t want to talk about it but it even got rejected by the ICA 23 conference) In the end, I am glad that it’s now in its own software paper. Although in the age of ChatGPT, I don’t know how far it still can fly.

I also like the rang paper because it is the first paper I developed at GESIS to contribute to my current Open Science research agenda. I have written about rang many times in this blog. I don’t want to repeat. The same can be said about rtoot.

The DJ paper with my long-time collaborator Jing is more of an opinion piece for the 10th anniversary issue of Digital Journalism.

I would like to thank all of my coauthors: Adrian, David, Hartmut, Jing, Joe, Kasper, Katharina, Marc, Philipp, Rainer, Scott, Wouter.

For the second half of 2023, there are still some papers in the prep line.

Postscript

Usually, I would write a postscript. But I feel that I would write something extremely bitter about how I perceive myself, how other people perceive me, or how I perceive other people perceive me. I will be 42 next month. In September, I will be in this country for six years. I think I should be at ease about my life. But instead, sometimes I feel more and more uneasy about my life. There’s always frustration about my career, in spite of the fact that I have given up my pursuit of professorship. There are always hoops to jump through: you must get x (let’s say a research grant) to have a stable career. Oh, you got x. Congrats! You need to get y now to have a stable career! My wife once said to me that I consider too much how other people see me. I agree. But in an environment where I have worked my arse off but my destiny is still controlled by some other “judges”, you can’t help but to value yourself through the lens of those “judges”. Can I please them? I don’t know. It’s paranoia enough to a point of almost like a persecution complex.

I said I don’t want to talk about ICA 2023. This is the only thing I want to talk about: During the ICA 2023, I strolled the streets of Toronto alone a lot. My tiny cheapo headphones almost exclusively played a 1996 song on repeat. It’s a song by the Hong Kong 90s indie band AMK. I like the band a lot and I’ve written about the band in this blog several time. Even when I was 20 something. During the ICA, I read that the front man of AMK (Kwan King-chung) is no longer with us (I didn’t know he died in February, he was 12 years older than me) and I think a big part of my youth is now gone.

The text of the song reads:

我問佢點解唔笑。佢話：生命唔係唱歌作樂。我問佢點解鐘意黑色。佢話：真實嘅黑暗，比虛偽嘅光明，更令人舒服。 (I ask her why she doesn’t smile. She says: life is not for singing and making fun. I ask her why she likes the color black. She says: the truthful darkness, compared with the hypocritical brightness, makes one more comfortable.)

I always wanted to voice out the truthful darkness inside me. But the reality is, the hypocritical brightness makes people more comfortable. So, I think I better shut the fuck up.

Powered by Jekyll and profdr theme