Hacker News Reader

Comment by nafizh | Hacker News Reader

It's funny to note that the first author signed a boycott [0] for publishing research in Nature's new journal (Nature Machine Intelligence) covering AI topics but is one of the first to publish there.

[0] https://openaccess.engineering.oregonstate.edu/signatures

View on HN

Step 1: Sign the Nature boycott.

Step 2: Get boycott popular. Less people submitted to Nature.

Step 3: Submit to Nature. Accepted due to less competition.

Step 4: ???

Step 5: Profit!!

Step 4: Increased h-index

Step 5: You get tenure and more grants

This is a News and Views piece, published in the journal Nature, about a paper in Nat. Mach. Int. The distinction is important.

We are talking about the paper this news piece is talking about.

That is an important distinction, but in this case the problem still applies.

The boycott doesn't apply to Nature, and arguably doesn't apply to non-publication writing like News and Views. Rather the boycott is of Nature Machine Intelligence, and the first author of the underlying paper in NMI is a signatory on the boycott list.

Similar: In the recent baby gene editing controversy around He Jiankui it became apparent that he just had co-authored a paper proposing a set of ethical rules around human gene editing, which he promptly broke himself.

https://www.wired.com/story/he-jiankui-crispr-babies-bucked-...

Is this some next-level of publishing scandal? First there was a reproducability problem, now there's a fake-and-jab? What's next?

Yeah, even funnier though is anyone can submit signatures to this with no verification. I just signed it for you (search for your username). Who knows which of these people actually 'signed' or not. Computer scientists!

The article is published in Nature, not in Nature Machine Intelligence.

The names may sound similar, but there’s a big, big difference.

It's published in Nature Machine Intelligence:

https://www.nature.com/articles/s42256-018-0002-3

But at least this article seems to be available as Open Access. I am not sure if this is temporary or not.

https://www.nature.com/articles/s42256-018-0002-3

Published in Nature Machine intelligence. Publicized through the regular Nature site.

Do as I say, not as I do.

Most researchers are not independent and don't have much choice where they publish. Also, prestige plays a huge role in academia. Getting in Nature is a big deal to them. That's one reason I really dislike academia. They are often playing games that are very far from genuine progress. It's all ladder climbing and groveling at their masters feet. Someone please disrupt academia.

You perfectly described all the non academic jobs I've ever had.

I have worked both in academia and in the industry. Academia is heaven when compared to some of the serpentine performance evaluations in the industry. I have even seen people in the same team undermining each other's work because the top 2 in the group would earn a bonus.

The real merit of academia is that people can be alone in a corner doing one's work, if they choose to. Industry enforces socialisation, conformity to norms, and shmoozing.

Also, the track record of the academic world vis-a-vis disruption is impressive. The oldest western universities are around 900 years old.

That said, the article seems to be a fluff piece about a good but not groundbreaking paper. (Nature MI is not Nature, not by a long shot. It does not have any prestige, yet.)

> The real merit of academia is that people can be alone in a corner doing one's work, if they choose to. Industry enforces socialisation, conformity to norms, and shmoozing.

That's a bit dubious. If you don't network in the academia I doubt you can get far at all, while you can very well just comfortably do your job as a technical employee even if you are not really into networking (if you don't have further ambitions). Networking is absolutely essential to would-be professors or anybody who wants long-term stable employment in the academia, and the competition is much fiercer.

Reply on news.ycombinator.com