Comment by bonoboTP

Comment by bonoboTP 3 days ago

5 replies

Prolific professors who publish dozens of papers at scientific conferences year after year also have similar methods. There's a pattern, there's a method. There's always some novelty of course but it's a production line nevertheless. (I'm talking about real successful academics with real status in the community, not fakes).

It also reminds me of the MrBeast document from the other day.

Once you hit the formula, you keep grinding it. Consistent high performers almost all do it, no matter the field. Whether it's songs, books, YouTube videos, science papers, blog posts, paintings, consultants, etc.

Of course science papers are a team effort but so are novels and music and paintings. Hans Zimmer didn't single handedly compose his recent film scores, he has lots of people working for him on the project. Master painters used assistants. Stephen King isn't typing every character himself.

Amd then there are one hit wonders who are at the right place at the right time, saying what needed to be said in that moment, and then sink back into obscurity or become a parody/tribute of their own self as they try to milk that one insight from their 15 minutes of fame.

Outlier level work by one person that keeps being novel in truly surprising ways, on a consitent basis, is so rare that there are probably only a handful of instances.

ghaff 3 days ago

As someone who gave hundreds of talks and wrote at least hundreds of articles etc. over the years, I basically agree. Some I look back on as being particular insightful or clever. But you pretty much have to mostly crank at least to a minimum standard. And, yes, there's a lot of reuse in various ways.

  • bonoboTP 3 days ago

    Yes, I'm not trying to disparage it too much, it's just the reality of the job. But I found that even junior PhD students can have a romantic illusion that real scientists just spend their time musing and thinking and sometimes they have a flash of insight which they then enthusiastically share with the others. But the more mundane reality is that papers are projects with their own life cycle. You have to be on time, you have to cater to where "the discourse" is right now, the metagame. But to be fast, you need a process, a formula. It's sales. Everything can be approached from a sales and marketing mentality and it tends to bring success to a scary degree. Some apply similar tools in their romantic lives too, where optimizing your dating profile is just the start. I tend to think this is not the real path to something fulfilling, but pursuing this question veers into religious territory.

    • ghaff 3 days ago

      I've even had this discussion with people in at least adjacent roles to mine. We could put together ten very serviceable conference submittals tomorrow that we could spend a week turning into very competent conference presentations within a week. They wouldn't be brilliant but they'd be "good" relative to the norm. It's just one of the things we can crank out like we crank out blogs/columns.

      • bonoboTP 3 days ago

        Depends of course on the field and the conference tier. I doubt you could write an accepted paper for ICLR/ICML/NeurIPS/CVPR/ICCV in a week from scratch. These typically involve writing code, doing experiments, evaluating on benchmarks and obtaining good, if not state of the art, scores.

        Typically this does involve 4-6 months of work even in high-output labs. It just involves many iterations, polishing the text and figures to "look like" an accepted paper by gestalt etc.

        I wasn't trying to imply there's no work or effort needed. I'm more saying that there's a lot of learnable, repetitive parts and if you get them right, your output can have a consistent good quality but nothing groundbreaking. Chipping away at problems, permuting the assumptions and tasks, throwing in the latest hot new method and combining it with another method etc. It isn't something that anyone could do, but it doesn't rely on big eureka moments or flashes of deep insight that one might romantically expect.

        • ghaff 3 days ago

          Certainly, if it needs to be some original/novel research that you've personally done of some sort, you can't just conjure that from scratch. A lot of what I've done over the years isn't that though.