Item — HN Reader

This topic is close to my heart. I do research on the fundamentals of 2D graphics, some of it cutting edge (especially GPU techniques), others just selecting the best known techniques (for example, right now I'm doing a bit of a deep dive into robust cubic and quartic polynomial root finders, not really academically publishable, but potentially hugely useful for others working on similar problems).

I made a serious attempt to submit some of my GPU monoid work to academic conferences earlier this year, got rejected. As someone who doesn't have the "publish or perish" incentive of actually being in academia, it's really just not worth it.

I'm also doing inquiry into UI, for example architectural patterns in reactive systems. The current state of academic literature on this topic seems to be terrible (though please point out counterexamples!). Most of what's written is marketing material for UI frameworks - there's an explosion of those, especially in the JavaScript world, but very little synthesis of the core concepts. I wouldn't even attempt to try to submit an academic paper on this topic, though I think it would likely be useful to the world.

It feels like there should be a space to publish work that is not novel in an academic sense, but useful. Wikipedia is not it (their articles on cubic and quartic equations are garbage when it comes to numerical concerns), Stack Overflow is not it, academic conferences and journals are not it. I'm using my blog, generally successfully, but it feels there should be a more systematic approach.

4 comments

I found myself in a somewhat similar situation. I learned to thing long-term:

(1) Publish initial ideas not in top conferences, but in second rate venues. Most publications in top conferences are, clearly, refinements of ideas that were first presented in inchoate form, in less prestigious workshops. Paradoxically, prestigious conferences don't like brand new ideas, but they like polished papers. And they like concrete open problems being solved. If a paper starts "We are solving the longstanding open problem from [17]" then that dramatically increases acceptance probability: most submissions to prestigious conferences are not rejected b/c they are wrong, but because it's unclear why they are more siginficant than all the other submissions. Solving a concrete open problem gives you social proof that the problem is significant (others worked on it) and hard (people failed). How do you create concrete open problems? By leaving them as concrete open problems in a previous paper.

(2) Create your own conferences. All now famous conferences started as small, low prestige workshops 10-20 years ago, e.g. the "Oakland" security workshop (now: IEEE Symposium on Security and Privacy).

I learned this from my post-doc mentor, who is now a famous scientist with a whole research tradition to his name, that he created, bootstrapping from a first workshop paper.

PS, is your monoid work online, is it this https://arxiv.org/abs/2205.11659 ? I'm interested in monoids on GPUs.

> PS, is your monoid work online, is it this https://arxiv.org/abs/2205.11659 ? I'm interested in monoids on GPUs.

Yes, that's the draft on arXiv, and I have a blog post in the pipeline[1] explaining it to a more general audience.

And thanks for the other advice, I'll consider it!

[1] https://github.com/raphlinus/raphlinus.github.io/issues/66

I am a graphics researcher who has been following your blog posts on your GPU vector graphics engine. I've been benefitted by your posts and I think some of the vector graphics renderer designs you discussed in your blog are genuinely novel and worth publishing as academic papers if written and evaluated properly. Please consider submitting them to a graphics venue like SIGGRAPH, EGSR, High Performance Graphics, or Journal of Computer Graphics Techniques if you find free time to do it. It will greatly benefit the community. ; )

By the way: graphics community does recognize contribution on fast and robust polynomial solvers! See the very recent work from Yuksel http://www.cemyuksel.com/research/polynomials/

I wonder if publishing papers onto something like github could become a tradition. Heck, you could even include an implementation. People could make pull requests for new ideas, or even fork your paper...

Academic journals grew out of the processes that academics used to communicate with each other and filter ideas through their community in the pre-internet days. It is weird that they've become the stamp of approval for all research.

The Programming Journal (https://programming-journal.org/) and its associated conference seem pretty good to me. At least I find papers there that are interesting to me and are not unnecessarily obtuse in notation etc.

On the subject of reactive UIs, I'm quite enjoying reading the WebSharper papers at the moment. The most recent is, I think, http://www.simonjf.com/drafts/reactive-abstractions.pdf

Hope that's useful to you.