Hacker Newsnew | past | comments | ask | show | jobs | submit | rkp8000's commentslogin


I love Markdown. I'm a bit surprised, though, that you still can't open a .md file by default in most web browsers. It seems like it should be quite trivial to have the browser automatically convert it to html and display it.


As I was griping above, you usually can't just view a Markdown file with formatting applied at all. I think MAYBE Notepad has been updated just recently to render it, but otherwise... you're looking at plain text with a bunch of formatting characters in it. Why? It's baffling. Where are the simple Markdown READERS?


Your plain text editor is the markdown reader. That’s the point.


It really would be nice to have a convenient renderer for it though. It's genuinely surprising something like firefox doesn't have a markdown reader builtin already.


I wrote one for my own use (on Linux): https://github.com/mvz/mdv


If you’re on a Mac, I recommend qlmarkdown. It’s a customizable quick look renderer for markdown (md) files.

https://github.com/sbarex/QLMarkdown


Then that's no point at all. You're not seeing the formatting that the embedded formatting characters call for, so why have them in there?


and it looks like poo, that’s the other point


Exactly. You can always count on someone to come back with, "That's why it's so great! You just read it in a plain-text editor!"

Why does anyone have to point out how dumb it is to litter your "plain text" file with formatting characters that are never respected?


That would require Markdown to be standardized. (There is the CommonMark standard, but it’s extremely complex and still ambiguous.)


You could still support a subset of the most common features like bold, italic, strike, bullets, links, etc.

Isn’t the beauty of MD supposed to be that if you can’t render it it should still look fine as plaintext?


Even these basics are not consistent. See my Markdown Monster:

https://git.sr.ht/~xigoi/markdown-monster/blob/master/monste...


There should be only one correct interpretation of that according to CommonMark. Software is faulty for sure and a lot of the these "markdown converters" are pre-AI slop code but at least there is a carefully written spec now.

(That dude who coined the name Markdown is being a dick about other people finishing his abandoned idea is another issue and not the fault of CommonMark.)


The problem for web browsers is that markdown is technically a superset of HTML.


Why is this a problem? To me it sounds like a it would be an advantage because you have everything you need to render it already built into the software.


Rendering is trivial. The issue is standards, and the DOM. No-one can write a Markdown implementation for the core of any major web browser in a form that is simultaneously acceptable to both their technical and political governance.

Best you’ll get is a plugin. Strictly arm’s reach. Translation only.


I'm not quite sure I understand what you are saying. Is the essence of what you are saying that it is hard to agree on a spec for the Markdown (and how it is translated to HTML or directly to DOM?) Or that this represents a technical challenge I don't understand?

I think you mean that any markdown byte sequence will also parse with some results using an HTML5 parser?

Content-Type should fix that.

Easy enough to associate *.md with one on most static servers too.


In what ways is it a superset? What can you express in markdown that can't be expressed in HTML?


Why is it a problem for web browsers?


I don't think so. I think it would be sufficient to document the exact markdown it supports and let the chips fall where they may. Yes, it would push markdown in a certain direction, but that's OK as long as it stays faithful to some variant most people already know. For instance whatever variant Github or some other major, main stream tool uses. And then just ignore the critics.

It'll certainly make some people angry, but if we try to please everyone we can't get anything done. And I suspect that it is the fear of not being able to please everyone that is the reason browsers do not have markdown support. It takes a bit of courage to say "this is the variant we'll implement".

HTML was originally said to be an application of SGML. It wasn't. It was vaguely inspired by its syntax and that is the only reason HTML saw wide adoption. Had they tried to actually implement anything close to ISO 8879:1986 we would NOT have adopted HTML for the web. Mostly because it would have been too costly to implement. (Anyone doubting that: have a look at the ISO standard. You can get it in what is essentially annotated form in Charles Goldfarb's "The HTML Handbook").

Of course, Markdown is nowhere near as complex. So this is where perfect is the enemy of good and we end up getting nothing.


I've taken to using Markdeep [1] for this.

You write your markdown file, but add the code snippet at the bottom of yor document and save it with a .md.html extension. Then when you double-click it it opens and renders in your browser.

I save my notes in a Google Drive, and it's now replaced all the note taking apps I've tried over the years

[1] https://casual-effects.com/markdeep/


This is a good call. I know it's been suggested multiple times over the years; I wonder what the rationale was for rejecting the format, or at least having the option to render a file when it's loaded. (Maybe a "display as HTML" button or the like would be required before it would be rendered.)


“Markdown” is a family of writing formats. There is no one “Markdown”. It’s completely unsuitable for direct inclusion in the web platform.

Related reading: https://www.rfc-editor.org/rfc/rfc7763.html text/markdown registration.


The overlap between these Markdown formats is actually larger than with many other formats. Possibly even larger than HTML’s overlap back when MS Explorer was the dominant browser.


> Possibly even larger than HTML’s overlap back when MS Explorer was the dominant browser.

No way. You were never left in doubt about whether a normal HTML tag would work, or whether tables were available or would become a jumbled illegible mess, or whether a line break in the source would become a space or a hard break. And that’s just the first three things that occur to me.


You have to be willingly ignoring CommonMark, these days.

I understand it doesn't have all the extensions one might hope, but to not parse the basics like the examples in the spec say is just doing everyone a disservice.


Yeah, also missing a built-in JS API for turning Markdown into safe HTML. Sure, there are lots of different implementations, but maybe start with something small at least.


Computing entropies of high-dimensional random vectors for a theoretical neuroscience study. The journey is mostly a repetition of (1) almost giving up because it's completely hopeless, (2) taking a hot shower, (3) realizing there might actually be a path forward, (4) almost giving up because it's completely hopeless.


A notable and interesting point of this article is that convolutions and correlations (convolutions without flipping the filter) are quite a bit more subtle on the sphere than on Cartesian spaces. For a convolution between a function and a filter on R^N you just "slide" the filter around, integrating at each shift, which produces another function on R^N. On a sphere, however, there is not a clear cut way to slide a filter around a sphere. For instance, there are multiple ways to slide a filter centered at the north pole to the south pole, which will result in different filter orientations.

More generally, the space of rotations, which is the argument of the convolution (analogous to the shift amount being the argument of a standard convolution), is 3D (3 Euler angles), whereas the space of points on the sphere is 2D (polar and azimuthal angles). Thus, whereas convolution over R^N returns a function over R^N, convolution over the sphere actually returns a function over the 3D rotation group SO(3). This has interesting consequences for e.g. the convolution theorem on the sphere, which is not as clear cut as simply rewriting the standard convolution theorem in spherical terms.


A (non-exhaustive) list of some notable papers:

Maass 2002, Real-time computing without stable states: https://pubmed.ncbi.nlm.nih.gov/12433288/

Sussillo & Abbott 2009, Generating Coherent Patterns of Activity from Chaotic Neural Networks https://pmc.ncbi.nlm.nih.gov/articles/PMC2756108/

Abbott et al 2016, Building functional networks of spiking model neurons https://pubmed.ncbi.nlm.nih.gov/26906501/

Zenke & Ganguli 2018, SuperSpike: Supervised Learning in Multilayer Spiking Neural Networks https://ganguli-gang.stanford.edu/pdf/17.superspike.pdf

Bellec et al 2020, A solution to the learning dilemma for recurrent networks of spiking neurons https://www.nature.com/articles/s41467-020-17236-y

Payeur et al 2021, Burst-dependent synaptic plasticity can coordinate learning in hierarchical circuits https://www.nature.com/articles/s41593-021-00857-x

Cimesa et al 2023, Geometry of population activity in spiking networks with low-rank structure https://journals.plos.org/ploscompbiol/article?id=10.1371/jo...

Ororbia 2024, Contrastive signal–dependent plasticity: Self-supervised learning in spiking neural circuits https://www.science.org/doi/10.1126/sciadv.adn6076 Kudithipudi et al 2025, Neuromorphic computing at scale (review) https://www.nature.com/articles/s41586-024-08253-8


A great pedagogical article on thermodynamic vs logical reversibility, for those interested: https://arxiv.org/abs/1311.1886 (Sagawa 2014).


Sagawa was mistaken in this article; he failed to appreciate the role of mutual information in computing, which is the proper basis for understanding Landauer's principle. I discussed this in https://www.mdpi.com/1099-4300/23/6/701.


Thanks for the link to your article!

If you don't mind my asking, how much does the role of mutual information in linking logical and thermodynamic reversibility depend on considering quantum systems? I.e. does your footnote 37, which discusses independent systems vs "subsystems of correlated systems" hold for classical systems as well?


There are a few folks working on this in neuroscience, e.g. training transformers to "decode" neural activity (https://arxiv.org/abs/2310.16046). It's still pretty new and a bit unclear what the most promising path forward is, but will be interesting to see where things go. One challenge that gets brought up a lot is that neuroscience data is often high-dimensional and with limited samples (since it's traditionally been quite expensive to record neurons for extended periods), which is a fairly different regime from the very large data sets typically used to train LLMs, etc.


This is just to say: the poem at the end is a play on William Carlos William's poem, entitled "This is just to say".


This reminds me of a project I worked on during my PhD, where you create a network of scientific documents and notes/threads via Markdown, with a similarly structured "rabbit-hole" linking system: https://github.com/rkp8000/hypothesize . I'm not really a software engineer so never made it ready for general use, but I'm very happy to see a similar idea turned into something real! Kudos to the authors.


MSR has a very clear and accessible tutorial on quantum computing for anyone interested in getting up to speed with the fundamentals: https://www.youtube.com/watch?v=F_Riqjdh2oM .


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: