interesting project; I’m not really sure how useful it is for field-specific stuff—I'm searching for “image reduction astronomy”, and it shows all sorts of related but not image-reduction work (including noise reduction which is not the same thing). I’m not really familiar with vector search enough to evaluate it well enough.
However I can give you the heads-up that the abstracts don't render well because (La)TeX is interpreted as markdown so that
Paper~1 shows something and Paper~2 shows something else
will strikethrough the text between the tildes (whereas they are meant to be non-breaking spaces). Similarly for the backtick which makes text monospaced in the rendered output but is simply supposed to be the opening quote.
Yes, I think vector search is tricky to navigate at times since now the onus is on the user to explain the problem well. However, you can copy paste full abstracts to get similar papers well enough.
I will fix the LaTeX rendering ASAP.
Thank you for trying out the site! Happy Holidays :D
I disagree. The spacing of the stylized TeX/LaTeX logo is very sensitive to the font used (especially the LA pair); most LaTeX users already don't adjust it when they use a font other than Computer Modern. Even when they can see the A halfway on top of the L, they don't change the spacing.
But with PDF you at least have the knowledge that the font will be the same no matter who sees it. Webpages have no such guarantees (and indeed some people disable web fonts completely for whatever reasons) and it will only end up looking ugly as hell.
On the web, the typographer's instructions are more of a suggestion, and we should lean into a more declarative style of layout, instead of trying to hack up something resembling precise position that will end up looking completely out of place anyways. I say this remembering in horror the many websites where people wanted the staggered look, and ended up producing a monstrocity.
> I disagree. The spacing of the stylized TeX/LaTeX logo is very sensitive to the font used (especially the LA pair); most LaTeX users already don't adjust it when they use a font other than Computer Modern. Even when they can see the A halfway on top of the L, they don't change the spacing.
I'm not sure with which part you disagree. If the sentence:
> And, if you're going to refer to the typographic capabilities of TeX (or to LaTeX), it's a shame not to try to reproduce the distinctive styling of the word that Knuth says proudly in the TeXbook shows off precisely those typographic capabilities.
… then, you're right, I overstated it by saying that you should try to reproduce the logo when just referring to the typographic capabilities of TeX. But I still think that, if you wish to write a post about how to make web pages look more like compiled (La)TeX documents, then it is a shame not even to try to reproduce the distinctive typographic marker in the name of the language itself. Of course there's space for disagreement even with that more mild claim—but can we at least agree that the solution implemented, that visually sets off the word "LaTeX" but in no way that resembles how it ideally should be styled, is worse than just leaving the word alone?
pandoc is seriously under-powered for the kinds of things that LaTeX and Typst can do. Much of the information in Typst/LaTeX source code would simply be ignored during the conversion. It is fine for simple documents, but cannot handle a lot of stuff.
Pandora offers a superset of LaTeX and any other markup languages as long as you’re not using it for final rendering. I have a Markdown+LaTeX project I’m working on and try to write mostly in markdown, but can drop down into LaTeX in-line if need be. I compile a tex file with Pandoc and render with pdflatex, bibtex, makeindex, etc., etc.
This is not the workflow being discussed. Yes, you can use Pandoc Markdown (in which you can embed LaTeX) and then Pandoc can parse the markdown and produce a LaTeX file. But the actual typesetting is still done by LaTeX; here Pandoc is basically just a preprocssing step.
what was being discussed is using Pandoc to convert from Typst to LaTeX, which it can't really do because its internal document representations are not as expressive as either Typst or LaTeX.
pandoc is much less powerful here since it can only parse a subset of LaTeX, as opposed to LaTeXML or lwarp which actually processes the LaTeX/TeX code expanding the macros etc, and then generates the output HTML/XML.
pandoc is nice for simple documents, but it is completely inadequate for many use-cases, which is why ArXiv is using LaTeXML and tikz.dev is using lwarp, rather than the more ubiquitous pandoc.
I am unable to reply to any comments, so I’ll add my reply to blagie as a top-level comment instead: Gilles himself documents his process of using Inkscape here: https://castel.dev/post/lecture-notes-2/
As a summary, he uses a custom keyboard shortcut manager¹ which allows him to compos multiple keystrokes (and also saves commonly used styles):
> For example, when I press `s` and `f` simultaneously, my shortcut manager will apply a solid stroke and a grey fill to the current selection. When I want the stroke to be thick, I press `s+f+g` together, where `g` stands for thick (as the `t` key is hard to reach).
well yes, but it likely will be a “telephone number” and not even a simple one like say `2MASS J19593766+2246141` which is 2MASS (the survey name) and the coordinates of the star in the J2000 epoch: 19:59:37.66 +22:46:14.1 .
The Gaia ones (which is the billion star catalogue) look like `Gaia DR3 1827256624493300096`; the number basically contains the rough coordinates of the star as a HEALPix index (along with some other data) so it's not really human readable, but is perhaps more suited for a survey with billions of sources.
Ironically, the Gaia catalogue is incomplete at the brighter end (very bright stars like Betelgeuse for example are not in the catalogue at all) so still needs to be supplemented by other catalogues (and can't be used as a single source of truth to which all other catalogues are cross-matched for example)
there are still people using it---a few examples of being used in Malayalam: https://typedrawers.com/discussion/4912/metafont-in-2023-nup... (the paper linked there has more information); the comments have work by another group which was presented at the TeX Users Group conference a few months ago.
- “weak gravitational lensing” with the visible light instrument (which has higher resolution than the IR one) to measure very precisely the shapes of galaxies, to enable a statistical study of distortions in their shapes, caused by weak lensing due to dark matter (and regular matter which they can observe “directly”)
- “galaxy clustering” with the near IR instrument to calculate distances to the galaxies (via their redshift) which they can use to map out 3D distribution of galaxies and compare to simulations for example (there is a nice figure on this page: https://www.euclid-ec.org/euclid-core-science showing a few surveys and simulations)
The images here are simply first light images (i doubt the horsehead nebula or globular clusters are part of the core science of Euclid); more images and spectra will be taken in the coming years to do the actual core science (which will require a lot more data)
However I can give you the heads-up that the abstracts don't render well because (La)TeX is interpreted as markdown so that
will strikethrough the text between the tildes (whereas they are meant to be non-breaking spaces). Similarly for the backtick which makes text monospaced in the rendered output but is simply supposed to be the opening quote.