Live data from Hacker News

The most expensive typefaces (2019)

medium.com

41–47 of 47 posts

Re: The most expensive typefaces (2019)

#41
post #26
post #13

Earlier quoted context omitted.

The point is simply that if I wanted to create a website and use Circular [0] (a nice geometric font made popular by Spotify) I would need to pay based on total pageviews which is an unknown, so the price is unknown, which is prohibitive in the sense that I'm now tied to a font license that penalizes me based on the success of my website. Or, I could use Inter which is similar, free, better (more weights, filetype op…

If you don't have an idea of how many users to expect, your server costs are an unknown too. I don't see how paying per view will change that.

A server requires energy and maintenance forever. A font does not.

Re: The most expensive typefaces (2019)

#42

Artbreeder ( https://artbreeder.com/ ) is likely to have a font generation mode eventually. This can generate high quality fonts (as well as lots of Unicode codepoints) and even interpolate between fonts in latent space. You can also take an existing font and find it in latent space, and then regenerate it. Take a look at the header of the Artbreeder homepage to see a small sample of font interpolation. (Apparently t…

> I would not want to be a manual font designer in the future.

I wouldn't want to be any designer in the future. Eventually there will be a tool to generate any design with easy sliders and stuff. You can already design logos in software that are as good as the best designers 20 years ago for free with a few YouTube videos, imagine what this type of generation tech will do in the future.

I think that along with the current easy to use API 3rd-party systems, like Stripe or Uberauth, we're basically going to go back to a ColdFusion style design tools for the "low hanging fruit" sites like regional grocery stores and below where there are generic templates for a whole e-commerce site and back end for your business with the front end included, but this time you can use all kinds of knobs, levers, and buttons to adjust it to look unique and personal for your business and the backend generation will happen based on your product data requirements and the like.

We already kinda have that for mom + pop stores and smaller businesses. The only companies that will need handcrafted web apps will be the out-of-the-ordinary businesses like Uber, or the huge companies that need more complex logic, like FedEx or Walmart. Everyone else will either do it themselves with whatever the Wix of this type of thing is or they will hire some farm of developers (like an Amazon warehouse, but with developers instead of pickers) to task one of their drones to do it.

On different kind of note, that page header font warping thing is super fun to look at while intoxicated!

Re: The most expensive typefaces (2019)

#44

Why is `n` missing from the screenshots?

Cut off one of the arches of the m, and you'll probably have a good idea what the n looks like.

Conspiracy Theory: Maybe the 'n' is actually extremely ugly, and they're concealing it, in order to sell the font anyway. Is it non-refundable?

Re: The most expensive typefaces (2019)

#46

Earlier quoted context omitted.

This is a good start (and baskervald is even better) but is not of excellent quality. I am not talking about the character shapes themselves, that are OK, but about the relationship with greek characters, with LaTeX math, and the quality of accented characters. What the world needs is an available version of the "SMF Baskerville" described by Yannis Haralambous here: https://hal.archives-ouvertes.fr/hal-02101627

I wonder if people would chip in to pay someone to do it right.

This is exactly what happened. When the french mathematical society ordered this font they probably paid a lot of money, and then it became a sort of "secret sauce" that they want to use exclusively.

Re: The most expensive typefaces (2019)

#47
post #32
post #29

Earlier quoted context omitted.

Unless gp doesn't use any sort of auto scaling what so ever. In which case, the cost will be that of the (known) provisioned amount

The equivalent to that would be to just buy e.g. the 100k users font license, and then stop serving traffic once you hit 100k.

Except that for most digital services / contents, a server is a core requirement, while any specific font isn't.

I would not going to risk something as secondary to my business as font become a runaway cost, or stop providing my service / content because of something tangentially related to what I offer. And it seems gp feels the same.

Post reply on HN