Hacker News new | past | comments | ask | show | jobs | submit login
Byrne’s Euclid – A reproduction of the celebrated work from 1847 (c82.net)
79 points by akalin on Dec 17, 2018 | hide | past | favorite | 12 comments

Tufte reproduced one diagram and the accompanying text as an example in one of his books. It might have been "Envisioning Information" but I'm not sure.

If I remember correctly he had lots of praise for the visualization and said that appart from the font it had a very contemporary look despite being from the 19th century. Tufte also showed a variant that added colored labels, but I don't remeber if he recommended it. Another thing I remeber is a sidenote about how complicated it was to get the colors right in print.

EDIT: I was wondering if Tufte's book was the inspiration for this but couldn't find a reference at first. The making-of blog post mentions Tufte's work:

> Byrne’s work was largely ignored and criticized at the time of publication but it has gained renewed interest in recent years in part due to a mention from Edward Tufte in Envisioning Information and a reproduction by Taschen.

[..] > Inspiration

> I can’t recall when I first learned of Byrne’s edition but it was likely from Tufte or seeing Taschen in passing.


There is also a version in ConTeXt (arguably a bit higher quality and with some very impressive programmatic typography): https://github.com/jemmybutton/byrne-euclid

The typography is definitely competitive. It's unfortunate that TeX and Metafont are so dated and so austere in terms of what they offer the programmer. It's a big barrier to figuring out what is going on. The author of the c82.net version is obviously a meticulous graphic designer, but his priorities are clearly so far away from the mindset needed to enjoy writing complex TeX that the comparison is a bit difficult to make.

Where the ConTeXt one falls down, aesthetically speaking, is the lettrines (which, by the way, are more usually called initials in English). They show the marks of having been procedurally generated rather than hand-drawn, for example in the way that the line width doesn't vary, and the way the curves appear to be piecewise circular arcs.

It's not possible to make beautiful organic arabesques with a few piecewise circular arcs, it just ends up looking crude and mechanical. Mathematicians are inclined to be insensitive to this problem.

Take the following invented curve, for example. The junctions between different arcs aren't smooth enough. No typographer would accept it, although it goes without saying that typeface design is not a highly complex intellectual activity comparable with math—it's about judging things by eye.


It's a shame that fragmentation of expertise means that the mathematicians and the typeface designers don't generally communicate.

I have an edition of this published by Taschen [1] and spent some time going through it (about 3/4 of the way through).

My main frustration is there are many mistakes. In an effort to keep the text aligned with the original, those mistakes are in-line and the corrections are at the beginning of the book in a prefix. This makes following the proofs a bit difficult since once you get far enough along the mistakes compound. The later proofs are built on earlier proofs. There are some instances where Proof C has a mistake that relies on Proof B that has a mistake, etc. And then flipping back to the corrections and keeping track of all of them is a bit difficult.

It's not a problem if you are just enjoying the book for its aesthetics but since I was very carefully checking my understanding of the proofs it was annoying.

I still recommend the book.

[1] https://www.amazon.com/Byrne-Six-Books-Euclid-Multilingual/d...

Interesting that the book uses the old-fashioned "long s" (that somewhat looks like an "f") despite being from 1847. The "long s" was already falling out of favor by the late 1700s.

The font is impressive. If you search for "ct", some of them have a fancy curlicue ligature. They're true to the original (compare https://www.math.ubc.ca/~cass/euclid/book2/images/bookII-def... with https://www.c82.net/euclid/book2/). And they seem to obtained merely with the css:

  font-feature-settings: "clig", "dlig", "liga", "onum"

The publisher William Pickering was apparently a publisher of high quality. But I have seen Byrne been described as eccentric. Maybe he demanded the use of the long s. It was still used in handwriting regularly that century.

Oddness and lavish mid-Victorian colour printing did seem to go together, yes: https://books.google.ie/books?id=SugpRWf_FB4C&pg=PA35&lpg=PA... https://en.wikipedia.org/wiki/Vestiarium_Scoticum . Like the frilly initial letters it's clearly an archaising choice meant to hark back to early printed Euclids (see the one in https://www.c82.net/euclid/about/ ): part of the nineteenth-century craze for style revivals.

In German text, it was still very fashionable, even into the 1900s. It is also the reason for the "sharp" or double s, ß, which is the result of a ligature between two of the long s letters.

It is still part of the orthographic rules for Frakturschrift, i.e. not using a long s in fraktur would have been an orthographic mistake. Since Fraktur was used into the 1940s, people were very much used to it.

I find it very hard to get used to. Could have been easier.

I used to work out of a lot of 16th century Italian manuscripts. After a couple of hours you don't notice it anymore.

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