# TeX family tree with timeline?

When I first started using TeX, I felt overwhelmed by the myriad of names — BibTeX, XeTeX, LaTeX, XeLaTeX, ConTeXt, LuaTeX, MacTeX, TeXLive — and more. I didn't know the packages from the programs from the distributions, or what begat what.

Since then I've learned, but once in a while I still come across names I don't recognize. I often wonder where I might find a “TeX Family Tree” showing how everything relates along a timeline. I've looked and looked, but have never found anything. So, I decided to bite the bullet and roll my own.

I'm looking for feedback:

• Are the relationships correct?
• Is anything major missing?
• What can be improved?

Clicking on the image leads to a PDF. I made two versions: Portrait and Landscape.

The chart spans 52 years — all the way back to 1962, when D.E.K. began writing TAOCP — and includes 17 versions of TeX, 12 versions of METAFONT, 17 derivatives of TeX (CommonTeX, Web2C, MLTeX, eTeX, TeX-XeT, TeX--XeT, eTeX, eLaTeX, encTeX, Omega, Aleph, XeTeX, XeLaTeX, ArabXeTeX, ConTeXt, LuaTeX, LuaLaTeX), 3 formats (Plain TeX, AMS-TeX, LaTeX), 7 tools (WEB, CWEB, dvips, pdftex, pdfetex, pdflatex, BibTeX), 8 distributions (PCTeX, DirectTeX, MiKTeX, proTeXit, teTeX, 4AllTeX, TeXLive, MacTeX), and 36 D.E.K. publications (TAOCP, Computers & Typesetting, Literate Programming, and so on). I learned a lot making this.

I included Knuth's writings because (a) I think they're integral to TeX's history and because (b) I find it fascinating how they fit into the overall timeline, epecially the gaps.

If anyone would like to experiment with modifications, please be my guest. The `graphviz` source is here.

-
Wow, impressive. Shouldn't there be some connection between LaTeX and the version of TeX it is based on? Also, this seems like a community wiki question to me. –  Canageek Jan 28 '12 at 16:49
First of all it was very difficult to follow the tree ;-). Nevertheless I think great work. You listed `bibtex` ==> `biblatex` but `biber` is missing. I think this sould be mentioned. –  Marco Daniel Jan 28 '12 at 18:10
I hope you don't mind, but I made the images smaller because 6600px wide image was really too much for browsers. –  morbusg Jan 29 '12 at 12:09
small addition: 1978, knuth's first published documentation of tex: "Tau Epsilon Chi: a system for technical text"; first appeared as a stanford computer science report, then published by ams as a pocket-sized yellow book with a large green TeX on the cover with a background "TeXture" of smaller tex logos in green. –  barbara beeton Jan 30 '12 at 20:11
many tex-related reports are listed among the [stanford comp sci dept tech reports]{infolab.stanford.edu/TR) for the 1970s and 1980s, but not that one. (the reference number is STAN-CS-78-675.) a google search for -- "tau expilon chi" knuth -- turns up quite a few references, but as far as i can tell there's no readable copy on line, almost certainly because of copyright restrictions. i will investigate the possibility of releasing the copyright so that the original stanford cs report can be posted. –  barbara beeton Jan 31 '12 at 13:59

Arno Trautmann has produced a nice document on the same topic, title "An overview of TeX, its children and their friends..." (available from ctan and github).

On most TeX distribution, it can be viewed directly by typing

``````texdoc tex-overview.pdf
``````

in a terminal.

-
+1 It would add value if you can eloborate with a simple oneline about the title of document and permanent link like this one mirrors.ctan.org/info/tex-overview/tex-overview.pdf or `texdoc tex-overview.pdf` at the commandline terminal. Otherwise links may vanish overtime –  texenthusiast Jan 8 '13 at 5:49
@texenthusiast: Below? In my setup there is nothing below , eehm, now mine comment. ;-) In other words: It would have been better to write something like “in a comment to his own answer” – you could even link it: “comment to his own answer”. –  Speravir Jan 8 '13 at 16:38

Kudos: this is great to see. I particularly enjoyed seeing the laying out of Knuth's work in a timeline. Did he really throw away his Tex and Metafont implementations in SAIL a year after writing them? Yes, I see he did.

There was a whole host of rival formats in the mid 80s to early 90s of which Latex became the winner. These might take up too much space to document in any detail. David notes that the NFSS is not put in the timeline, which is of obvious importance to the development of Latex. Some other of the most important format-independent macro packages might also be worth including.

Context also deserves a tree. There's not a good history of it that I know of, but Hans Hagen documents that it is based on INRS Tex in his TUG Boat interview. I understand that early Context incorporated a lot of macros from TABLE, a big table-setting macro package from the 80s, and I think it might also have incorporated the macros from Pictex.

-
I had a really hard time finding old information about ConTeXt. I didn't put this on the chart but I think it was originally called `pragmatex` prior to its public release in 1996. (This is accidentally showing up in the chart as `ConTeXt0`.) –  Todd Lehman Jan 30 '12 at 16:58
@Todd: Ah, I hadn't spotted that Context was nestling among the implementations - that's not really right, because Context mk. II is a regular Tex format that runs on Xetex and Pdftex, and Context mk. IV is a mixture of Tex and Lua scripts that runs on Luatex. I'd group it in yellow with Plain Tex and Latex. Have you looked at tex.ac.uk/tex-archive/info/tex-overview/tex-overview.pdf? –  Charles Stewart Jan 31 '12 at 14:13
Likewise, I think you can drop some nodes: lualatex, pdflatex, and xelatex are just slightly tweaked formats of Latex for the respective engines. –  Charles Stewart Jan 31 '12 at 14:19
@Todd: I searched CTAN for mentions of pragmatex. You have quite a few things in your chart not in those charts. I guess I should put that document in my answer. –  Charles Stewart Feb 1 '12 at 9:40
Taco has been maintaining an archive of ConText releases at Suppelec –  Ulrik Vieth Mar 9 '12 at 13:26

I think you need a key, it wasn't always clear what the colours stand for, for example e-LaTeX/pdfe-LaTeX, Aleph appear to be marked as implementations like pdfTeX/Omega, but they are (more or less) just formats built over the relevant engines. Similarly I'd have expected to see context colored as a format like latex rather than an implementation (although it's always come with additional programs and scripts in addition to the TEXformat at its heart).

The most influential format precursor of LaTeX2e is omitted, namely the NFSS-LaTeX formats distributed in parallel to LaTeX2.09 in its later years.

Several influential implementations are omitted, early implementations on PC included ArborTeX and SBTeX (and most influential around the time of the LaTeX2e design, emTeX) The commercial Y&Y TeX and VTeX formats were notable for early support of scalable font formats.

Notable implementations on mac should include TeXtures, I think it was the only Mac implementation for a long time. The implementation timeline ought to start from the beginning, not from PCTeX, VMSTeX various mainframe TeXs, etc.

Sorry no dates in this comment, just some initial observations..

-