Personally I hate Tex/LaTeX and I think it should be replacted by something else long time ago. First of all syntax is horrible, if you don't work with it on daily basis try to figure out what macro you wrote one year ago is doing. Each time you have to jump to manual and learn almost everythig from very beginning. Next thing is lack of utf-8 and TrueType fonts supports (I know there is XeTeX and LuaTeX) but today such features such be in very core of text system no in some software branch. Extensibility is next thing and here again it is very poor unless you use something modern like LuaTeX or you are an expert in TeX macros. Packages dependency hell is next thing side by side with stupid compilation process with meaningless error messages. Am I the only one who think this way?
The problem with systems like Adobe Indesign is that they are “What You See Is All You Get”. For a company brochure with special typography and particular Pantone ink colors used in offset printing there may be nothing better. However, I’ve owned a license to Indesign for over a decade. I’ve done tutorials, bought half a dozen books on it and maybe produced one document using it.
In Indesign I wasted so much time figuring out how to get the layout, the formatting, the figures, code samples,
mathematics, bibliography, even page numbering, index and footers I wanted that I had to give up.
The power of bibtex, TikZ, and other packages in the TeX ecosystem make it possible to use a system that produces just the sort of documents that I want—and the software is infinitely less expensive (it doesn’t cost anything).
TeX isn’t without its difficulties. I’ve been programming for over 50 years and still dread diving into the macro language based code for complex packages—but at least it’s available, something you can’t say about these other proprietary systems.
TeX was written by a Computer Scientist (perhaps the most eminent Computer Scientist) and it shows. Its real power is revealed only to those able to program. This is a shame, because using it has made me appreciate good typesetting—something I’ve found difficult
to achieve in the kind of papers I write when I use other tools.
I use MS Word when my recipient needs it that way, I use Apple Pages or Google Docs for simple documents when I don’t care what they look like, and Markdown or Org mode for my own notes. For anything important or for something I want to have as archival source (TeX is a purposely frozen format) I use TeX and its related tools, LaTeX etc.
Your comment reminds me of the people who shit on C++. Some of the criticism may be true, but there simply isn't anything out there that could possibly be considered a feasible replacement.
Also:
>lack of utf-8 and TrueType fonts supports
There's UTF-8 support in the standard pdfLaTeX compiler and TrueType support in XeTeX (which is far far from a trivial issue, which is why there are two branches, if you don't need TT fonts you're better off with pdfLaTeX which has more support for e.g. microtypographical adjustments).
I don't think that's true. Both Latex and C++ were replaced. It's just that their best replacements were more specialized tools so that left a bit of space for both to survive.
It's just like that example of cars mostly replacing trucks in the US. Trucks aren't obsolete, but most don't need them so choose alternatives that fit their life better.
Latex was replaced by Microsoft Word, Adobe InDesign, and Markdown. Each one is different but for people who might have used Latex for a certain use case, one of those more specialized tools might be better. But there will always be a need for Latex itself.
That just proved my point. Saying markdown replaces LaTeX is like saying javascript replaces C++. They're not even in the same league, they're tools for completely different purposes.
One could argue that Adobe InDesign, Quark XPress et al are the replacement. They have the typesetting and layout capabilities that common software lacks.
I feel like there isn't even a straightforward way to learn the principles of the syntax of LaTeX, in full. I've been an occasional user for years, I love the results, and I can do basic math typesetting off the top of my head. Anything more involved (diagrams especially), I really do need to start from scratch with some new package to get things working. Fortunately, stackoverflow tends to be a one-stop solution for me most of the time.
I don't believe I'll ever have a deep enough grasp to be able to do those more complex tasks on my own, and that really bothers me. I can imagine some syntax improvements, but like a sibling comment said, I don't see the mindshare shift happening any time soon.
Next thing is lack of utf-8 and TrueType fonts supports (I know there is XeTeX and LuaTeX)
UTF-8 is supported by all modern engines. It's just that it wasn't the default for pdfTeX so you had to add one line to enable it. But this has changed in 2018 [1], so UTF-8 is now th default LaTeX encoding even when using good old pdfTeX.
I consider XeTeX and LuaTeX as branch from main software line LaTeX. Ech of them have some extra features I would like to see merged into one software. To use unicode I have to switch to XeTeX but it is not such simple transition, to have bibliography with unicode I need to use some other tool (like Biber) because Bibtex it is not unicode aware. And there is a lot of other exceptions/hacks like this where such basic scenarious should just work out of the box in XXI century but they are not.
> XeTeX and LuaTeX as branch from main software line LaTeX
No, that is mistaken. As the FAQ says, "Development of pdfTeX has mostly stopped" https://texfaq.org/FAQ-enginedev, in favor of LuaTeX (and XeTeX maintanence). But users use what they use and the people who put together the distribution do not normally shut things down, so pdftex remains. (I admit I often use it.)
I'm not much of a biblio person myself but I find that biblatex and biber work for me. I think I remember someone saying that they could use more developers, but in any event I'm sure that someone who volunteered to pitch in would be welcome.
If a person would like to see some more development but does not have the energy or the skills then the best thing to do is to join a Users Group, such as TUG at https://www.tug.org/. Or you could simply make a contribution so that some developer could get funding.
We are teaching this combination since 7 years now to physics students and I have been using it for nearly all documents I created. That includes bachelor and master thesis, dozens of beamer presentations and currently wip PhD thesis.
Is pdfTex clearly labeled "do not use" and development has stopped completely? Also, do lualatex/biber/biblatex all have feature parity and are clearly documented as the "true" latex projects that are the only ones which should be used currently?
If there weren't two "yes"'s in response then OP's clear criticism will remain a problem. It cannot be escaped by advising the user to jump to a different branch based off the last downside they happened to describe about their chosen branch. (E.g., "I want unicode" -> "use lualatex", "I want WYSIWYG" -> "use SwiftLatex", "But what about unicode" -> "use lualatex", ad infinitum)
TeX/LaTeX is a rich ecosystem that has grown over a long time.
Over the last couple of years, pdftex, xetex and luatex have coexisted and pdftex was the stable choice and the other two were functional but in beta stage.
In the last years, xe and luatex became stable and pdftex entered maintenance mode.
WYSIWYG is an antithesis to LaTeX, but I'm sure it's possible to base something like SwiftLaTeX on luatex instead of pdftex.
This has changed only in the last few years.
You listed a few requirements, and while many are achievable using pdftex, only lualatex tikz all the boxes.
* unicode out of the box
* otf font support
* unicode in bib via biber / biblatex
* easy extensibility by writing lua instead of plain tex
there is more.
But only because luatex is out of beta, does not mean pdftex suddenly stops working or needs to be deprecated.
OK, maybe he means that, but XeTeX and LuaTeX do have them at the core. Users use what they use, but the developers have for a decade worked to provide utf8-native and system font-enabled engines.
