On Mon, Oct 23, 2017 at 12:40 PM, N. Raghavendra wrote:

I am a mathematician, and have been using LaTeX on a daily basis since 1992, when I was a graduate student. I have recently started using ConTeXt, and am very happy with the facilities it provides, and with its self-contained system which does not require one to load external packages with subtle interactions.

One problem that I am facing is that the documentation of ConTeXt is often not enough for me to figure out what to do when a problem arises. I understand that documentation is perhaps not a priority for the project. As Hans Hagen wrote in `? Context',

"... writing can get a lower priority in a time when quick and dirty answers can be found on the internet, mailing list or wiki. ... For what it's worth: whenever I have to solve a problem with a program (or language implementation) I run into cases where I have to look long to find (non conflicting) information. It just comes with the problems one wants to solve and TeX (ConTeXt) is not different."

However, there is not a great deal of information on the Internet about solving ConTeXt problems. Whenever I ran into a LaTeX problem in the last several years, I used to find a solution, or a substantial step towards one, that was already documented on the TeX-LaTeX Stack Exchange. Unfortunately, it does not contain a similar amount of information about ConTeXt. In any case, I have started posting ConTeXt questions there, see https://tex.stackexchange.com/q/397463/146025 and https://tex.stackexchange.com/q/397607/146025

These are the resources that I have found very useful: 0)interface files tex/texmf-context/tex/context/interface/mkiv 1) test suite cont-tst.7z http://www.pragma-ade.nl/download-1.htm 2) sources of the pdf of the standalone, tex/texmf-context/doc/context/sources/general 3) sources of modules tex/texmf-context/tex/context/modules 4) source code of mkiv tex/texmf-context/tex/context/base/mkiv 5) the archives of emails https://mailman.ntg.nl/pipermail/ntg-context/ The ml is ok of course, but at some point you will develop your own style of doing things, and probably you will encapsulate it into a module. -- luigi

On 10/23/2017 12:40 PM, N. Raghavendra wrote:

I am a mathematician, and have been using LaTeX on a daily basis since 1992, when I was a graduate student. I have recently started using ConTeXt, and am very happy with the facilities it provides, and with its self-contained system which does not require one to load external packages with subtle interactions.

Hi Raghu, my experience with LaTeX was brief when compared with yours: about seven years. I think I may be using ConTeXt for almost a decade. My background is in humanities. I cannot code, although I may read and understand very basic scripts. I simply don’t do math (it’s all Greek to me). The most useful ConTeXt feature for me is the handling of XML files as source files for text.

One problem that I am facing is that the documentation of ConTeXt is often not enough for me to figure out what to do when a problem arises. I understand that documentation is perhaps not a priority for the project. As Hans Hagen wrote in `? Context',

"... writing can get a lower priority in a time when quick and dirty answers can be found on the internet, mailing list or wiki. ... For what it's worth: whenever I have to solve a problem with a program (or language implementation) I run into cases where I have to look long to find (non conflicting) information. It just comes with the problems one wants to solve and TeX (ConTeXt) is not different."

When I started learning ConTeXt, I thought it wasn’t well documented. I changed my mind since that. I understand that writing documentation may not be a priority for Hans, becase: - He is the main developer of ConTeXt and LuaTeX. - He fixes the bugs users or developers report. - He kindly implements reasonable requests for new features. - He checks the mailing list and replies to some of the questions. - He still writes documentation for many of the ConTeXt features (not only literate programming). In my personal case, many bugs and new requests were fixed or implemented in less than a week. (I wish it would be the same at work, where we have paid support.) I’m not the only user that has experienced that. Hans may be not writing documentation, because he focuses on other (more important) aspects of ConTeXt development. I would like to write a manual for ConTeXt. But it will be in Spanish, I may need years, and it will be limited to my own experience. The wiki is an informal invitation to the users to document the issues the were able to solve. Hans explained his take on the documentation some years ago: https://mailman.ntg.nl/pipermail/ntg-context/2010/047500.html.

However, there is not a great deal of information on the Internet about solving ConTeXt problems. Whenever I ran into a LaTeX problem in the last several years, I used to find a solution, or a substantial step towards one, that was already documented on the TeX-LaTeX Stack Exchange. Unfortunately, it does not contain a similar amount of information about ConTeXt. In any case, I have started posting ConTeXt questions there, see https://tex.stackexchange.com/q/397463/146025 and https://tex.stackexchange.com/q/397607/146025

There is no secret that ConTeXt is way less popular than LaTeX. It is even trickier to search for it (since context is also a rather common word). I don’t know how many ConTeXt users are in the world. Probably the number of LaTeX users is a hundred times the number of ConTeXt users. This explains a lot of things in terms of written explanations. Let me mention a single case. TUGs are TeX Users Groups, but most of its participants are LaTeX users. The TeX Spanish mailing list has almost 400 participants. There may be other ConTeXt users on that list. But LaTeX is by far the most common topic in the vast majority of messages. My impression when I explain there basic ConTeXt features (such as modes) is that I’m describing Martian life to the generation of my grandparents (they have no clue of what I’m speaking about). StackExchange and similar platforms are great, sure. But being responsive to questions for users in the mailing list and in SE may be too much for most of the users. At least, my day has only 24 hours ;-).

As for the mailing list, I posted three requests for help here in the last one week. When I bumped my first message after more than a day without any response, Tomas Hala kindly came to my rescue. The other two messages, posted four and two days ago, have not elicited any response yet. This contrasts with Aditya Mahajan's experience, which he describes in his interview at http://tex-talk.net/2012/08/textalk-an-interview-with-aditya/:

"Most questions on the context mailing lists are answered within minutes ... look in the main manual and the wiki to see how to achieve the particular effects that you wanted, and if you cannot figure it out, ask on the context mailing list."

When the interview took place, the replies in minutes in this mailing list were the rule. But these times are over. Sometimes people don’t reply because they don’t know the answer. As newcomers, we all may suffer from the condition “my question to the mailing list went totally unnoticed”.

I understand that we are all busy, and it is not easy to find time to answer novice queries on the list.

Replying to questions takes times. It doesn’t matter (at least to me) whether the question is basic or more complex. The relevant factor is whether I can give a reply or not.

Lastly, I found the Wiki useful sometimes for quick and dirty solutions to problems, and sometimes as a source of documentation. Yet there have been several cases where I could not find what I wanted in it.

Welcome to the club. It is fine that you ask what the wiki can do for you. But when you find the solution to your problem, consider what you can do for the wiki. Expanding the wiki is the reverse action from replying questions. The wiki requieres time too.

For example, I could not find the answers to my last two queries on this list and to my two questions on SE, in the manual, on the archives of this list, or on the Wiki. Again, it might be that people don’t know the answer.

These problems mean that I have to spend much more time experimenting and searching before I find a reasonable way to get the things I want with ConTeXt. Given the beauty of the results that ConTeXt returns, I don't mind that. However, I wonder if there are others who have similar problems learning about ConTeXt, and if they have some words of advice for me. I am also interested in ways to help people with similar difficulties.

In my experience, what you learnt from LaTeX won’t be helpful to use with ConTeXt. Unless it is plain TeX, ConTeXt is diverse from LaTeX. My last statement may be obvious, but it drove me crazy almost for years. I couldn’t understand why I was a newbie in ConTeXt (and I couldn’t achieve the most basic things) after having used LaTeX for more than five years. If you want to help people to introduce themselves in the use of ConTeXt, start writing a manual. Maybe not now, but don’t wait too much to compose an introductory book on ConTeXt. I hope it helps you to understand the ConTeXt ecosystem. Cheers, Pablo -- http://www.ousia.tk