March 15, 2019 at 11:56

Prevent obsolescence of teaching materials

Briefly about the situation in universities (personal experience)


To begin with, it is worth stipulating that the material presented is subjective, so to speak, “inside view”, but according to the sensations, the information is relevant for many state universities of the post-Soviet space.


Due to the demand for IT specialists, many educational institutions have opened the relevant areas of training. Moreover, even students of non-IT professions got a lot of IT subjects, often Python, R,less fortunate many have to learn “dusty” learning languages ​​like Pascal.


If you look deeper, then everything is not so simple. Not all teachers keep up with the "trends." Personally, when I was studying in the "programmer" specialty, I was faced with the fact that some teachers do not have relevant lecture notes. More specifically, the teacher threw the warden on a flash drive photo abstracts, hand-written by some student. I am completely silent about the relevance of such materials as manuals on WEB-programming (2010th year). It remains to be guessed as to what is happening in technical schools and the worst of the worst educational institutions.


Eventually:

  • They print a lot of irrelevant information in pursuit of quantitative academic indicators;
  • The release of new materials is unorganized;
  • Often “trending” and relevant details are overlooked due to ordinary ignorance;
  • Feedback with the author is difficult;
  • Updated editions are rare and irregular.


"I do not agree - criticize, criticize - offer .."


The first thing that comes to mind is the introduction of systems based on the Media Wiki engine . Yes, yes, everyone heard about Wikipedia, but it has a reference enceclopedic character. We are more interested in educational materials. Wikibooks is more suitable for us. Among the shortcomings can be identified:


  • compulsory openness of all materials (quote: "Here, in the wiki environment, educational literature is jointly written that is freely distributed and available to everyone.")
  • the presence of some dependence on the rules of the site, the internal hierarchy of users.
    Many wiki engines float in the public domain, but I think that you don’t even need to start talking about the possibility of deploying a wiki system at the university level. From experience I will say that: a) such self-hosted solutions suffer from fault tolerance; b) you can forget about system updates (with very rare exceptions).

For a long time I pondered to no avail how to improve the situation. And once, one acquaintance said that he had long ago printed a draft of a book on A4, but had lost the electronic version. I was wondering how to convert it all into electronic form.


It was a textbook with a significant number of formulas and graphs, so popular OCR tools, such as abbyy finereader , helped by half. Finereader produced chunks of plain text, which we began to put in plain text files, breaking into chapters, marking everything up in MarkDown. Obviously, we used git for the convenience of cooperation. We used BitBucket as a remote repository , the reason was the possibility of creating private repositories with a free tariff plan (this is also true for GitLab ). For formula insertions, Mathpix was found . At this stage, finally turned to the side of "MarkDown + LaTeX", as the formulas were converted to LaTeX. For conversion to pdf used Pandoc .


Over time, a simple text editor began to be missed, and began to look for a replacement. I tried Typora and several other similar programs. As a result, we came to a web solution and started using stackedit , everything was there, from synchronization with github to LaTeX support and comments.


Specifically, a simple script was written as a result for which I am ashamed, which performed the task of assembling and converting typed text to WEB. A simple HTML template was enough for this.
Here are the commands to convert to WEB:


find ./src -mindepth 1 -maxdepth 1 -exec cp -r -t ./dist {} +
find ./dist -iname "*.md" -type f -exec sh -c 'pandoc "${0}" -s --katex -o "${0::-3}.html"  --template ./temp/template.html --toc --toc-depth 2 --highlight-style=kate --mathjax=https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML' {} \;
find ./dist -name "*.md" -type f -exec rm -f {} \;

He does nothing clever, from what can be noted: he collects content headers for easy navigation and converts LaTeX.


At the moment, there is an idea to automate the assembly when making pushes to rap on github using Continuous Integration services (Circle CI, Travis CI ..)


Nothing is new ...


Interested in this thought, he began to look at how popular it is now.
Obviously, this idea was not new to software documentation. I saw quite a few examples of training materials for programmers, for example: JS courses learn.javascript.ru . Also interested in the idea of ​​a git-based wiki engine called Gollum


Many saw repositories with books written entirely in LaTeX.


Conclusion


Many students rewrite the compendiums several times, which they wrote many and many times before them (I do not question the benefits of writing by hand), each time the information is lost and updated very slowly, not all compendiums, as we understood, are in electronic form. As a result, it would be cool to fill in the abstracts on github (convert to pdf, web-view), the same offer to do for teachers. This would attract students and teachers to a “live” competitive github community to a certain extent, not to mention an increase in the amount of digestible information.


For example, I leave a link to the first chapter of the book I spoke about, here it is and here is a link to the rap itself .

Tags:

Also popular now: