Experience of creating Reshebnik based on MediaWiki

This article is a detailed commentary on the topic “Karachai-Balkarian Wikipedia written in six months starts with 900 articles . ”
In short, he would have sounded like this: “The work has been done tremendously, but if a person is interested, this is not a problem. But then you just wonder how it was all done. "
Further it is expanded - about the experience on an example of wiki-reshebnik + Pi . Who cares, I ask under the cat. Carefully, PR :)

I always liked mathematics, and in the university I often did calculations for matan for money. When I got the Internet, I immediately got the idea: “to collect all the tasks I have solved and put them into the network”. The idea, as often happens, went into a long box, and 5 years later (in April 2008), after the end of the uni, was implemented as an “Internet Reshebnik + Pi” based on the MediaWiki engine.

The advantages of MediaWiki for me were:
- formulas using Tex-markup ;
- the ability to always easily fix added tasks or roll back erroneous edits.

I decided to fill each task with a separate article. The main task for the solution is chosen problem book Kuznetsova LA (Collection of tasks in higher mathematics (standard calculations)), according to which we were given calculations in Kiev Polytechnic. I am sure he is known to many.
')
Then for some reason it seemed that it was enough to start and immediately join a crowd of like-minded people, and the site would begin to replenish like an avalanche. But this did not happen.

Tasks and solutions to them were typed by me. The work went slowly also because the solutions that had been developed during the course of study were either distributed or lost upon moving, and I had to decide everything all over again. But as I wrote above - I like it and this is not a problem :)

To draw attention to the site, I created a Vkontakte group , where at that time I spent a lot of time. The group aroused interest at the expense of interesting tasks, and Kuznetsov's problem book remained mainly my interest.
Interesting tasks and now are on the site: Interesting tasks and puzzles .

Six months later, the site already had more than 500 tasks with solutions, of which 95% were added by me. Interest in users' PlusPi has awakened, as it is not difficult to guess, before the winter session. Then a few more people appeared who added solutions, and it was damn nice. Unfortunately, the new editors did not last long, but about 300 solutions were not added by me, and this is quite a lot.

For the year of existence, the task book already contained about 3000 tasks, of which 2000 were with solutions. 85% resolved and added by me. After a year of the site’s existence, I’ve had less time and haven’t added any solutions for the last 10 months, but I made sure that the information wasn’t overwritten and added correctly.

If I had been told earlier that I would solve and type so many tasks with my hands, I would not have believed it. And now it is a reality and does not seem to be something complicated.

A few of my conclusions during the life of the site:

• users do not like to add tasks and solutions;
• users like to correct errors in tasks and are even ready for this to understand the markup;
• It's easier for users to find my soap on the site and send me a scan of tasks than upload a picture to the site.

Anticipating questions, I immediately answer a couple:
1. Why this is needed if there is a “Wolfram alpha” .
Unfortunately, he decides only in one way, does not always describe in detail the decisions and forgets about explanations in Russian :)
2. You help lazy people! Let them decide for themselves.
Who seeks will always find. And you can easily check whether the person himself decided or wrote off. To do this, at the university, we had the protection of payments.