In late 2007, I /so to say/ cleaned the perl code and submitted it under GPL to SourceForge http://sf.net/projects/tclub It takes a while to install and configure it on a Unix server, and I believe the instructions are good enough for the task. One good thing about that release is that I put all strings into a single file, so that translating the interface into another language is just a matter of translating all strings and putting them as a new file. As two years passed since then, the current version of the program which is running on Motivation site differs much from the one released in 2008, as I wrote more lines of code and generally refurbished the outlooks. However, the older one containes a lot of useful code so anyone experienced in perl may need little time to use it for a start up of their own projects, let's see, in bowling, table tennis or snooker.. (?), to say nothing of tennis. So far, so good, I leave this intro and will follow the narration, starting from this year, under a different title: Cloning.
One may say "10 million people and only 100 players - this is not a good statistics at all". Partially agree with that, but keep in mind that only one of 7 people who join the club profile list by registration would start actually playing (15%). So we have now 170 player profiles , 120 profiles active (that is, visited the club in last 30 days), of them 96 played at least one match; in order to get the club rating calculated, any player must amass at least 24 rated matches in as many as 104 weeks, i.e., must have a rated match played at least once a month in average; so the rating as actually caluclated for 30 players, each one, in average, playing some 40 matches a year; these are the "core" players of club, 25-30% of it.
-- this picture shows the number of visitors (green) and pages seen (blue), daily average, for the past 12 monthes. The stats for both have grown 200-300 percent.
Currently, according to the DB of four years in Moscow lounge, 133 players have together played 1792 matches in singles and /only/ 21 matches in doubles; of these 1800 matches, 82 per cent were "rated" matches; the match is rated if A)at least two full sets were played B) the match is first or second individual (i.e. non-tournament) between the same players in the ongoing year. So we do not rate one-set matches, that is, do not take into account one-day tournaments. Instead, we do two or three "long" tournament every year, with usual best-of-three-sets matches. Such tournaments would not be possible without a software that imposes flow control upon the routine of matches assignments, dates, negotiations between players. We have used it for 12 tournaments, and now the thirteenth is under way.
I must admit that I have never made any good web design in my life /and probably never will/, so whatever was the substance under cover, the cover looked a) chaotic b) unprofessional. I didn't even try to use CSS, just printed lines on screen. Many times I was ridiculed by the ungrateful users, saying, hey, this site of yours looks so bad, you must be an idiot. Then I say, - maybe you know a webdesigner who would do that sort of work for free? Now, stop talking uselessly and enjoy what you have so far. So it lasted about three years, and last summer as we had had around 100 actively playing people, and adding new fuctions created a load on the site, it became obvious that some changes had to be made. I rewrote the engine again calling the new interface a 'MINI' to mention that it's less consuming and more fast; in the same time, heavy use of css and mootools was to add better looks to it.
- this picture shows how dramatically the load dropped after the MINI interface was used starting January this year; and monthly numbers for visits remained generally the same, but the traffic was reduced 10 times.
As a nuke reactor needs a mass of radioactive stuff for operation, so my new program needed a a big city for a start. I was lucky to have a 10 million Moscow, Russia around, with its hyper-expensive courts and short summer. I visited a nice tennis forum where I had myself made a few acquaitances before and left a link to my site, just saying something. In two weeks some 20 or 30 people joined, which was a big progress, and the "reaction" has been self sustained ever. Though some additional services were not present at that time, but generally the program has now the same meaning as it had four years ago. That is, to organize players and keep record of the matches, then to form a bulk of statistics from the matches and to calculate a rating.
This is a comment. Ok, back in 2005 I was developing a huge web application for an institution to control their disbursements. Following a tedious TOR, in half a year a 30,000-line chunk of perl was written while gradually I was musing over how effective these IT technologies would have been if applied to fields other than money spending.So in spring of 2006 I partly rewrote the disbursments engine to add a support for flow control that occurs in amateur tennis society. Let's say, how effective is waking up and calling on the forum, hey, is there anybody to play with me today? In percent, that figure is 10. But if one gets in use of a specific interface and puts an ad with desired match proposal, this message deliverd to other player by email through subscription, then empirically this figure rises to 60-70%! Here we are with a 70 per cent chance to find a partner for a close date, almost never missing good weather conditions and playing a mean of 40 matches a year. This is statistics taken from now 4 years of it.