Board for everybody who is interested in BrainKing itself, its structure, features and future.
If you experience connection or speed problems with BrainKing, please visit Host Tracker and check "BrainKing.com" accessibility from various sites around the world. It may answer whether an issue is caused by BrainKing itself or your local network (or ISP provider).
Storing a few messages in db doesn't really affect the operations of the site (unless something really stupid has been done when setting db up) - and as Fencer pointed out, the problem is not with slow operation, but rather a bug in the server.
As for me becoming a paying member, it's a bit off topic here, and as far as I'm concerned a closed one too (for now anyway).
Thanks harley, but I'll wait and see a bit further. 10-18 dollars (or even more) is not much, but the trouble of sending it and such is a pain in the ass... Besides, I don't really need any of the extra services at the moment (maybe when my holidays start - this really messes up my work). Anyway, I like to see how the site's operating for a bit longer before I go to all the trouble.
Fencer: what? Testing everything all at once is the _only_ way to do it!!! :D
OK, I appreciate what you're doing, I really do. And I can really understand how having a regular, full time job can seriously affect your normal life = this site. :)
BTW, is the deadlock caused by the Mortbay Jetty application server? It's about the only component on your technical info page I'm not familiar with, and would really like to know if it has some bugs (for further use)...
Edit: Hmm... It seems to say that it's the database driver... Forget I asked. :)
Correct me if I'm wrong, but I seem to recall that you had two processors in your servers, and any java thread would run on only one of them... Is all of the processing done on one thread or does downtime only happen when two (or more) threads are maxed out?
Anyway, killing those threads after given time-out is probably not a good idea, as there might be other reasons for 100% cpu usage (like: a lot of users - killing the thread would only make things worse). Of course if that's all that can be done, then that's what must be done... :(
I'd really love to try and help you out, but:
a) I don't have the time
b) I don't know enought of your system
c) you seem quite competent your self :D
Thanks for the site anyway. Maybe some day I'll even pay something (for the record, the reason I'm not a paying member is that I don't like to pay on the net with anything except Nordeas "button payment" - and that's only implemented here in Finland, possibly in some other northern european countries too). :)
Simple solution is quite easily implemented. The server knows how many users are logged in at any given moment, and propably some value can be set, after which (auto)login for non-members redirects the user to the page I mentioned before...
The "correct" way to do this would of course be to monitor the actual strain on the servers - much more difficult, although possible.
Edit: Fencer of course knows best. I have no insights to the software. :)
Edit2: 10-30% was a rough estimate, or as some people like to call it, a guess. :)
"I will address this directly to The RAt, since you spoke on behalf of the non paying members, but any pawn is welcome to answer this:
HOW do you suggest Fencer increase revenue? I am sure you realize that 250 memberships does NOT provide enough revenue to run this site. We all hope that Fencer will run the site forever as a hobby, getting almost no sleep, just because he is a fan of board games. But, that might be too much to ask. So, I repeat my question: How do you suggest increasing revenue? You oppose measures to restrict what pawns get, calling such measures "attempts to discourage pawns from coming to the site," so what do you suggest?
I'll help you out with this one: To increase revenue, there has to be either 1) more memberships or 2) a higher membership fee charged. "
I agree. There has to be an increase in revenue. But the matter is a bit more complicated than what you make it sound like. Here are a couple of points I use when I think of how to do it:
a) there has to be enough players on the site for it to actually work
b) most players are not willing to pay to play - I mean that there are always places where you can play for free (I was actually going to put up one for myself and a friend of mine, but then I found this place...)
c) There are always some who are willing to pay, maybe 10-30% of the users - depending on things like price and what you get for it
d) increase in price reduses the number of players willing to pay
e) decrease in what you get if you don't pay increases the percentage of willing payers, but might actually decrease the number of people doing it
f) each player can be said to cost x $ to the site
Now, the obvious answer is to increase the amount of players, as about the same percentage is willing to pay. This would lead to increase in revenue... Unfortunately more players means more processing power required and therefore increases expences too. Furthermore, this is not easily done...
Other alternative is to increase the percentage that is willing to pay. This is what has mainly been discussed here (I haven't read the latest posts yet, sorry if I'm mistaken. I'll answer this one first). It might work, usually it does not. IYT tried it, and it didn't work very well, and they had a huge amount of players to start with (compared with brainking). I don't suggest this line of action.
The last and toughest alternative that I can think of is increasing the total number of players while decreasing the cost of a single player. To achieve this, the following seems essential:
1) reduce the amount of processing and data required for using the site. I mentioned before that there is a huge amount of useless information on the screens. I believe that processing times could be cut to a third by simplifying the design (depends on how it's implemented, of course). Just look at the main page and ask your selfs, how much of the data is dynamic and how much of it do you really need. Amount of HTML seems about ok to me...
2) improve the usability of the site - I mean the frequent down time experienced... It might be a good idea to start blocking out us non paying members if the site is getting full - with a friendly (and quick) message informing us that the site is nearing it's peak operating level and therefore only paying members are allowed to log in.
3) advertice. It's expencive (unless you spam - do that and I'm gone).
What this really means is hard work and a possibility of a potential outcome. You propably already know, or at least are quickly finding out, that it's difficult to make money on a site like this. Usually the best you can hope for is to make it finance itself... And then again, some do make it. :-)
I'm a relatively new member here, transferred from IYT. I had two reasons to leave IYT:
1. regular down time (propably due to heavy usage)
2. restrictions for us non paying members (20 move limit).
This site isn't much better in aspect one, but I've enjoyed the possibility to really play here, not just hang around for 10 minutes a day. And I know I'm not alone...
I also know that us non paying members dont seem that important, but if we leave... A site like this needs a certain amout of players to work, and I think that brainking is hanging on the edge here already. I mean there are only 95 players online at this very moment.
So I advice you to think very carefully any changes to policy that might lower the amount of users on the site (I work for a company that gives advice on such matters - this one's free :D).
Anyway, good site (not perfect, but good). Keep it going. :)