If my code isn't stable and isn't fast enough to not cause insane amounts of lag, then why would anyone want to use it? My UI is designed based upon me using the 'grandparent' test. I take my grandparents(who think facebook is powered by some sort of weird technomagic) and have them try to use it. If they're unable to use it, then i call it an instant failure. If they can then i move on with it. End usability for UI design anyway is designed to be fast but also not to just die. Server code and such is consistent, and probably makes it a bit harder on some of the others out there but my UI design has to pass the grandparent test. Since i don't have anyone else nearby(i live close to them since i help take care of them and also take care of their animals) it's easy to just have them test it.
If the UI sucks, i also know that NO ONE will use it. How blizzard was able to get past such failures is beyond me. I don't know of anyone who could possible do that. It's insane to think they could launch with so many crashes and long queues and still be able to get out of the date. It's complete insanity to me. I know i'm not blizzard and thus don't have that kind of IP power behind me so when i do launch, it has to launch and not come crumbling down with only 20 people per server(not sure if that's the exact number but one of my friends from way back said that that was the cap on each server during the early days something similar to it anyway).
Since i have no idea how in the world you're so easily able to do multi-quotes i'm unable to break up your message and directly respond to it. I have to do it as i am, and authorization was one of the things that i got a kick out of a long time ago. It used to be for er... let us call it fun activities. But now i'm keeping up on the news about vulnerabilities/breaks in algorithms, and also making sure to validate every single variable that comes from anyone. I've already implemented anti CSRF, i escape every string along with using strip-tags for the anti-xss side of things. And then i use sprintf() and insert the variables by then passing it through that into mysqli. I also, if the person has javascript enabled, hash their password 30 times using sha512 before sending it up stream to the server(since i cannot warrant a ssl certificate yet since it's currently warranting me 0 dollars) but once i can i will be getting it.
And you are very lucky to have someone like that there. I've mostly got this forum as far as some things go and then basically studying the PHP manual, learning about/keeping up with the latest attacks and trying to work around it. And the memory consumption thing was about PHP itself but i didn't say it then since my mind seems to sort of jump around a fair bit.
On the point about not too much OO code it may be that i don't look for a lot of code, i do, do a quick look into the PHP manual before i attempt to write a function or check out PECL to see if someone's already implemented it before i attempt it so, it may just be a thing of familiarity with the platform. Since You've been working with PHP for i don't know how long, and i've just begun a couple months before i first signed up on this forum. So the amount of resources/knowledge of said resources is going to be very skewed. One of the other things is that i'm more the type of person to try to solve the problem myself if it's not already inside of the main libraries that ship with a specific language.
On the release schedule, i hope to, upon actual release of this thing release patches every four months at the least and also do yearly 'expansions' which should add a ton of content and if there's a serious issue with balance i'll try to push it out rather fast. For balance itself, i'm planning on logging outcomes of fights with people and then using that information to see how it went. Regular patches/content updates are a must for this thing to succeed. I hope to later on, be able to actually let people customize their UI, but for now it's just the stock one(which is functional but probably could use some improvements but i'll find that out once i start alpha testing.)