A web-based MUD?

[Aragon]

Well, it would *have* to be open source if its php, right? :(



I'd probably only license it to a select/small number of servers, and require contracts or something to help prevent piracy (shrug?)



You can use ZEND to encode php files. ;)



http://zend.net/



ps, contact me on AIM if you're interested in some help building this... :)

[EvanElias]

You can use ZEND to encode php files. ;)



http://zend.net/





Sweet! Didn't know about that. Will definitely give it a look!



Is the encoder freeware though? Or is it super-expensive?







ps, contact me on AIM if you're interested in some help building this... :)



Awesome, thanks :) I probably won't be starting on this project til late January or so, but will definitely contact you about it then.

[Sirdude]

tell me what i can do to help and you got it evan

[Aragon]

I think it's pretty expensive to get the encoder. I didn't mention it because last time I looked into it was a few years ago and I wasn't sure if they changed anything ;)

[EvanElias]

Hmm, yeah, seems like the Encoder is still really expensive... they don't even list the price, it's one of those "call for price" things (which usually means huuuge cost)



Well, dunno whether or not I'll go through with a project like this anytime soon... depends on how much free time I have next semester...



Well at the very least, within the next month I'll have a web-based data file viewer for DoorMUD on-line (kinda like MudView I guess), since I can do a project like that as the final project in the databases class I'm taking... it'll probably be in PHP+MySQL, although PL/SQL+Oracle is a possibility too (professor seems to prefer that, but PL/SQL is kinda hideously old and pascal-like)



If I do it in PHP, I might later improve it to make it a web-based DoorMUD *editor*... that'd be pretty sweet :) And then I could re-use that code & data definitions for the purely web-based MUD if/when I get around to that project.



Another option I'm considering for a web-based MUD is writing it in C++ and using the MySQL API... and then having traditional bbs door support, as well as rlogin and telnet servers, AND a socket-based PHP interface to allow it playable from the web too! That'd definitely be the supreme MUD, methinks... but it would be a huuuuuge project in terms of time commitment, so I really can't say if I'm going to go through with it.

[Aragon]

if you're going to write it in c++, might as well have your own db structure so you don't have to rely on outside applications. It will be eaiser to install and maintain. (though harder to code...)

[EvanElias]

True, but I've begun to realize that a SQL-like relational model really seems to work *perfectly* for a MUD... it's just a really good fit. For example, using a relational model, you don't end up with crap like MajorMud's 10-spell limit... in fact you don't end up with any array-based limits at all, which is great!



Also, using SQL, you can very easily obtain statistical values from the data... for example, computing the average exp value, monster HP value, damage value, etc for an area is just a 5-line SQL query. Similarly, you can easily calculate how much optimal damage each class can do at each level, which is insanely helpful for game balance purposes. (I bet that if Metro could do this, MajorMud wouldn't have all the balance problems that have plagued its entire existence ;)



So yeah, there's downsides of course, speed/resources being an issue... altho I was surprised to learn recently that tests have shown MySQL executes *significantly faster* than MS SQL Server, and is even on par with Oracle (which costs tens of thousands of dollars).



And of course you're right, the biggest problem is requring outside software, which will significantly reduce the number of systems running the game. On the other hand, maybe this could be a good thing... DoorMUD's biggest problem right now is that there's more boards running the game, than there are people playing the game (doh!)

[Aragon]

When you talk about MySQL vs. MS SQL vs Oracle performance, you have to mention what the test conditions were ;) For huge databases, Oracle will kick MySQL's arse. Also, it depends what kind of queries are being performed most of the time. Don't remember much more - i didn't really pay attention during my DB Class :-p



But basically, it's like saying "Linux is Faster than Windows"...



For a truly web-based mud (one that runs thru CGI - yes PHP thru the web is considered a CGI application), I agree, MySQL is probably the way to go since it's the easiest and since you're running it on a web server anyway, why not take advange of the resources available.



For just an average Joe Schmoe that wants to run the game from his HD, requiring him to install & configure MySQL (or whatever SQL) is a turnoff.



Personally, I say use MySQL. The advantages outweigh the drawbacks.



advantages of MySQL:

- easy to code

- easy to backup

- easy to edit

- db in plaintext (maybe)

- more portable (php, perl)



disadvantages:

- relying on an external program

- setup is more complex

- db in plaintext



advantages of proprietary format:

- easier installation

- easy to backup as well (just backup data files)

- probably more secure



Disadvantages:

- harder to code

- might not be as portable

[War Hawk]

Install the mySQL odbc driver for windows.... and now you have a portal from a windows program to a mysql server...... :)

[EvanElias]

Install the mySQL odbc driver for windows.... and now you have a portal from a windows program to a mysql server...... :)



umm... already discussed that... :)



Another option I'm considering for a web-based MUD is writing it in C++ and using the MySQL API...



same difference, except mysql API is a more direct route to the db calls than odbc.