Well, I'm sad about the loss of linux support for this EXTREMELY COOL PROJECT, but I can adapt :smile.gif: I intend to try and get it working in wine once you guys release. This is EXACTLY the kind of tool server admins need.
You should look at the Java mod called simply "Minecraft Map" or "Minecraft Minimap" depending on which version it is. It is on my site, http://minecraftam.com, and the forum thread for it is in my sig. It may do what you want even better than this project will, and it is already functional and released.
If you want to try the new AutoMap in wine, you will need to install the XNA framework first in it, so good luck!
I specifically am looking for a minimap along the lines of Minecraft Map, for SMP and with real time terrain updating so I can see and stop griefing quickly. Perhaps I misunderstood the feature list - will AlphaMap be updating the map as it changes? Or is this just a new form of displaying the existing map, like Cartograph or AlphaVespucci, with overlaid player positions?
Minecraft AutoMap does update in real-time, but it gets all its data out of the client, which can only have so much loaded at once. Therefore the window of real-time updating terrain is very small compared to the whole map.
Such a mod could be possible on the server, but it is not a high priority for us at this time. You would probably be better off with Minecraft Map (my Java client) and grief alert for now. Also, if you find it may be useful, I have written a Hey0 plugin that logs the locations of all connected players every 10 seconds to a file that is easy to parse. it is on the minecraftam.com website. I haven't tested it thoroughly though, so don't just stick it on an important server without trying it for a day or two somewhere else.
Alright. I did find a way to determine whether or not the Minecraft client is on a multiplayer server. I even found a way to get the server address out of Minecraft and into the AutoMap (by modifying another one of Notch's classes). But there is a problem. Anyone with even minor technical skills can bypass this security and use ore detection in under an hour via at least two different methods.
One very obvious way is to decompile the AutoMap source code just like we all do with Minecraft source. From there it would be trivial to find the ore detection code and disable the server check. <-- This is remarkably easy to do with all .NET apps.
The other equally obvious way is to redirect the query to the SMP server to the hacker's own SMP server, which is configured to say "Yeah, go ahead and use Ore Detection." This can be done by a 10 year old with any one of a number of very simple proxy programs.
Just in describing these two I have come up with a third way, in which the hacker would decompile the Minecraft mod and have it report his own server address to AutoMap for verification, whilst still connecting to whatever SMP server he chooses.
Further, nobody but private server owners and random oddballs are going to install another mod every time the server is patched just to let their players mine all their diamonds without looking for them.
That said, the only viable options are to disable ore detection entirely and delete the code to make it harder to hack-back-in, or to allow it to be used anywhere. You all have voted, and it is obvious that most of you want ore detection, so ore detection will be in, unrestricted, in the first release.
I want to suggest releasing an OREDETECT version and a NODETECT version separately. Giving people the choice seems good, right? And it'll save you some time, but probably not bandwidth. Not that that would be a concern. If you have problems, you can resort to mirror uploads.
The only issue was whether or not to allow it in the first place due to its controversial nature, and it has been decided that it will be allowed. It is, of course, a feature that can be toggled on and off as desired. That is no big deal. No need for two versions.
It works now, but it seems that AM conflicts with the Rideable Mobs mod, (dm.class), crap.
Awesome. I'm sorry to hear that we have a mod conflict. It is bound to happen I suppose, until Notch or some enterprising Java developer gets around to delivering a client-side plugin API.
We could, in theory, modify one of the classes that extend dm.class, but there is no telling what kind of side effects this may have. Loss of player names or other players in SMP is a good possibility.
Another possibility that I do not like is that we make separate installers that know how to patch all the various versions of dm.class that have been produced by Notch and by modders.
I've been keeping an eye on Xau's work (he is the guy who runs the HD texture fix) because he has hinted that his patcher may soon support plugins, and in a way that would make them not step on each other.