Ragezone Open Development Server.

[gzuz]

Okay,

Main reason for this thread is that i have a spare server...

I have no real use for that server other than "fannying" around with priston tale every now and then.

Now, i feel a better use of this server may be to open a development server for pristontale, the only way you will get a user account on this server is to be a ragezone member and actively show that you have an interest in the community. The whole idea behind this is that i want to centralize development so that we can work productively together, the server will be a client/server based server where you will connect to via vpn client console. I will (when i get a new hard drive for the server to support all user accounts) be issuing 50 accounts for the time being, to users who have shown interest in developing this community. Each account will have 4gb of private storage, as well as access to the PristonDev directory drive, this will be 200GB of space where will be filled with priston tale files you can utilize. I will have to work out a ranking system so certain users have certain rights depending upon the amount of contributions opposed to the amount of items taken.

i just want to get started on something that could kick off a huge developmental jump.

The users of wow have managed to completely rewrite the server.exe and database.

Q: What is stopping us?

A: lack of a centralized developmental team.

The server will feature a full MS Visual Studio 2010 Suite, along with Java and access to alot of the adobe softwares (photoshop, illustrator, dreamweaver)

Perhaps this project is something people wont want to do, but i feel we could benefit so much from it.

Look into the source of the mangos wow server, or even the SWTOR project that is just starting now on ragezone.

It is what we need to do to get to somewhere bigger, if we have a workable area, we can all communicate, collaborate with our work, and share sections of our own code with one another.

Answer in the poll whether you would like to start this project with me or not.


Thanks
Gzuz

 

[bobsobol]

i feel a better use of this server may be to open a ragezone development server

Woa! Hold on, right there. I've seen this done with other MMOs on RZ, and MentaL takes a *very* dim view of pServers attaching the RaGEZONE "trade mark" without explicit authorisation from him before the venture starts. *If* he gives consent to use the name, he will probably want to provide the hosting.

So first of all, if you want to tie it in to RZ that much, talk to MentaL first. If you're not that fussy about a name, then a community server can probably still have something like a "valued partner" status. But again, speak to MentaL first! It's rude to simply assume that he'd be prepared to put his brand name on something he has no control over and could (probably will) reflect on the community he has worked hard to build up over more years than any of us have been active.

Remember, he is not only "da boss" around here, he is also "community member No. 1". This place was created by him. It was his idea. His work, and aside from his music and GF (and the pup) has been the majority of his life all these years. He has fought several legal battles and paid legal fees to gain the expertise required to keep our home open to the public, shifted servers on several occasions to avoid legal shenanigans and is constantly topping up the hardware, caching and redundancy facilities needed to support such a large database and active front-end user activity. Donations and sponsorship may make that financially viable, but dolla doesn't do the work.

You can't expect him to just trust that you will be so dedicated to this development, and any association (however terse) can be picked up by legal eagles forcing him to explain the *exact* nature of the relationship. That's embarrassingly difficult when you aren't aware that there is one.

Now... this "VPN Console" thing. Why? MaNGOS has a GIT (previously SVN) repo and forks. Developers dev on their own systems and post patches for approval. The project managers bug test patches in the betas and merge them into the release editions after bug reports on that section fall off. This is reasonably normal practice, and it usually works.

Aside from anything else, it means that every MaNGOS dev in the world (including me) has a copy of the MaNGOS repository (and the UDB one) on their system should someone behave maliciously. All those devs are not going to be stingy with those files in getting the repository back up.

 

[V for Vendetta]

Well this seems like a good proper idea but i think you need to talk to ''The boss'' which is in this case ''MentaL'' so you'd rather wait till he he posts a reply or you mail him that choice is yours good luck.

 

[cloak]

I think co-operative development would be essential. I do not approve however, of this "ranking" system if it ends up being something where only a select few would have full access to developmental files. I think there should be full transparency for everyone, otherwise, what is the point? It may not even be your intention, but I don't like opening any possible door for elitism, thats what got us where we are in the first place.

That aside, you're offering free hosting, that is a great deal. One problem is it kind of takes away from people that want to run thier own local server. So would exporting developed files be an option?

Another thing is what files do we work from? The MPT ones? I'm running off of one hour of sleep atm, I may be taking your words out of context even, I don't know. I just want to make sure this ends up being what it should be and doesn't steer away from what im thinking it is.

Bob makes a good point about the RZ name lol. We could call it something else though and vote on a name. I think personally it's better if everyone devs off thier own cpu and like Bob said a repository in conjunction with posting all updates on the forum.

We could make an official server dev thread and base it off the MPT files since there are the most recent by far. I think we can at least arrange a stick if needed to keep it on top, have everyone contributing post finds in olly and hex, post links to updated files, and then every few days or so someone can re-pack everything and repost it with updated link in the opening post.

Just keep the thread moving on and on and on. But keep the chatter to dev as much as possible. Thats my 2 cents.

 

[MentaL]

I know your intentions are well but we have a bad history with servers, plus I don't really know you. Although I am open to ideas with people I know and trust.

2

 

[V for Vendetta]

Why MentaL some say that RageZONE has the best server History when it comes to Legends of MIR so i can't say that the history is bad after I've heard some people say that the days of the Legends of MIR servers where so good that they actually miss them I'm not a legend nor a Legend of MIR veteran but mostly the certaint posts proves it all.

 

[gzuz]

i dont mean to call it a RZ server, i understand that its not exactly the thing you want tied to your brand if it all goes wrong, my intention was simply a server where from which the users are selected from ragezone alone, i have never met a developer as trusted as i have found here at ragezone and personally i dont want to recruit from any other dev forum than here, im not suggesting the server be tied to ragezone in any way if you dont want it to be, that wasnt the idea, just that the users i know and trust from here could be included in the project. Sorry if i came across as rude but that wasnt intentional.

 

[MentaL]

Why MentaL some say that RageZONE has the best server History when it comes to Legends of MIR so i can't say that the history is bad after I've heard some people say that the days of the Legends of MIR servers where so good that they actually miss them I'm not a legend nor a Legend of MIR veteran but mostly the certaint posts proves it all.

Mir doesn't count and yes I ran the best ;-)

2

 

[gzuz]

Sorry i think i got ahead of myself there and got carried away with what i was writing. Well the idea was that i have that spare server sitting there that could be more productively used to create a centralized work environment for the more enthusiastic developers that i know here. That was all, i dont mind about ties or binds, i just want to get some good work done.

 

[bobsobol]

my intention was simply a server where from which the users are selected from ragezone alone

There is an "implicit" reverse association in that which says "If you want to get developing with us (get files?) then you *must* go through RaGEZONE."

This alone can attract signups of people who are "undesirable" to RaGEZONE. They don't want to be here, they just want your files.

Additionally, auditing *true* RaGEZONE members would require database connectivity to information on RaGEZONE servers which may be in breach of our privacy agreement with users.

--- EDIT ---
Basically, engage brain before opening mouth. There are many implications in what you suggest which require careful consideration.

 

[V for Vendetta]

Well just saying MentaL *If* it would count then i would certaintly count that with RageZONE because compared to other forums with their servers have they never ran a server so good like RageZONE no other forum can run a server for so many years only RageZONE could due you obviously because I've never seen a forum that has been running since 2000/2002 and have a server with it so you should actually be proud of yourself that you've pushed it so damn far.

 

[gzuz]

Fair enough, well, i suppose there could also be a private signup for the dev server on the web server, but i would really ultimately check their background from what i have seen them release or contribute here, whether its officially tied or not i think that if we started out with a team of head users to recruit new devs based on what you know about them / what you have seen them release / the work you have done with them in the past. I dont really want to cause any controversy over this, its just really an idea that i want to develop into a private team that would focus to supply solely the ragezone community with up to date and high quality releases. I dont want to tread on anyone elses toes, or p*ss anyone off, just wanted to propose an idea that could be shaped by any user here who has an interest, with the use of the resources i have (web server + data and game server). I dont care if we use my server, i dont care if we burn through hard drives and processors, as long as we get somewhere, And that, is all i want to say. Whether its a GIT or an SVN or a VPN! i dont care, i would just like to be a part of something bigger that is going to bring our community forward and expand the possibilities of what can be done, also itd be kinda nice to create our own server for something so theres no way we can be shut down by any other company for using their db or code.

 

[bobsobol]

I think co-operative development would be essential. I do not approve however, of this "ranking" system if it ends up being something where only a select few would have full access to developmental files. I think there should be full transparency for everyone, otherwise, what is the point? It may not even be your intention, but I don't like opening any possible door for elitism, thats what got us where we are in the first place.

The original suggestion was SVN, and I'm happy for that, though GIT rules for "forks" and merges, SVN is quick and easy start place.

Does RZ not have a hierarchy? How Elitist is MentaL, or me? (wow, that's a can of worms ) How "transparent" are posts? Can anyone delete the work or others?

Usually you have repository managers and open repositories. Everyone can get the files, officially accepted devs can submit patches, only management can actually "appy" (or merge) those patches.

This is necessary because patches can conflict, and there really isn't any way to completely automate "conflict resolution" in source code... let alone a hex dump.

I think that's the kind of setup gzuz is suggesting.

We could make an official server dev thread and base it off the MPT files since there are the most recent by far. I think we can at least arrange a stick if needed to keep it on top, have everyone contributing post finds in olly and hex, post links to updated files, and then every few days or so someone can re-pack everything and repost it with updated link in the opening post.

Just keep the thread moving on and on and on. But keep the chatter to dev as much as possible. Thats my 2 cents.

Again, MaNGOS devs "chatter" on forums, and unofficial devs submit patches to forums if they think it's good, but don't know if it will tie in with the mainstream and such. I absolutely think that's a way to go, and something of a PTDK should be put up here to help devs work with / on the files. Something to make concise patch files in a standard format, the dump / undump tool for binaries etc.

 

[gzuz]

Ultimately i was thinking of picking up and starting a complete server overhaul. Obviously there should be at the early stages some development of existing files so that we can release what we have learned in a central place so that everyone involved can access this based on an indexing system so you can find related items in the correct place. This would help for understanding when it comes to coding the server. The reason i have suggested this is that i have seen that Sheen doesnt want to continue with his PTEmu anymore, and to continue this community we need that. There is no point spending 2 years to only catch up with what the latest feature is and fully support every part of it, when we could spend 2 years coding, and have double the features that the official game actually has. Maybe im wrong but it would seem to me that any community that is actually flourishing is one that has a team based developed server. I dont know if anyone else wants to do what i want to do, but the whole thing that sheen had started with his server emu is the track we need to be continuing. Development will be slow and difficult if we do not head in that direction. Id much rather write a few lines of code to represent a new feature than spend hours searching for hex values and memory addresses changing JMP commands to different offsets in a pretty much botch job to get where we wanna be. If we redesign the server we could make it much more resource friendly to allow it to handle much more concurrent connections per server, run multiple threads per host to allow different servers with different focuses ie PVP PVe and Raiding. The rest could be modifications upon the client via dll's, but to have the understnading of how the server will use and process the data will be a big help when sending packets from the client. As i say, i wanted something to be shaped by other users, not just what i think. Dunno if you like the idea i have, but as i said, whatever works, im happy as long as we are getting somewhere.

 

[bobsobol]

Sheen should probably say this himself, because he words it differently (having a different perspective on it) but he wanted people to contribute to his source, but was "torn" about sharing the source which he felt was primarily "his". IMHO those two desires where in conflict... but that's all "water under the bridge" now. His last open source repository was archived and rehosted, so we could pick up from that if people wanted to.

I believe, and I don't know if he will want to confirm or go public with it, but I believe he has started reworking his ideas in C++. Again, alone.

Right from my beginnings here, and all my stuff on making DLLs and plugin systems, I hoped we could get a server (and client) for which we gained more and more source by removing routines and importing equivalent code from libraries which we have *real* source code for. The idea being that each time you write an equivalent routine, that is one less part which would need to be implemented in an alternate host. Ultimately making the "alternate host" simple to write from scratch and provide source for, thus removing any binary reliance. In the meantime, you have a working server which will test the code of each module to the limit. ^_^

Vormavs DLLs are still problematic in the way they require offsets for data (and code?) in the host. That is still too tied to a specific type of host for my liking. : The problem with the C# emulator is that it cannot be intermixed with existing server releases at all to produce a comparable server *now*. :/:

In short, I agree.

 

[gzuz]

Incase anyone is interested, i started work on the site about an hour ago

You must be registered to see links

 

[bobsobol]

Unrealistic goals I think.

You could do that, but it would corrupt every character on your server if you did anything other than "add" levels. Changing the experience of lower levels "on the fly" would just break every existing player.

It could be done if you force logged everyone out, modified the exp of players to match the new table before allowing them back in... but you may as well take the server off-line to do that.
--- EDIT ---
Things you could store in the server SQL would include player data, warehouse data, shop inventory, NPC locations and data, monster spawns, monster details, PostBox data, logs. Additionally, if you implement "extra packets" you could distribute item lists, maps, ageing failure rates, mixtures and other crafting recipes to the client. But those would take some considerable re-working of both client and server.

Maintaining compatibility with official clients is a "popular" thing, so if you do that, that should be your first "fork". (ie. main fork does not "improve" the original feature set... it may work better, take less memory, be easier to configure and so on, but will not add or remove features, the second fork will be fixing fundamental flaws in the original design)

That's stored in the client, not the server... I'm not sure what you imagine achieving with that.

a share in all revenue generated from Support Donations.

This is asking to be shut-down.

Any emulator must be distributed purely in source form, for no profit and purely for educational purposes.

I believe individuals can donate / offer reward for bug fixes, but these must effectively be a private transaction which does not pass "through" the project, which must be a non-profit organisation. Advertising or sponsorship, even charitable donation is acceptable, provided it only ever goes towards running costs, and never individuals or labour cost. (as far as I'm aware)

Binary versions of the project setup will also be available

Again, that's dubious. Leave that to RaGEZONE members to make "repacks" is my recommendation. (have you noticed that MaNGOS, SD2 etc will never offer binaries, even though you can find them all over the web? There is a reason for that. )

 

[gzuz]

skill damage is not stored client side, delays are, but i never mentioned that. These will be stored in a database, as will item values, the pre existing dat files will be stored in a database. The website isnt the final cut, as i said i started it an hour before posting the link. I filled it out with crap text so that it wasnt a bare shell. I never said anywhere that the emu would be distributed as a binary. I mentioned i would add other useful source code's that users could modify or if they wished could be distributed as an executable, which i have no worry of being shut down over as all those things would do is modify dat files and download information from the net, or update database values, in which case there is no solid connection to anything breaking any kind of rule or law.

I plan to use the SVN to be the sole distribution method for the project. I dont understand the issue with support donations however, ragezone takes donations to keep the servers running and no one has an issue with that... i probably wont even end up taking donations as i never have with my servers previously and probably wont start now.

If smart database programming is used you can update on the fly experience, you do not need to disconnect anyone, only force the client to update its variable values of what it has stored, if an increment is made in experience gain needed or a drop then the server can have a built in sub to handle such an event. Thereby re-sending the correct values to the client and keeping everything running smoothly(provided you have a client with an adabptable xp table) obviously at present we do not have such a client but i believe forward thinking never hurts.

Hold tight and i will get the site into a more release worthy state, the main reason i am posting it here is that i am enthusiastic about starting the project and i wanted to get some design input of what users who might want to be involved would like to see, basically because i cannot speak for everyones taste. The content on the site at the moment is irrelevant, i am trying to get a partially filled shell so i can visualise how the project will look upon completion.


Edit--------
Removed the issues you had from the site.

 

[bobsobol]

Yea... the "design input" is kinda what I'm offering. Only, design of "words". New wording is much better.

People are constantly trying to shut RaGEZONE down, and MentaL works hard to keep it up. The service(s) provided by RaGEZONE are not those of a server, or a development project. That's part of why he has to be very careful what the RaGEZONE name is put to. Essentially it's an MMO developers "speak easy". So if you shut RZ down, the development would continue, but communications would have to move to eMail, IRC or MSN or such.

pServers which take donations are "dubious" in legality, if those donations can be shown to go to individuals. (ie. as profit or earnings) They must only be used to fund hosting, and accounts must be able to prove that. However, you can make "personal donations" outside the organisation. (to a friend, for any or no particular reason) Very complex, I know. The fact is, friends don't ask for donations on a website. You can offer "rewards" for anything you like to strangers, and because you are asking for a service, rather than being offered (marketed or sold) that service, that's okay. (AFAIK)

As for the dynamic experience, that's what I meant about it needing to be a "fork"... because you then need a client development project, and once you are changing the client, it's "not the same game". (apparently ) You could do it "off-line" via a launcher. You would probably also find that a normal client would DC players if the server started telling it it's Exp was all wrong.

 

[Vormav]

Vormavs DLLs are still problematic in the way they require offsets for data (and code?) in the host. That is still too tied to a specific type of host for my liking. :

Those that add new functionality must also add new code =P

This look like interesting project and since you are providing "space" you can name server whatever you like but I would call it OPT, OpenPT or DPT, DatabasePT since you planing to move "everything" to database
I just wanted to ask if this will be something like rPT? PT code rewritten to source and injected as .dll?