All those years

[Tolakram]

You're still not thinking about this rationally. So we monitor their changes and somehow police this rule? So if they didn't we could take our team of lawyers and sue them? If they were a hugely popular server with competent changes they would be destroyed by being taken off the portal?

Give me a work flow scenario that forces someone to do something and will work for a project this small? Open source projects depend on the community to survive. It's a big enough job to manage the code and make sure complete crap does not get into the core. Policing a database project would be 10 times harder.

[geshi]

You're still not thinking about this rationally. So we monitor their changes and somehow police this rule? So if they didn't we could take our team of lawyers and sue them? If they were a hugely popular server with competent changes they would be destroyed by being taken off the portal?

Give me a work flow scenario that forces someone to do something and will work for a project this small? Open source projects depend on the community to survive. It's a big enough job to manage the code and make sure complete crap does not get into the core. Policing a database project would be 10 times harder.

Yeah maybe not -every- server on Portal, just those tagged as Supported maybe..

It would be relatively easy to police it, if someone logs on a server and sees they have a livelike feature which isn't in the public DB then give them a 2 week warning and remove them from Supported after.

Sorry, I've just been pretty annoyed at this issue for a few years now.. the public DB still doesn't have artifacts, working ml abilities, latest dragon weapons, astral weapons and much more.. it doesn't even have the three dragons...

[Blue]

One of the issues for the ones who work on the public database is not getting the credits they deserve for their (hard) work.

It can be demotivating to put a lot of detail work into the DB that a new server just leeches, and gets praised by the community for all "their hard work" they put into the server, completely ignoring who really did all that. I experienced that to some extent when I worked on DOL. You put the energy into the project, others earn the credit while giving very less back, if at all. I think thats also a reason why Argo loves to stick to a server, so he gets the proper feedback for his work.

Theres no easy solution for this. In the past I think I discussed with Etaew a system where you can't just download the whole public DB without having "credit points" or something. So you have to help on the DB first, earn points and be allowed to download stuff in relation to what you contributed. For example, make one zone live like, download 10 others in return. Give and take. It shouldn't be onesided like it always was. The keypoint is to get people to help not just leech.

[Argo]

Geshi,

to put a thing straight right, loyalty is not a thing that someone gets for free, it has to be earned. Take for example Uthgard, i was GM there for only a short time and i left for own reasons, and still, Uthgard has my loyalty aswell. Why? Simple because during my time over there they were a bunch of awesome guys in many ways and they were good to me. Allthough the opinion about the Uthgard team is very controvers discussed along the comunity you wont ever hear a hatred word from me about them. I would even contribute to Uthgard, unfortunately they have their own system of adding things to a DB. My loyalty is quite dificult to explain, take freyad, Leodagan is one of the persons i loved to share my DB with. I had my reasons for it too. When i contributed my DB to the public DB i had reasons too. But what i wont ever do is to spread my db that i worked on for years continuously along a bunch of ungreatful leechers. The reasons for that are obvious I don't mind helping servers with things but i am not a worker for a comunity that doesn't know how to say thank you. And to make one thing clear, i do not solely care for Storm, in the first way i care for my own DB, for my own world of DAoC. I wanna restore it to the best of my abilities and whatever results from my work on my DB finds its way too those that i am loyal too for what reasons ever. With that having said,

kind regards

Argo

[Maze]

Contributing database assets is a major issue for me for a few reasons.

The work we've done on the server isn't 100% applicable to every server, it's tailored to a specific era of DAoC. It's not always live-like.

Some of it is guesstimated. I don't feel comfortable submitting work that I'm not sure about.

Where would I submit it to? Would I just share the whole database? Patch specific zones of mobs into the public database?

And lastly, there's just simple self interest. I'm glad to contribute to DoL as a whole, but the work to put into a DB is tremendous and there honestly should be assets that are kept to oneself simply so you can compete.


The issue here is there's just no actual specific project for building a working public database. There has to be a specific methodology/goal for it. There's got to be a standardized way to submit database additions.

If there's something like this set up, I'd be glad to help out with it, even if all the work isn't really applicable to my own project.


Oh, and what exactly is the deal with Storm? Is it just the official test server or do we actually want players on it? Why is it so dead? You'd think the official server would at least see some respectable activity on it...

[Leodagan]

For me some trouble about committing DOL DB change is that slight differences into "DOL Core"-Code make the data useless without implementing these change to the user Code base...

Recently I added the new Pulsing BT "Immunity" Mechanism to my Code (on live : Livings can only receive a new PBT each 3 seconds from apply time) this forced me to subclass Bladeturn SpellHandler and thus my Spell Database with now "Live-Like" PBT-skill can't be shared with Storm or other Shard (even if I share my Code update, it's a lot of work to merge with SVN trunk)

Another example is giving a summon spell to a Mob NPCTemplate, if the summoned Pet use Heal spell you will have a lot of errors on DOL SVN Core, I fixed this on my side, but with a quick-and-dirty untested patch, if I release part of my Mob DB and it crashes other peoples server that don't use this code fix (or any other kind of fix) it would be more work for them to find the issue than just building their own Mobs with available resources...

If Code-Merging with DOL Trunk need some DB Change to Storm/Storm D2/Public DB, then the patch is regressive, if DB Commit need Code changes to prevent errors or just to work as expected then it's useless for most "common" users...

If some Code change the Database schema (Table Definition) without providing migration scripts or datadump you're just breaking everyone shard, if you provide datadump with table fields that aren't implemented in current Trunk it's useless to a lot of people !

I don't want to spit on the quality of DOL Trunk or Public Database, but changing either and giving support/docs/migration tools is a lot of work, the database definition is not stable (it doesn't match BCNF normalization) and the code structure is a mess honestly (a lot of object-oriented logics are wrong and doesn't allow for easy patching)...

Even the recent changes proposed by Darwin for 1.114 DAOC Client needs a lot of code to be merged, and need database changes requiring both Coding skills and DB maintainer Skills.

Public assets are a starter "pack" it would be impossible for most shard owner to release all their works and merge them to some "live like" public DB, because most shard owner have custom code / custom database content / custom fixes, and a lot of this work would break current "Demo" Storm RvR Server... (which honestly I don't know who is really Administering it...)


Like Tola said, this isn't a professional project, with paid-employees, the provided code/tools/data aren't professional level and thus working with it is not an easy task for a community contributer, why would I put more effort into fixing my code issues with DOL-(Storm)-Core in Mind (that would need some effort into weird Database change/values) aside from just giving away my updates to who-ever is able to use them and merge them with their own work...

For my part it's been 8 month of almost non-stop Code Auditing/Code Editing/Database Updates/Object Tree fix/Game Rules Plotting and I still don't have an own-satisfying Live-Like shard... putting so much effort into my own project is already using a lot of my freetime, I can't put as much work in trying to release "ready-to-use" "Storm-Complient" "Cleaned Datadump" work assets

[Tolakram]

D2 should not stop any development. I can take it off the core if a change breaks it, all I need is a few days warning, I comment out a line in a batch file, and all is good. Just wanted to get that out of the way first. D2 is a pile of scripts on top of the core, and it gave me a way to test some of my core changes and have fun at the same time.

Ok,

Storm is different things to different people. I always saw it as a Pendragon like server. A few fervent users and sometimes a bigger crowd, but it was a testing ground for core modifications and I made it a play ground for a while testing my instance madness (all Dinberg's fault ) prior to launching D2. It's humming right along in my basement, some sawdust on the vents due to my wife's woodworking hobby ... I should really sweep it out from time to time.

I am willing, with Graveens ok, to grant anyone access to Storm who has shown a positive contribution to the project and a mature attitude, which would include any purple name IMO. I have shown frustration in the past when people have broken Storm, which is probably the exact wrong attitude to have, but it demonstrates the conflicting goals of Storm. On the one hand I want it running 24/7, but on the other hand I want it used to help further enhance DOL development.

One of the issues with Storm is that it runs a good amount of custom code, like most other servers, that tends to break when any significant change is made to the core. This requires someone with access to the scripts to make changes required to fix it. In times of rapid development this developer becomes the Storm slave, spending most of their time fixing scripts.

One take on the purpose of Storm might go like this:

Dev tests change on local test server.
Dev commits change to DOL, restarts Storm to grab new code, and verifies change is working correctly.
Rinse and Repeat.

The above idea conflicts directly with the goal that Storm attracts players and is stable. It also conflicts with the idea that Storm can have a decent population needed for testing these changes. A server that's unstable and constantly changing will never keep a large population, and without a large population Storm does not really function as a valuable test server.

So there we are.

I have been tempted multiple times to abandon the current Storm (not the users, just the extra features) and put up a core only server with minimal maintenance scripts, running it as bare as possible to replicate live. I doubt this would add to our population, but it might make it easier to maintain during development cycles since there are no scripts to break.

Now let's get to the ugly part about the DB. Remember one of the conflicting goals of the official server is to attract enough people to make it a valuable testing server. So many of us worked hard to make the DB as live like as possible, in order to attract more players. Lets say we share this DB, completely, with the community. Our fear, or at least my fear, is that someone grabs it and takes the population away from us. All that hard work with nothing left for us.

I don't think it matters much anymore because most of the servers I was worried about don't have the competence to last long anyway, and the rest of them I'd gladly share what's needed because I know they could take a simple Heidi export and use it without crying asking what SQL is.

Most of these decisions were not mine. I came along after most of the core development was completed and took over Storm when we lost our previous host. Storm has a lot of history and I'm not at liberty to be the decider as to what happens in the future. Most of the work put into Storm is not mine.

[Graveen]

Let me (once again.. ^^) sum the way we are dealing with databases.

4 years ago, we RETIRED from the DB management. On purpose. We exactly don't want to support something as boring as db stuff, and 'i-own-the-truth' threads where boring.
2 years ago we RELEASED Storm DB. Can we count the contributions it involved ? 0.

Notice, this was the trigger for no more releases on DOLSetup.exe. Who cares ?

There are no points in arguing over the community. There are no points in weakening the community in having strict rules for portal. This is something that should be accepted: the DB work is server-related. Storm shares it sometimes, but this is not something we would like to be systematic because my point of view is that the contributer must choose to share or not.

Notice, despite what you could think, some contributers did EXCELLENT job on public DB. Leodagan, Misimouse, got stuff that is far over any DBs.

[Etaew]

Speaking on public DBs, I'm tempted to start exporting DBs from DAoC DB if people want to help work on that as a central repository and management.

[Overdriven]

Without responding to every post on this thread, I'm going post specifically about the public database: This is an issue which has come up time and time again, with every new generation of DOL developer/DOL DBA/DOL Enthusiast and has caused clashes at various levels of seniority in DOL - Which makes the conversation even more fun for me - As I've seen a few of these myself.

A brief history lesson - If I've left your name out of here or anything major, apologies, summarising.

- joined DOL just before we moved to SQL, when we ran on XML and it was painful.
- Akira at the time was running the DB for DOL along with DH and a few others.
- (IIRC it happened about this time) XML was dropped for MySQL and the sky opened up and things were good.
- Akira left DOL (As far as I was aware) and I eventually took over, DH etc was still active.
- Various people joined (Tired and quite a few then regular commiters joined) who made the update process mental, we were roxx0ing the DOLness.
- It got quiet, there were various phases of activeness/in-activeness/Tired and I switching roles as we came and left
- It got quiet, years passed, people left, people left and the work that Akira, DH, Tired, Argo, myself, Geshi, DF, Viper, Phen, DW etc (Some of who are still active) became less and less.. Updated.
- We're here today.

The IDEA of the DOL Public Database was fantastic, absolutely fantastic and I hated it. An open source project which released databases (and Code, but who cares about code..) regularly with as up to date(live) as possible information was fantastic. Then something was raised, and raised again and raised again:

"What do I get out of it?"

Now, this irked me a fair bit - Probably because it was my team (at that point) who were doing a metric tonne of work along side the core guys (ISAT, Etaew, Biceps, Batlas n'co) and it was basically: New server -> uses our DB/Code -> Gets 500 players -> No one gives a shit that we did all the work and got no thanks for it.

That gets a little annoying. Releases become less and less common. People leave. No more updates. People come back. Cycle starts again. (And then a thread with all this appears and I post something longer than "LOL ETAEW IS SMELLY")

A project like DOL, as many people will know, is a PITA. Lots of servers come out which are buggy, aren't updated, don't deliver the experience we want and various bad things which make DOL look stupid - At the end of the day, nothing will change. There's never going to be the drive to have something 100% complete (Uthgard is a different story) where the public are involved.

If DOL had the same team (A mix of now and a few years ago) and had a fully fledged (HI ORIGINAL IDEA FOR STORM) core server, with a decent upkeep, a decent release cycle and kept up to date with live (With the help of other "core" servers) AND had a DB which only got released quarterly (ALONG WITH CODE) - We'd have probably done better.

A server with 500-600 active players, who got the most up to date experience, without having to worry about other little crappy servers stealing their population = Happy DOL staff = More work, better updates and we'd have probably been further ahead than we are now...

[Leodagan]

Speaking on public DBs, I'm tempted to start exporting DBs from DAoC DB if people want to help work on that as a central repository and management.

I'm pretty much interested in your work around Database, I really like your blog having some nice "views" of database skills/spells.

One of my most up to date source website when trying to rebuild live-like behavior.

[geshi]

So many of us worked hard to make the DB as live like as possible, in order to attract more players. Lets say we share this DB, completely, with the community. Our fear, or at least my fear, is that someone grabs it and takes the population away from us. All that hard work with nothing left for us.

Well yeah, but everyone I've spoken to (including myself) thinks the same way.. which results in no one sharing their stuff... it's a vicious circle sadly! I'd like to see Player count removed from Portal and replaced with the admin's ContributionCount or something, but I doubt it'll ever happen I've been quite vocal over the last few years on why displaying player count on Portal is an absolutely awful idea but this thread isn't about that

People have said it's difficult/time consuming releasing livelike stuff to the community.. is it? not from what I've found.. maybe complicated stuff but things like spell updates, item updates etc are pretty simple to gather & release.

[Etaew]

I think we tried an experiment once removing the count, I can't remember what happened as a result. Either way doesn't Uthgard control portal dev now?

[Maze]

I think restructuring Storm is a good idea. Who actually would be involved in making that decision?

Player counts are a very odd thing. Players go where players are, so you need players to get players. It's almost a catch 22 in our community.

Removing player counts does end up removing some relevant info from players. A way to promote servers that deserve it is an interesting thought but it would require a bit of a design change.

and Uthgard controls portal dev? When did that happen?

[Etaew]

and Uthgard controls portal dev? When did that happen?

Once I stopped working on it, those guys are awesome. Not sure if it is still in their hands though.