0.7.0 and DB Milestone 1

[John Adams]

Guys, we're targetting the weekend of Jan 30th to try and release 0.7.0 + DB.

I am going to try REAL hard to repop QC and OO (specifically, at least fix the damn signs), and Scatman and I will take a pass through every popped zone from newbies through level 10. If everything isn't 100% perfect, that's ok. We can patch content after the first batch is released.

Zcoretri, I have all the (new) spells parsed, and I think they are on Dev named properly (??) but you let me know. It's been so long, I might have reverted something unintentionally.

[John Adams]

This is going to sound like a hair-brained scheme, but Scatman eluded to it the other night on IRC and I gave it some thought, and (unless I misunderstood him), I agree... here's the idea.


[center]DB Release and Reset[/center][/b]
We toss the Tess DB out as-is, bundle up the current LIVE db and scripts, and just deliver it. The data is ALPHA because the server is still ALPHA, so it will not be perfect and folks need to accept that.

Then, on Dev, we wipe the DB completely and wait til after this SOE re-do of everything (yet again) in May.

I'd personally still like all our old zones populated (unless SOE is removing the ability to enter those zones anymore). Scatman also reminded us that we used our current 28 or so zones to basically build the functionality of the server, and he was planning to go back through _every zone_ anyway and re-tweak everything.


Maybe it's just best to nuke it and start over, one last time. Won't take us nearly as long this time, because most of the systems are in place. This is an option, not a requirement. So if you think it's a bad plan, say so. I won't get my butt hurt

Sometimes it's just refreshing to start with a clean slate. And QC/OO are definitely going to be a work of art this time around... to bad no one will ever see them.

[Zcoretri]

Then, on Dev, we wipe the DB completely and wait til after this SOE re-do of everything (yet again) in May.

WTF are they doing now? I will probably answer my own question shortly

EDIT: They are thinking of removing the 2 original starting cities? OMG!

Sometimes it's just refreshing to start with a clean slate. And QC/OO are definitely going to be a work of art this time around... to bad no one will ever see them.

I agree here

[elcapototal]

well...

first things first: I'm kinda dizzy from reading so many posts from so many different points in time that I'm no longer sure what we wanted to accomplish before releasing 0.7. My two cents on the matter: I think we should clear out the current bug list in bug tracker and then release. In these past few days the list has gone down alot. Personally I've been tackling about 5 bugs a day which means in about 3-4 weeks or so I should be able to finish them all (assuming that I'm the only one working on them, which is not true).

I'm not a big advocate of throwing away work without a good reason, but before I can make an educated statement about wether it is a good idea to throw out the entire DB and start over or not, I would like to understand what's involved in re-creating the content. Are we talking about re-importing all data from the logs and then re-creating all scripts again? if so, I think it is *not* a good idea to do so. I think we can fix what we have in less time than it takes to recreate it... but again, I'm not sure.

[John Adams]

I'll let Scatman explain why content is "iffy at best". Some of his greatest work is in the last few zones he did, because the server HAD functionality to support what he wanted to do. Nearly everything before The Ruins was done when the server merely spawned mobs you could hail. So no, the data is not going to be missed if we nuke it. The scripts themselves are probably fine for the most part, especially as you said, just fixing what's broken/enhanced since then.

For re-spawning data, that's not hard. I have every log parsed into a Raw database, and we simply move the data from one DB to another, this time thoughtfully (previously we just shot an entire zone over, and deleted what we didn't think we'd need). Now, our server supports spawn "parties", before it did not. This is why QC/OO are being re-spawned right now.

There are far more arguments for doing it than not. The greatest reason it would be a no-brainer (imo) is that we know what we're doing now, and have a server core that we do not have to stop content for 2 mos waiting for functionality just to finish the zone.


I can speak to releasing 0.7 - and I fully agree, there is no release til Zexis and I come to an agreement on what is acceptable. Bugs, definitely need to be squashed. You have indeed been doing a great job there, so I thank you for helping out again. Yes, there are a ton of posts from a year+ ago about what needs to be done, but there are specifically "Priorities" threads that are (or were) stickied at one point that lays out what we intended to do for this release.

The database was NEVER going to be released until Beta (0., but that was when Beta was an achievable goal in 2009. Clearly we missed that goal, and considering all my whining about a dead community, it's obvious we cannot wait til Beta to release our data. There are "Phases" for our data that are also clearly laid out in the Content forum, those too should be stickied, along with my many "Tasks" posts. Hopefully this points you directly at some goals rather than blow your mind with a ton of obsolete chatter


ElCapo, you have any questions at all about what we should be doing, simply ask anyone. Bugs is a great place to start though. That's what Zexis wants to tighten up first before any new features.

And one last thought. While I'd love a jumpin community, I am not really in a major hurry to release. I understand the problem, and know what we have to do to fix it. If it takes 6 more mos, I'll just suck it up and be happy that work is being done. It's when nothing is happening and we're not getting closer to releasing this bitch that I get my panties in a wad.

[ZexisStryfe]

I think this is a good idea as well. Once 0.7 is released we are no longer going to support pre Sentinel's Fate boxsets anyway (for the SVN) due to the massive changes between SF and TSO. Doing this will give the community something to do while we drive ahead, we just have to make it clear that if you patch to newer than 0.7 you will lose compatibility with older clients.

On this note, i was wondering something... EQEmu has what world version a server is running when you choose what server to play on. I think that might be a great addition to our login server is at all possible (I know this might require LE's intervention). This way people will know what versions they can access with the old client vs. the new.

Regarding bugs, first things first I want to get all the critical and high priority bugs resolved. Once that is done I want to start from the oldest bugs and work forward (stopping to resolve new criticals as they come up). I have a feeling most of the old bugs were resolved and just not closed. Once we get to a reasonable number of bugs we have a few systems that we want to do some work in and then we can release. Personally, if we could do this at about the time Halas is released and the starting zones go away, that would be perfect.

[Scatman]

There are many reasons why it would be easier just to start over:

1) When we first started I was doing only content. I had never seen LUA before so my scripting for Queen's Colony, Outpost, and a few other zones are atrocious.

2) I added a lot of generic scripts after we discovered you could include other scripts and use their functions. For example, all guards say the same thing. I wrote a script to determine the race and gender of a qeynos guard and say to appropriate thing in 1 function. Before, we would have to check the race/gender of a guard, find the 4 or 5 voice overs for that race/gender combo and hard code them. Plus, as we've progressed, we've added more race/gender voice overs so some over have 3, some have 4, it's really a mess. With these generic scripts, we can add them to 1 script and ALL scripts that include them are affected.

3) As John mentioned, 3/4's of our content was done before the server really had an capability to support all the quests we wanted to do. Because of this, I coded in a lot of ugly work arounds. Literally, almost every single quest in Queen's Colony, Outpost, and most of the noobie cities have work arounds. I started going back and fixing these but it's really a pain.

4) Our Item IDs have changed. Because of this, almost every quest was broken. There are still a lot that are broken because of this.

5) Documentation. I started documenting IDs in scripts because of the situation above. So if a quest involved any item ID, I'd document what item it was. Most scripts do not have this and I'd like to see every script have it.

Each of these reasons isn't a good reason why we should start from scratch but rather it's the collection of reasons. Going back looking through every script, remembering what's going on, figuring out what's broken can be a real pain in the ass. The script's coding styles are all completely different because I was still trying to figure out the best way to handle scripts and it's just really ugly. I was literally going to go back and rewrite almost every quest script and spawn script because of this. Most people that try to learn off these scripts are clueless because of how complicated the scripts were at first.

[elcapototal]

ok, these sound like convincing reasons. Let's nuke it!

[John Adams]

Woot. Ok, after we're done with 10,000 other things, here's what you'll see coming from my house:

content-db.jpg

[Scatman]

From lighting a match and farting?

[ilythor]

Looks like I missed a fair bit, glad to see that poll came in handy! Also, from a CM's Point of View, it looks like a lot more writing for me!!! Yes! Since this is a pretty major development, I think some sort of massive public expose is in order. I won't post anything now, just saying it would be unfair if we kept this under wraps!

We can do this people.

[John Adams]

Not to stifle your creativity, Ily, but no one has a database now so they are not missing anything if we choose to blow away our current "learning curve" content to replace it with the actual Beta release a year+ from now.

What we can do (eventually, not now) is lay out our plan - once we know what it is - including the release of our current Alpha content, the release of our 0.7 code, and our plans to re-vamp everything after that is released and in the hands of admins to play with. The only reason I am giving in to the original plan is because everyone tells me the only reason our community is failing is that there is no data.

I'm going to prove that even with data, no one is going to help us anyway

[deadlyasp]

Most people that try to learn off these scripts are clueless because of how complicated the scripts were at first.

Amen! -.-

[John Adams]

Thanks for bumping this, DA. It just reminded me, if we're nuking our DB, there is no sense continuing to fix bugs in 0.6.5, since I have thrown the DB over the wall already.

LUA Script fixes we should still do, because when it comes time to nuke the DB, I will be preserving existing spawn_id's like I did with QC/OO and hoping to respawn the new content using old spawn_id's (so we do not have to edit every script).

Lastly, if we can come to an agreement on what the C++ Code needs for 0.7 to make Release, we can also start working from our new Public SVN - as the Private and SourceForge one will be going away.

Thoughts? Questions? The sooner we do all the above, the sooner we can feel productive again.