Author
Message This forum is locked: you cannot post, reply to, or edit topics. This topic is locked: you cannot edit posts or make replies.
Swiftoak
Developer Emeritus
02 Feb 2005

Location: Kah-nah-duh

So I've been thinking about a few things lately. First off, everyone seems so busy IRL (which is cool, no skin off you), but as a result, things have slowed down. Queries have gone unanswered, showcases have been forgotten sometmes, and there's a large queue of interiors in reviewing. I'm proposing we take a few steps to remedy this, referring back to the Management Proposal's reviewing structure. While we probably need to rethink our entire system of task management (I.E not the claims browser/forums in their current state), there are things I think we can do to make things more efficient, in the short term.

1. Claim/Showcase Reviewing Speed: Yeah people can be impatient, we do take too long to review shit/respond to queries. This is a problem I've remedied by simply reviewing and responding more (even though I hate it). Or getting more reviewers. I think the problem is people dont like reviewing because it's so tedious. I agree. A potnetial solution is to not get so anal about whether a potion is off by 0.1 degree off a table. Also getting overly nitpicky in the showcase can draw people away. I'm saying we should proably tolerate minor placement errors because you most likely need to zoom into the CS to see them. I'd much prefer people having the skill to create something tasteful and cool, rather than a person's ability to place potions on uneven de_p, or ability to sink rocks_01-04 100% into the ground. Look for the major things, leave the minor anal things for betatesters and playtesters. Chances are if most people don't notice, it's not a big deal in the first place. Also, atm we can roll the Plan Compliance and Technical reviwers into one, because our current active recviwers are all either Lead or Senior Devs in charge of a section, so they would probably be able to check or both types of errors in a single review.

2. Claim Approval Speed: This is problematic because right now, we only have a single Final Reviewer for each kind of Data. Haplo for Exteriors/NPCs, TF for Interiors, Arvis for quests. We also need to approve ints based on priority (IE ones needed for current sections in production), and not (last to first). Basically wat I'm proposing is that we first appoint more Final Reviwers for each type of data, and second, trust their word. So this would mean we wouldnt need to give Final Reviwers mod privilidges, but if they say "Approved" any Lead Dev can simply hit the buttons. For Interiors, I'm suggesting Rats and Jule for potential candidates. I'm suggesting rot and arvis for exterior finalers, since they have a good eye for spotting errors.

3. Developer Promotion: This is problematic as well, in a sense that right now, we have showcases that must wait for a forum administrator to push buttons, or a final reviwer to (currently, the same two people hold both positions, and are super busy IRL). I think the solution here is similar to what I am suggesting for my point above. More final reviwers obviously, and the ability for Lead Developers to promote to any group. Or maybe even remove the need of a final reviewer to authorize a promotion. Basically any two recommendations from lead/senior developers should = a promotion.

In any case, have all forum permissions from now on deviate from the usergroups, so that the only part of promotion that is held to forum administrators is the actual rank change (which is just a simple user title change). This allows us to promote people faster, without giving everyone they keys to the forum admin panel. (Though maybe it might be a good idea to have a third person on top of Haplo and TF to manage forum upkeep regardless). I would suggest Adanorcil if that's the case, because he not only is the website manager, he is competent and trustworthy.

_________________
"Idleness and lack of occupation tend - nay are dragged - towards evil."
-Hippocrates
Post Fri Feb 21, 2014 6:46 am Send private message             Reply with quote                   up  
Gnomey
Lead Developer
19 May 2006

Location: In your garden.

I'll admit I'm hesitant in respect to the idea of loosening TR's reviewing standards. On the one hand I think TR's high standards are one of the project's selling points, (admittedly also a weak point and often a bottle-neck), on the other hand I find that, after the first few interiors, modders seem to get the hang of it and avoid most errors. I'm worried that loosening the standards will lead to modders not improving much and, as a result, sloppier work. For my own part, working with TR has allowed me to improve on various skills. If the project's standards and demands weren't as high as they are, I don't think I would have improved as quickly or as much.

That being said, I'll be the first to admit that a large part of my hesitation is probably balking at new ideas, and vanilla Morrowind shows quite well that a product doesn't need to be technically perfect to provide an excellent experience. I think the question of lowering TR's technical standards is certainly worth debating.

As for the other points you make, I can flat-out agree with them.
Post Fri Feb 21, 2014 3:39 pm Send private message             Reply with quote                   up  
rot
Lead Developer
21 Oct 2012



It's true that too much nitpick CAN draw good recruits away, but the ramifications extend way beyond showcases. The point of all this anality is both making sure new modders can be to some extent trusted with not screwing up too much in their claims, **and** making the reviewers' job less unbearable. And for actual (non-showcase) interiors, the bottleneck is not int-makers but reviewers too. So

Swiftoak Woodwarrior wrote:

This is a problem I've remedied by simply reviewing and responding more


This is still the only real solution (and yes I hate it too)


Quote:
for exterior finalers


Appreciate the thought but I'm not a terribly good choice, if only for the hard reason that my machine can't handle many TR exteriors :<

Still on more efficient finaling, no less important than all those is Data, but I don't know whether anyone else but Haplo would even know how to handle it (the Data error threads having never been maintained probably doesn't help on that)
Post Fri Feb 21, 2014 4:18 pm Send private message             Reply with quote                   up  
Rats
Lead Developer
03 Jul 2012



Being super busy IRL so don't have time for lenghty answers. Agreeing with Swifty that there's a problemz with the speed stuff gets reviewed/responded to. Right now we have three new showcase guys with recommendations for promotion (not that they've waited for long, though). Agreeing that a more relaxed attitude towards minor/irrelevant bleeders/floaters might not be an utter disaster quality-wise. Agreeing with rot that the only real solution is to review more.
_________________
stars are out tonight.
Post Sat Feb 22, 2014 8:42 am Send private message       Send e-mail       Reply with quote                   up  
arvisrend
Lead Developer
04 Oct 2010

Location: substitutional world

I fear I have very little time to respond to this now; accordingly it will be short.

Yes, the reviewing speed has plummeted lately and I'm to blame for this in the quests section. I will catch up eventually, but this isn't my last term before the PhD and it's not likely to get better afterwards. The unfortunate fact that rot and I are the only two people reviewing quest claims means we plainly and simply won't get Almalexia done, no matter whether or not we integrate Mournhold. It's a pity :/

1. I agree that we are looking too much at angles and not enough at the big picture. It seems an unfortunate tradition to review things in the CS and not go ingame for a single moment; as a consequence, we tend to have doorways too small to pass and caspering half- and quarter-columns (just to name two things hard to spot in the CS but obvious ingame) but hell are our potions rotated well. I hope this will get better with the flat de_p furniture we have now and will hopefully use on the western half of MW. (I still don't see a reason to avoid bleeding-into as opposed to bleeding-through, by the way. But I'm not much of an inter...)

2. Yes, we need more finalers. rot is competent both at dialogue and at scripting and should be a finaler for quests (although in this section nothing is really formal, so he can tell me anytime to wave through a claim he has reviewed and I'll just do it). Rats is as good a judge of dialogue as we can have. Adanorcil can also be fully trusted on dialogue. IMHO we should encourage more people (including outsiders) to playtest our quests, even at the costs of exposing our TR_Mainland.esp. About exterior finaling, I might be a good floater-spotter (yes, my secret is going ingame, as before) but I'm not sure if you want to trust my judgment on style, and besides I don't know some of the technical guidelines. [EDIT: removed a sentence or two.] Checklists for reviewing would probably help a lot: reviewing is not supposed to be a creative job, and it absolutely shouldn't be pulling creative energy from our modders unless the reviewer decides to add more content (which is fine, as long as people don't feel forced to).

3. +1.

4. I don't think handling Data is so difficult that noone but Haplo can do it. But it's a job that is best done by one person only (you can't really work distributed on a .bsa file), and Haplo, while slow, has never been out of reach for a significant timeframe, which makes him one of the more reasonable candidates for this job. Do we have more such candidates?

Oops, this didn't end up being short after all...


Last edited by arvisrend on Sat May 31, 2014 2:17 pm; edited 1 time in total
Post Sun Feb 23, 2014 6:04 pm Send private message             Reply with quote                   up  
rot
Lead Developer
21 Oct 2012



arvisrend wrote:
Data is so difficult that noone but Haplo can do it. But it's a job that is best done by one person only (you can't really work distributed on a .bsa file), and Haplo, while slow, has never been out of reach for a significant timeframe, which makes him one of the more reasonable candidates for this job. Do we have more such candidates?


Ya that's what I was thinking (Hapslow :p still being best on that for those reasons),
the problems with data are rather
- the tons of things that still need fixing,
- all the objects in need of review and inclusion
- all the objects that are so near review they really just need someone to push them / make it known there's interest in them
- all the objects that need to be designed/done

so basically there's a need for an objects reviewer and no active/willing/good candidate (in immediate vicinity)
Post Sun Feb 23, 2014 6:46 pm Send private message             Reply with quote                   up  
Display posts from previous:   
This forum is locked: you cannot post, reply to, or edit topics. This topic is locked: you cannot edit posts or make replies.

 
The content of this site is © by the Tamriel Rebuilt community. Morrowind, its expansions, and its content is © Bethesda Softworks.
Forums powered by phpBB © 2001, 2005 phpBB Group