Oolite Bulletins

Neelix wrote:

Wildeblood wrote:

Here's something to think about: suppose everyone with a user login at oolite.org had a hard limit on the number of OXZs they could add to the system. Say 10. No exceptions for anyone. Would that encourage a focus on quality, rather than quantity, of OXZs available through the manager? (No, let's say 5.)

Seriously? You expect that to make any difference at to the quality of what's offered at all?

Seriously, as in is there any prospect whatsoever of the powers that be doing such a thing? Of course not. Is it something you should think about? YES. If you can't nominate your five best OXPs, why can't you? ("They're all equally brilliant; everything I do is brilliant," is not an acceptable answer.)

Neelix wrote:

If anything that would just discourage me from sharing my work...
I make my OXPs primarily for myself, and while I enjoy being able to share them with others and get feedback I think having that kind of limit imposed would just cast a very off-putting shadow over the whole thing...

Who suggested any limit on anyone's ability to share anything? We're discussing a particular method of sharing. A method which is designed for its users to avoid any interaction with OXP authors, this forum or the wiki. A method which strongly implies by its nature that the content available through it will "just work". But some of it doesn't work well, and some doesn't work at all.

The in-game OXZ downloader is effectively making OXPs part of the core game. When it doesn't work well, that no longer reflects badly on the individual OXP author - who is now reduced to anonymity - it instead reflects badly on Oolite itself. I don't want the player's experience to be, "I tried that Oolite game, and it looked okay, so I added some of the expansion packs. Then it started to act weird, so I deleted it."

Wildeblood wrote:

Neelix wrote:

Wildeblood wrote:

Here's something to think about: suppose everyone with a user login at oolite.org had a hard limit on the number of OXZs they could add to the system. Say 10. No exceptions for anyone. Would that encourage a focus on quality, rather than quantity, of OXZs available through the manager? (No, let's say 5.)

Seriously? You expect that to make any difference at to the quality of what's offered at all?

Seriously, as in is there any prospect whatsoever of the powers that be doing such a thing? Of course not. Is it something you should think about? YES. If you can't nominate your five best OXPs, why can't you? ("They're all equally brilliant; everything I do is brilliant," is not an acceptable answer.)

Neelix wrote:

If anything that would just discourage me from sharing my work...
I make my OXPs primarily for myself, and while I enjoy being able to share them with others and get feedback I think having that kind of limit imposed would just cast a very off-putting shadow over the whole thing...

Who suggested any limit on anyone's ability to share anything? We're discussing a particular method of sharing. A method which is designed for its users to avoid any interaction with OXP authors, this forum or the wiki. A method which strongly implies by its nature that the content available through it will "just work". But some of it doesn't work well, and some doesn't work at all.

The in-game OXZ downloader is effectively making OXPs part of the core game. When it doesn't work well, that no longer reflects badly on the individual OXP author - who is now reduced to anonymity - it instead reflects badly on Oolite itself. I don't want the player's experience to be, "I tried that Oolite game, and it looked okay, so I added some of the expansion packs. Then it started to act weird, so I deleted it."

While I don't believe your proposed mechanism would lead to higher quality OXZs (but rather irate authors), the point you make merits more discussion.

Crowd feedback, a.k.a. "rate this OXP", could be a way for players to take the safe route and only install 5-star OXZs and could help authors to understand better what problems players face (if they provide comments with their rating). Of course we then have a third place to register with and read and so on… But if it is player-centric (vs. author-/enthusiast-centric like the BB) then it might still have its place.

Wildeblood wrote:

Who suggested any limit on anyone's ability to share anything? We're discussing a particular method of sharing. A method which is designed for its users to avoid any interaction with OXP authors, this forum or the wiki. A method which strongly implies by its nature that the content available through it will "just work". But some of it doesn't work well, and some doesn't work at all.

I see you already have 10 OXZs.. so when you get your next brilliant idea for a new expansion pack, which one of the already available ones will you sacrifice? And what about the one after that? And the one after that? ad infinitum..

Before long, you'll have maxed out your allotment with nothing but totally brilliant OXZs, and be begging for the limit to increased, so you don't have to choose which one to ditch.

Wildeblood wrote:

Neelix wrote:

Wildeblood wrote:

Here's something to think about: suppose everyone with a user login at oolite.org had a hard limit on the number of OXZs they could add to the system. Say 10. No exceptions for anyone. Would that encourage a focus on quality, rather than quantity, of OXZs available through the manager? (No, let's say 5.)

Seriously? You expect that to make any difference at to the quality of what's offered at all?

Seriously, as in is there any prospect whatsoever of the powers that be doing such a thing? Of course not. Is it something you should think about? YES. If you can't nominate your five best OXPs, why can't you? ("They're all equally brilliant; everything I do is brilliant," is not an acceptable answer.)

No... I wouldn't try to quantify that, it's not really my place to decide which small selection of my published OXPs should be available to the players. Who am I to decide that for them? Rather it would just mean that I'd stop publishing any through that method at all.
If I can't upload all my OXZs to the manager there's no point bothering with the manager at all, and I would just either publish only through the wiki and forums. I'd also be tempted to just pick up my bucket and spade and leave the beach... I'm here because it's fun, if I feel excluded in some way it reduces the fun level and if I'm not having fun there are other things I can spend my time on.

The problem is that setting such a limit does not address the quality issue at all.

You can't fight a fire by calling an ambulance for the person next door who doesn't need one.

- Neelix

As I said above, the problem is the over-hasty rush to publish OXZs coupled with perhaps a lack of care on the part of some authors. Limiting the number of OXZs per author is not going to 'cure' this and is in so many respects a bad idea.
I have already made some value judgements regarding my original OXP portfolio - I don't need a silly limit to focus my thoughts - and have actively decided that some will not be converted to OXZ. I have done this because the point made above about the managed OXZs being almost part of the core game is valid and I will not therefore include anything in my managed portfolio which which does not fit in with the core game, or is 'excessive' in some way. I have also made sure (admittedly after a bit of tweaking after the event, and while recognising that some aspects could still be 'modernised') that they are all fully up to date with all the new features and methods, and that they have all been tested as working properly, at least within the core game, and that any obvious or predictable clashes with other OXPs has been allowed for. There is clearly a caveat here that it is impossible to test with all OXPs and combinations of them.
The whole OXZ/manager thing is new, and there are bound to be a few teething problems with it. We must accept this and live with it. But this does not excuse authors from their responsibilities, and this is where the bulk of the problems seem to be coming from. To be blunt, too many shoddy, poorly written and un-tested OXZs are being uploaded. And this does need to stop if this game is to retain its credibility. Wildeblood is absolutely right in pointing out that bad OXZs are going to adversely affect the way Oolite is seen.
If we really do care about and love this game, this does need to be sorted out as soon as possible.

Sounds like first thing we should do is REMOVE oxz files .

Maybe we could start by removing all oxzs where the owner of the manifest is not the author or maintainer ?

Next step could be to post a list of all oxzs that haven't been updated since say 2 months before 1.80 was released.
Then we can comment on whether those oxz can stay or not.

Firstly, to pick up on KW's point, I would welcome a purge of OXZs to remove any 'dodgy' ones. However this should be done by the 'responsible adult', and not by the Devs or any sort of commitee - I would view this as censorship - so it might be time for a bit of soul-searching and honesty from some of us.

cim and others have expressed surprise at the number of OXZs already available, and again this brings me back to my comments about the 'rush'. There is no need to rush OXZs out. OXPs are still a part of the game, and there is no real reason to update an OXP to OXZ format except to add it to the managed AddOns. Far better to leave the OXP available while it is fully overhauled, updated and tested for v1.80, at which point it can be converted to OXZ format.

This process does take time of course - I am well aware of the hundreds of hours I have personally spent sorting out the OXPs of mine that have been converted. Except for the simplest OXP, there is a hell of a lot of work involved. Plus a lot of new methods and features to take into account. TBH this side caught me out a bit at first and some of my early OXZs were frankly not up to speed. I sorted them out quickly, but in retrospect I should have spent more time early on familiarising myself with the v1.80 features etc before I started.

So there might be a need for some sort of guidance as to the sort of things to consider when converting OXPs to OXZ.
If it is a ship OXP, have the roles been adjusted to suit v1.80 and the new role set? Has the shipdata been updated to the new 'template' system?
If things are added to systems, has the new populator been used?
If it uses ships (as say part of a mission) are the ships used still available, or if not has the OXP been adapted to use the new core ships or has it had the necessary models, textures etc added to make it work, and is the shipdata correct and up to date.
Is the scripting up to date? For example has missionScreenOpportunity been used instead of missionScreen?
Lots to consider and lots of work for some people. But really, this degree of consideration and effort is the only thing that will sort-out this mess in the long run.

Here's a thought: on the Oolite.org download page are offered the standard version and OXP developer version, with extra features like TAF enabled. What if the download manager in the standard version didn't list OXZs with a version number below 1.0? Then I could upload Half-Baked 0.3 without feeling guilty.

LOL @ half-baked 0.3, even though I think my OrangED HUD and Engine Trails have numbers below 1.0 only because I don't consider them complete

Also, I'd remove some OXZs from the list - namely, Lave Academy (we have the tutorial in 1.80) and Equipment Aide (duplicates functionality which exists in 1.80).

About the ships, I'm working on a comms OXZ with phkb and we've noticed that many OXP/Z ships don't have JS AI, therefore they won't react properly to comms. Another thing to consider?

Zireael wrote:

Also, I'd remove some OXZs from the list - namely, Lave Academy (we have the tutorial in 1.80) and Equipment Aide (duplicates functionality which exists in 1.80).

Really?

Can I urge again, to the current devs, aside from any extra beta-testing of OXZs and any limits or purges of what's available, what is required most is a module inside Oolite that checks an OXP hierarchy or OXZ zip file for internal consistency.

Such a module would need to:

0. check the OXPZ requested to be checked has a manifest
1. disregard any files, OXPs, OXZs that aren't explicitly stated as dependencies.
2. ensure that any files, OXPs, OXZz that are dependencies have also been consistency checked (recursive checking here)
3. check that all files referenced by the OXPZ and its scripts exist and have the right filenames (errors with inconsistent cases an example here.)
4. check that the scripts within the OXPZ being tested are sane and 'use strict' (I know about the stopping problem: I know a script cannot be completely tested, but simple errors can still be caught here.)
5. produce a list of any errors or problems and suggest corrections, possibly produce a manifest for the OXPZ, including maybe (in the case of an OXPZ that passes all checks) a checksum or hash that can be used in step 2. in future.

Once an OXP hierarchy passes consistency checking, a lot of the common problems should have been eliminated, and the hierarchy can be confidently zipped and released as an OXZ knowing it will not easily break the user's experience.

Zireael wrote:

About the ships, I'm working on a comms OXZ with phkb and we've noticed that many OXP/Z ships don't have JS AI, therefore they won't react properly to comms. Another thing to consider?

I'm a partner-in-crime on a number of those ships.. the ship designer doesn't do scripting, and I'm not up to speed on that aspect.. send me a PM explaining what you need, and I'll see if I can get the stable I'm responsible for up to spec.

Zireael wrote:

LOL @ half-baked 0.3, even though I think my OrangED HUD and Engine Trails have numbers below 1.0 only because I don't consider them complete

Perhaps WIPs simply should not be released as OXZs at all. Release them as OXPs to get the testing and feedback, then when you are fully happy with them, release them as OXZs.

What I'm reading here is a failure to communicate and to a certain extent talking at cross purposes.

We have 2 different types of ease here:

1) ease of use.

2) Ease of design and alteration.

I have to say the oxz system seems fine and dandy when it comes to option 1. So in effect we are talking about the issues found in relation to making option 2 workable. Now I realize that some people have worked with and had some reasonable success with this system. BUT they are failing to communicate it to people who don't understand. This is not unexpected, teaching is not the job of a developer.

BUT I feel the documentation on HOW to oxp and now oxz has ALWAYS been lack luster. Now I don't know how much we go in for "baptisms of fire" or "rights of passage" around here any more. But to my mind if an 8 year old can't pick up the wiki and be building something workable within a week then we have failed to supply the element of oolite that sets it apart from MANY other games and certainly all the clones.

To change the world.

To filter through the board looking for one, off hand comment, is not good enough. This game desperately (and to my mind has done for several years) needed walk through's. The more the better. Problem is, it's boring as hell writing a wiki that goes through steps you memorized years ago in fine detail. Writing trouble shooting wikis even worse!

But someone is going to have to, and probably sooner rather than later. Sorry I really didn't want to end this post on a downer, but the "better" oolite gets without proper documentation of process and explanation, the more you cut out the young talent who will take this game to the next level. We need to make it fun and make it easy, to play, to adapt AND to build.

ClymAngus wrote:

BUT I feel the documentation on HOW to oxp and now oxz has ALWAYS been lack luster. Now I don't know how much we go in for "baptisms of fire" or "rights of passage" around here any more. But to my mind if an 8 year old can't pick up the wiki and be building something workable within a week then we have failed to supply the element of oolite that sets it apart from MANY other games and certainly all the clones.

To change the world.

To filter through the board looking for one, off hand comment, is not good enough. This game desperately (and to my mind has done for several years) needed walk through's. The more the better. Problem is, it's boring as hell writing a wiki that goes through steps you memorized years ago in fine detail. Writing trouble shooting wikis even worse!

But someone is going to have to, and probably sooner rather than later. Sorry I really didn't want to end this post on a downer, but the "better" oolite gets without proper documentation of process and explanation, the more you cut out the young talent who will take this game to the next level. We need to make it fun and make it easy, to play, to adapt AND to build.

Lots of very good points in there..

While I'd say the situation is not quite as dire as the above makes it sound, I'd be the first to agree that there's considerable room for improvement. Heck, that was part of the motivation behind the guides I've made for joystick tuning and OXZ-ifying. It would be nice if some of the more experienced OXP producers could come up with something along the lines of what ClymAngus is proposing. Heck.. I'd be amongst the first wanting to make use of them myself.