Oolite Bulletins

Muchas gracias, Admiral. <pulls numbers out of sombrero> I reckon this could save me at least a couple of minutes per system (much more in dodgy/busy systems). On a forty-jump courier run (docking at thirty systems), that amounts to a whole hour - which could be very handy, bonus-wise! As for assassins - meep meep!

Let's attack the question from a different angle.

Some OXPs are popular. My guess is that for instance BGS, Additional Planets (+extra stations), Random Hits, Anarchies, Commies, YAH have been installed by at least 50% of the players.

I say those OXPs should be part of the official distribution. Either in the main installer or as a separate, all-in-one package of "standard" OXPs (only if their authors permit, of course). As those OXPs will probably be tweaked (to achieve consistency or balance for instance within the standard set), they will also have to be community-maintained.

The idea is to use popularity to determine what players like and want. Not things that people think would be nice to have (and that may or may not work), but things they already have adopted. Acknowledge them and use them as a basis for the next step forward.

Doing so will help OXP makers with determining where to put their efforts. I'm not a prolific OXP maker myself, but it seems to me that one reason why one releases OXPs to the public rather than them keeping for oneself is (aside from as a general thanks to the community) that one thinks it helps with improving the game. But what is "the game" exactly? The plethora of existing OXPs makes it a blurry picture. OXPs can be a double-aged sword in this regard. Having a standard set of OXP lets makers have a better idea of what is the actual "user experience" for most players, and how to improve it.

We can use the poll feature on this board in order to measure the popularity of OXPs. It's not perfect because not all players are lurking here, but that's a start. If nothing else a significant part of the regular posters are OXP authors, and as they are those who invest efforts in improving/expanding Oolite, it's only fair if they are somewhat over-represented in the poll. In a first round, one lets people freely nominate the OXPs they think are the most popular. Then we use this data to make a multi-choice poll, asking "which of those OXPs are installed on your system".

Astrobe wrote: ↑Sat Jan 21, 2017 11:28 am

The idea is to use popularity to determine what players people like and want.

Well, current trends suggest that this approach doesn't work well for most people where politics are concerned (Brexit/Trump), and I don't think it will work well for a game either.
Players should be left to decide for themselves what they want free from any peer-pressure or misguided 'marketing' attempts. OXPs are additions to the game, and should not be bundled up with it, even supposedly 'popular' ones. None of the so-called 'popular' OXPs you list feature in my game. I actually don't like most of them and would not add them as they stand, except Extra Planets which I sometimes add in for a bit of variety.
Just because a lot of people like something that is no reason to give it any sort of priority or precedence - there are probably as many players (like me) who don't want them.

The problem is, by making expansion packs "official" we'd be implying that they should be used. Of the six you list, all of them except one (BGS) have significant effects on gameplay - mostly by providing extra stations.

Astrobe wrote:

The idea is to use popularity to determine what players like and want.

I'm not sure that polls on this forum would be representative of the wider player base - I don't think we're "normal" users! If there was a way for users to leave feedback/reviews on the wiki, that might help - but I don't think it's feasible. The best solution might be to make sure that each OXP page has a prominent "How this OXP will change your game" section, separate from any background fiction.

Smivs wrote: ↑Sat Jan 21, 2017 12:23 pm

Astrobe wrote: ↑Sat Jan 21, 2017 11:28 am

The idea is to use popularity to determine what players people like and want.

Well, current trends suggest that this approach doesn't work well for most people where politics are concerned (Brexit/Trump), and I don't think it will work well for a game either.
Players should be left to decide for themselves what they want free from any peer-pressure or misguided 'marketing' attempts.

No, it is entirely different. It's not about voting for what they want next, it's about knowing what they have already chosen and adopted.

OXPs are additions to the game, and should not be bundled up with it, even supposedly 'popular' ones.

Suppose we run this poll and find out that 90% of the voters actually have for instance BGS installed. Should we really continue to consider it as an addition to the game? Then what would happen if it is suddenly taken down by its authors? Don't you think the game would suffer from this loss?

Oh, and I may be wrong but I think it already happened in the past. Didn't the current Cobra model or station textures begin as OXPs?

None of the so-called 'popular' OXPs you list feature in my game. I actually don't like most of them and would not add them as they stand, except Extra Planets which I sometimes add in for a bit of variety.

I don't have Random Hits either. Now what?

It's not about you or me. It's about the common preferences. Knowing them helps with moving the game in the right direction. 'Right' being, again, about satisfying as many people as possible. Moving in the right direction is necessary in order to attract more players. More players are necessary because a small fraction of the player base becomes contributors. We need contributors because Oolite needs to keep up with hardware and software improvements. If you don't do that, the spiral reverses: the game stagnates, it attracts too few new players so you don't get contributors, so your game looks more and more outdated. Many open source games have died from this.

And look, it happens that Oolite is that awesome extensible game now because someone contributed scriptability. You have nearly 500 OXPs to like or dislike because the game managed to attract many players that later became contributors. So it is worth giving some consideration about what the majority thinks.

Just because a lot of people like something that is no reason to give it any sort of priority or precedence - there are probably as many players (like me) who don't want them.

There are reasons to give them a particular status; I've given the main ones. If you think none of them are valid, then please explain why.

And if this suggestion is accepted, you'll most certainly have the choice of disabling the OXPs you don't want, or simply don't download the 'standard OXP' bundle if we go for that option.

Disembodied wrote:

The problem is, by making expansion packs "official" we'd be implying that they should be used. Of the six you list, all of them except one (BGS) have significant effects on gameplay - mostly by providing extra stations.

I don't understand what your objection is. If this list turns out to be right - that all of them have more than 50% (or 60% or 80%) adoption rate - doesn't that make it unambiguously obvious that people really do want those gameplay changes? What's wrong with acknowledging facts and take them in consideration when one asks oneself where the game should go?

I'm not sure that polls on this forum would be representative of the wider player base - I don't think we're "normal" users

I've addressed this point in my original message: it's fair that forum users are over-represented because they contribute to the game in many ways. However, this means that newbies are under-represented. I think we are smart enough to take this into consideration.

Astrobe wrote:

Disembodied wrote:

The problem is, by making expansion packs "official" we'd be implying that they should be used. Of the six you list, all of them except one (BGS) have significant effects on gameplay - mostly by providing extra stations.

I don't understand what your objection is. If this list turns out to be right - that all of them have more than 50% (or 60% or 80%) adoption rate - doesn't that make it unambiguously obvious that people really do want those gameplay changes? What's wrong with acknowledging facts and take them in consideration when one asks oneself where the game should go?

If a large majority of users already have these OXPs, then I think that would be evidence that there's no need to bundle them in with the core game - pretty much everyone who wants them is getting them. I think it's also important to recognise the difference between installing a game-changing OXP when you are an experienced pilot with a solid ship under you, and installing those same game-changing OXPs as a brand-new Jameson who has yet to develop a clear idea of the basics.

My objection, ultimately, is about balance. I think the core game needs to be balanced - which, by and large, it is. By bundling game-changing OXPs with the core game, we're implicitly telling people to install them, and to unbalance the game from the get-go.

As it stands, we don't have any facts about OXP use, nor do we have any reliable means of getting them. But I think we can see that popular OXPs suggest that e.g. players like having more career options, and much of the functionality of the UPS OXP made it into the core game in the form of courier/parcel contracts. A similar development, allowing players to take on specific bounty/assassination contracts, would be a good idea - but that's a very different thing from saying that the Random Hits OXP should be bundled in with the game. Also, players like having different stations to visit, in different types of system - but integrating that with the core game would require a lot of adjustments to the trading model to avoid creating money pumps.

I disagree with your basic point, in that I don't agree with the concept of officially sanctioned OXPs. For any reason let alone one based on an inherently flawed 'popularity contest' idea.
This was discussed when the Manager was launched, in terms of quality control when some felt that OXPs should be 'vetted' before being allowed into the Manager. One of the reasons given against this was that 'officially recognising' some OXPs by default diminishes all the others. OXPs that 'come with the game' would be seen as 'safe' or 'superior' and others may be treated with more suspicion, particularly by new players. All OXPs are equal and none should have any sort of special status.
No, far from promoting broader appeal, such a scheme would most likely be very negative for the whole OXP scene. At the moment folks can choose the OXPs they want, and avoid those they don't. This can be done easily and mostly from within the game. I really don't see a problem here.

I'm on the side of "the official game has to be balanced" in every way to play.

That some oxps aren't balanced for everything is not important as they are chosen by a player for a certain balance that player likes to use in the kind of games he plays.
(For example, I don't snipe. So it doesn't matter to me if sniping is off-balance.)
But if we delivered some "official oxps", they would have to be balanced in every way of playing.

That said, dropping the idea of "official oxp", would it be much work to have some statistics?
I imagine it would need the oxp manager, when downloading the most recent list of available oxps, to upload its "used oxps list" with a unique id.
That way, the oxps makers could work on what's really used :-p

Astrobe wrote:

I say those OXPs should be part of the official distribution.

They are. You must enable them with the expansion manager.

Astrobe wrote:

'Right' being, again, about satisfying as many people as possible. Moving in the right direction is necessary in order to attract more players. More players are necessary because a small fraction of the player base becomes contributors. We need contributors because Oolite needs to keep up with hardware and software improvements. If you don't do that, the spiral reverses: the game stagnates, it attracts too few new players so you don't get contributors, so your game looks more and more outdated. Many open source games have died from this.

Seriously though, I feel the developers made real progress in this respect with the manager, I see no reason to saddle them with oxp maintenance when they have have made it so easy to access those oxps yourself.
Seems to me like the best way to satisfy everybody is to allow them to make their own choice.
I don't see Oolite downward spiraling anytime in the near future anyway, not with nearly 500 mods available in the manger and the awesome team currently maintaining it for us. Sure maybe the backgrounds could use a little refreshment like the ship textures, but if you're going to add Random Hits, Commies and Anarchies, what about Dictators, the Feudal States and RRS, what about Station Adds? I think if you could really accurately poll every user you might find up to 50 or more of these most popular oxps, maybe even 100. How many do you include?

Disembodied wrote:

much of the functionality of the UPS OXP made it into the core game in the form of courier/parcel contracts.

I can't remove the parcel/contract function from the game if I don't like it, unless I can remove it successfully from the core. I can ignore it but I still see a dozen assassins sitting on the witchpoint sometimes. I'm not complaining or asking for the removal of this function, I am just making the point that once something is added to the core it may not be easy to remove or change, at least for an average user with few coding or modding skills.

Day wrote:

I imagine it would need the oxp manager, when downloading the most recent list of available oxps, to upload its "used oxps list" with a unique id.
That way, the oxps makers could work on what's really used

Some of the wiki pages have a counter that lets you know how many times an oxp has been downloaded, I think there was even talk of linking that info to the manager so a user could get an idea of what is popular and I'd support something like that for sure. I am all for allowing users to make informed choices. I don't think this was something that the devs thought might be very easy to accomplish, though, otherwise I think it would already be done after it was brought up.

Disembodied wrote:

players like having different stations to visit

I have been struggling with scale in this area, if I make a station that's very complex it ends up so large that it is clearly visible from the witchpoint yet if I shrink it down to a more reasonable size the dock becomes such a massive feature it ends up looking somewhat silly.

Smivs wrote: ↑Sat Jan 21, 2017 3:33 pm

For any reason let alone one based on an inherently flawed 'popularity contest' idea.

No, it's not a popularity contest. I thought it was clear it wasn't but maybe I expressed myself poorly. It's not a popularity contest. It's not an award-winning event. It's not a way to shut down dissident opinions. It's not a way to force things on people.

This was discussed when the Manager was launched, in terms of quality control when some felt that OXPs should be 'vetted' before being allowed into the Manager. One of the reasons given against this was that 'officially recognising' some OXPs by default diminishes all the others. OXPs that 'come with the game' would be seen as 'safe' or 'superior' and others may be treated with more suspicion, particularly by new players.

An OXP which is used by, say, 80% of players already has a special status. In other software with extensions/plugins, they are sometimes called "must have". Even 50% would be quite an achievement, judging from the wide variety of tastes one can see on this board.

Some of the Griff/Solo textures made it in the core game with the 1.80 release. Quoting the announcement:
"Core ship models and textures replaced with significantly improved ones by Griff/CaptSolo"

Now that's special treatment. But few people complained about it because... Those textures were popular anyway.

No, far from promoting broader appeal, such a scheme would most likely be very negative for the whole OXP scene.

There are many programs with extension/plugins systems. Browsers for instance. They even have a full-featured review and grading system.
I am a former Flight Simulator player. It is extensible too, but has no integrated review system. But you have dozens of third party sites that let you sort extensions by popularity, downloads, etc.

This is practically the same as having some extensions installed by default - at least for the persons you seem to consider, that would just download the most popular stuff without thinking.

Seriously though, I feel the developers made real progress in this respect with the manager,

Well, that's the problem with doing a good job: people ask for more.

I see no reason to saddle them with oxp maintenance when they have made it so easy to access those oxps yourself.

Yeah I feel a bad about it, but usually popular OXPs have their own maintainers. Also popular OXPs are usually quite stable, because they have been thoroughly tested and their features "approved". Maintenance is only really required when some breaking change happen in the API (see market definitions for instance).

Seems to me like the best way to satisfy everybody is to allow them to make their own choice.

If a large majority of users already have these OXPs, then I think that would be evidence that there's no need to bundle them in with the core game - pretty much everyone who wants them is getting them.

Yes, and the manager does it beautifully. Now that this problem is solved, let's move to the next problem; which is in a nutshell: "where should we go now?".

Take for instance an issue discussed right here previously: slow lanes. People have different ideas about it: masslock changes, or something outlandish like injectors that deplete your shields instead of your fuel tank for instance. Having that choice is certainly a good thing. But if the masslock change becomes very popular, then it may influence the way ship makers determine the weight of their ships. As a result you have informal dependencies between OXPs; balance that is determined with the implicit assumption that you use variable masslock. A corollary is that the injectors thing will become less relevant (if it ever was) because it doesn't depend on weight so it makes things way too easy in this new OXP ecosystem (for instance).

The gist of it is: sometimes you can't maintain "competing" alternatives. There has to be a winner, some choices have to be made.

I say that the wise thing to do is to acknowledge the choices made by the players and encourage (not force) everyone to go that way. Because we have talented but few contributors, we have to rationalize/optimize our efforts.

I think it's also important to recognise the difference between installing a game-changing OXP when you are an experienced pilot with a solid ship under you, and installing those same game-changing OXPs as a brand-new Jameson who has yet to develop a clear idea of the basics.

Good point. I believe OXPs that affect in significant ways the balance are naturally less popular, though. I agree that newbies, "training wheel" and "Hardcore" OXPs are a special cases.

By bundling game-changing OXPs with the core game, we're implicitly telling people to install them,

I see it as a good thing if those OXPs are selected based on popularity. It means that the majority of players like them, so a new player will probably like them too. Nothing evil there.

We can even decide to make for instance "Starting choices" a standard OXP to deal with the special case of newbies, even though maybe it doesn't reach the 50%. By doing so we are improving their first experience and have better chances to retain one more player that has a tiny chance to become a contributor later. The alternative is that the newbie is likely not to bother with going to this board and silently ditch the game because "it's stupidly hard and it's outdated graphically anyway".

and to unbalance the game from the get-go

Oolite is not player centric, which basically means that it doesn't care about being fair and balanced. The real deal is progression: how long you have to play before you can take a 1 on 1 fight instead of fleeing from everything; how long it takes to buy the equipment to "iron ass" your Cobra. etc. One could even dig further consider progression through equipment upgrade versus progression through ship acquisition. But I digress.

I think if you could really accurately poll every user you might find up to 50 or more of these most popular oxps, maybe even 100. How many do you include?

I don't know... Let's see the figures first?

But I'm more worried of the opposite: people here have very different ideas, so it might be that only a few OXPs reach 50%. If none, then I was completely wrong right from the beginning.

Some of the wiki pages have a counter that lets you know how many times an oxp has been downloaded

I don't trust those figures: they don't take into account un-installations, re-installations, and there are secondary download sources in some cases. Same goes for the Manager.

Some of the Griff/Solo textures made it in the core game with the 1.80 release. Now that's special treatment.
But few people complained about it because... Those textures were popular anyway.

Not a good example, as the plan was always to get Griff's textures into the core game.

Astrobe wrote:

let's move to the next problem; which is in a nutshell: "where should we go now?".

Take for instance an issue discussed right here previously: slow lanes. People have different ideas about it: masslock changes, or something outlandish like injectors that deplete your shields instead of your fuel tank for instance. Having that choice is certainly a good thing. But if the masslock change becomes very popular, then it may influence the way ship makers determine the weight of their ships. As a result you have informal dependencies between OXPs; balance that is determined with the implicit assumption that you use variable masslock. A corollary is that the injectors thing will become less relevant (if it ever was) because it doesn't depend on weight so it makes things way too easy in this new OXP ecosystem (for instance).

The gist of it is: sometimes you can't maintain "competing" alternatives. There has to be a winner, some choices have to be made.

I say that the wise thing to do is to acknowledge the choices made by the players and encourage (not force) everyone to go that way. Because we have talented but few contributors, we have to rationalize/optimize our efforts.

Personally, I think these choices - for any putative Oolite 2.0 - have to be left to the devs. They are, after all, the ones who would have to put them into practice within the core game. I don't think such major changes to gameplay as "how/if masslock works in the game" are best achieved via an expansion pack. We can make suggestions, and the devs do listen to, and ask for, opinions on the forum, but we can't tell them what to do, democratically or otherwise.

OXPs can work here to test concepts, to see how things might play out and what knock-on effects they might have to the in-game economy, NPC AIs, etc., but properly integrating these kinds of changes into the game would involve lots of labour within the core. If something drawn from or inspired by an OXP genuinely makes the whole game better, for players at all levels, then I'm all for building it into the next version of the game - but ultimately that choice isn't up to me, because other people are going to have to do the work.

Astrobe wrote:

It's not a way to force things on people

It can be though, depending on how its implemented. I would be stuck with those new ship textures without the great contrubutions of other oxp authors, now Griff/CaptSolo's textures are awesome and I am happy about their inclusion in the game but I think there is more to consider than popularity. Just because something is popular doesn't necessarily mean its better, or even good for that matter.

Astrobe wrote:

I am a former Flight Simulator player. It is extensible too, but has no integrated review system. But you have dozens of third party sites that let you sort extensions by popularity, downloads, etc.

Who maintains these third party sites volunteers, or the makers of the Flight Simulator?

Astrobe wrote:

This is practically the same as having some extensions installed by default - at least for the persons you seem to consider, that would just download the most popular stuff without thinking.

I think one of the things this board has going for it is that people are encouraged to think for themselves.

Astrobe wrote:

We can even decide to make for instance "Starting choices" a standard OXP to deal with the special case of newbies, even though maybe it doesn't reach the 50%

I would support the addition of start choices, for one thing it doesn't change the core start by default, the first choice is the same as the core.

Astrobe wrote:

I don't know... Let's see the figures first?

But I'm rather worried about the opposite: people here have very different ideas, so it might be that only a few OXPs reach 50%. If none, then I was completely wrong right from the beginning.

I am not opposed to that, I would recommend allowing users to post their oxp/z list from the latest log and count duplicate entries, list the top 10 or so with a link to a full list. Users should only be able to make one post but should be able to edit it to overwrite their old list if they choose. I am not sure if such a thread could be configured this way on this board but if so maybe it could be stickied under Expansion Packs or something. It might be nice if the oxp titles in the list linked to their wiki page if one exists.
I am not sure numbers gathered from such a thread would be any more accurate than the numbers on the wiki pages but at least it could give users something of an idea about what is popular as the numbers on the wiki pages do. There is no reason that you couldn't use the info to create a popular oxp set and put it in the manager. Generally I don't care for bundled oxps but one created this way may have merit.
If the devs use this information to help decide what should happen with Oolite thats their choice to make, but I still don't think we should have a popular oxp set that is installed by default.

While it's true that I can be a little torn between designing an oxp to suit my game or one to suit a more general release I think that such a price is mostly worth paying for if the benefit is flexibility.

In terms of suitability for inclusion, I think there's quite a number of distinctions to be made amongst game add-ons or improvements that may be relevant here.

1. Optional : adds to the game but only as a choice - if you don't 'flick the switch' then nothing happens. e.g. purchasable ships : it couldn't be done in the original elite but if you don't buy them then this addition has no effect.

2. Cosmetic : only changes the game's appearance with no additional gameplay or interactions added. e.g. adding Griff's models to the core game along with Frontier style contracts (pre assassins).

3. Avoidable : limited to a particular area or aspect of the game. Similar to #1 except that it requires active avoidance in order for the game to be unaffected. e.g. commies, deep space pirates.

4. Distractionary : Similar to #3 although cannot be avoided altogether due to unpredictability/frequency of distribution. e.g. additional planets / hoopy casinos, they exist as visible entities and navigation considerations even if you don't interact with them directly.

5. Additional : something new is added to the game to be encountered where it may. e.g. fuel injectors can be employed by both the player and non-player ships, oxp ships would generally fit into this category.

6. Active : has near constant or largely unavoidable game influence. e.g. ship configuration, many of the market altering oxps

The changes aegidian made to the game compared to the original elite fall mostly into the optional category with one in the additional category (which I seem to recall him saying he should have made oxp). As far as I'm aware, the only oxp that has made it into the core would be classed as cosmetic (if using the categories above).

If more oxps were to be core then I would favour them to exist within either the first category or (if easily changed via oxp) the second. The higher the number of category, the less likely they are to be suitable for inclusion in my opinion. This would exclude many fine oxps but then if oolite remains a highly moddable game then an oxp (or one inspired by it) can likely remain in some form for as long as game development continues.

OXPs can work here to test concepts, to see how things might play out and what knock-on effects they might have to the in-game economy, NPC AIs, etc., but properly integrating these kinds of changes into the game would involve lots of labour within the core. If something drawn from or inspired by an OXP genuinely makes the whole game better, for players at all levels, then I'm all for building it into the next version of the game - but ultimately that choice isn't up to me, because other people are going to have to do the work.

I never said that the dev team should make changes to the core game. Nor did I say they have to do the community's every bidding.

My suggestion was about including popular, working and proven OXPs in the Oolite distribution. That's all. It could actually just be a collection-type OXP if one wants to minimize the workload for the devs.