When will updates stop breaking mods?

[severedsolo]

29 minutes ago, Fraktal said:

Hey, I'd be happy if the modders would just stop putting MiniAVC into the CKAN releases of their stuff.

There is no different release for CKAN. CKAN just indexes the release when we push something out and grabs it from the same source.

Having said that, my personal preference would be for MiniAVC not to be bundled at all (and I don't bundle it with my mods). I provide a version file, and if you want that functionality you've probably already got AVC installed anyway. That's down to personal preference though.

[Curveball Anders]

1 hour ago, Fraktal said:

Hey, I'd be happy if the modders would just stop putting MiniAVC into the CKAN releases of their stuff. Putting it in the standalone release is fine and makes sense, but CKAN already does the version checking and won't even let you install an incompatible release in the first place unless you go out of your way to allow it, so not only is MiniAVC redundant in the CKAN release, but 99% of the time a pointless annoyance as well due to the mod working just fine despite the constant demands to downgrade my KSP version.

I just erase all miniAVC.dll files

[Superfluous J]

1 hour ago, Fraktal said:

CKAN already does the version checking and won't even let you install an incompatible release in the first place unless you go out of your way to allow it, so not only is MiniAVC redundant in the CKAN release, but 99% of the time a pointless annoyance as well due to the mod working just fine despite the constant demands to downgrade my KSP version.

The purpose of MiniAVC is not to help you decide if the mod is okay to run or not. It's so you won't bug the modder with bug reports on wrong versions.

I also eviscerate every instance of it I find on my install. Erasing's too good for it. But I understand why it's there and if it keeps a modder from going insane with dumb bug reports so be it.

[linuxgurugamer]

1 hour ago, Fraktal said:

Hey, I'd be happy if the modders would just stop putting MiniAVC into the CKAN releases of their stuf

You do realize, of course, the there is no special "CKAN" release?

CKAN only catalogs releases, I'm  not aware of ANY mod which has a separate release for CKAN.

If the MiniAVC bothers you so much, you have two choices, other than manually deleting the files:

1. Install the full KSP-AVC, which deactivates all the MiniAVCs.

2.  Install the mod ZeroMiniAVC, which will remove all the MiniAVC DLLs

[Tyko]

7 hours ago, Fraktal said:

Hey, I'd be happy if the modders would just stop putting MiniAVC into the CKAN releases of their stuff. Putting it in the standalone release is fine and makes sense, but CKAN already does the version checking and won't even let you install an incompatible release in the first place unless you go out of your way to allow it, so not only is MiniAVC redundant in the CKAN release, but 99% of the time a pointless annoyance as well due to the mod working just fine despite the constant demands to downgrade my KSP version.

There's a mod called ZerominiAVC that takes care of that.

[Lisias]

22 hours ago, Tyko said:

There's a mod called ZerominiAVC that takes care of that.

And there's the UNIX way:

find . -name MiniAVC.dll -delete

 

I made a bash script that I run every time I install/update something. On the list of things it does, there's the command above.

(and I removing that from everything I distribute too. It must had made sense at the time, but nowadays…)

[Kerbart]

Where AVC gets really annoying is that a mod generates a warning that it is INCOMPATIBLE with your current version. And the mod author responds happily with “I tested it, it works fine, no worries. I’ll recompile when I have the time.”

 I’m not complaining about the mod author not dropped everything to recompile. It’s a hobby and they don’t owe me anything.

What is annoying is that there’s no “don’t show this again” checkbox. Perhaps it’s helpful for the bug reports, but “the boy who cried wolf” is resulting in me deleting .version files and that cannot be the intention either.

[MaverickSawyer]

To answer the question posed in the OP: Never. 

[Superfluous J]

1 hour ago, Lisias said:

And there's the UNIX way:

find . -name MiniAVC.dll -delete

 

Or the Windows way:

del /s miniavc.dll

...though I have find as well (gnu ftw) so I could use that if I wanted

Edited November 20, 2018 by 5thHorseman

[releansol]

Never. This is the circle of life!

[Hodo]

I fixed the updates breaking the mods I run by not updating...  Still running 1.45

[Curveball Anders]

9 minutes ago, Hodo said:

I fixed the updates breaking the mods I run by not updating...  Still running 1.45

I fixed it by always install any new version in a new directory, playing around a little waiting until my wanted mods are updated.

Continuing to play my 'real' save in the 'stable' version.

Then swapping when the new becomes 'stable'.

 

[panarchist]

On 11/12/2018 at 12:37 PM, Overland said:

Ive been around since 0.25..

(snip)

Its the updates that break mods everytime.

If you've been around since 0.25, you've likely already seen that mod-breaking changes are a LOT less frequent. As others have said more eloquently than I can, it's impossible to prevent mods from breaking - especially in the case of mods like Kerbalism which do end runs around a lot of KSP's code and functionality. Since Squad started hiring several mod developers, the changes made have been a lot more respectful of mods in general, and I have a lot of mods under 1.3.x and 1.4.x which still run just fine under 1.5.1 without any changes. Honestly, since 1.4.2, there have been 5 hotfixes/releases that did NOT break the majority of the mods out there - the major exceptions being the ones co-opting graphics tasks, shaders, and other things that KSP normally handles natively, and those mods which are purposely version-locked by their creators. (which I am in favor of) I'd like to see better support for modders, and always will - but I think it's important to recognize the vast amount of progress towards that which Squad has made in the last year. Awesome work, folks!

[MaximumThrust]

I also play since 2014 and around the 0.25 (before the ISRU - my best reference to remember), and really think the 1.4.x and 1.5.x updates shouldn't have been made, stopping in the versions 1.2.2 or 1.3.1 would be better. The last meaningful update IMO was the Commnet. Mods now are far better than the stock game and the "new" parts. The need for updates uses a lot of time, energy and motivation from the modders, that could be producing new content instead. We had a boom of new things around the 1.3 era, that are kinda dead since the 1.4.0.

I personally use a heavily modded 1.3.1 install that won't be updated, and don't have much hope that KSP will get any better than this for me. A lot of modders still supports it, but some don't, and I can't get every mod I want anymore.

Since was announced that the game will be updated every 3 months, would be great that the community chose (by democratic vote) one old version to support and offer backward compatibility for it, I would be fine even if it's not the one I'm using right now.

[Lisias]

1 hour ago, MaximumThrust said:

Since was announced that the game will be updated every 3 months, would be great that the community chose (by democratic vote) one old version to support and offer backward compatibility for it, I would be fine even if it's not the one I'm using right now.

And once such democratic consensus are reached, how you would force the modders to follow it?  

It's nonsense. For the add-ons that are Open Source, the solution is clear: fork it and update it yourself, or use the fork from someone else that did it. While every License is being respected, nobody will have problems.

Some add-ons are working flawlessly since 1.2.2 (I tested ir on 1.2.2, 1.3.1, 1.4.5 and 1.5.1) without modifications or even recompilations. For at least some add-ons, this will not be a hard work.

Edited November 22, 2018 by Lisias
uh.. bad grammars….

[sturmhauke]

Updates are not mandatory, nobody is forcing you to get the latest version. If you like 1.3.1 or whatever the best, just stick with that. There are plenty of mods designed to work with older versions.

[MaximumThrust]

1 hour ago, Lisias said:

And once such democratic consensus are reached, how you would force the modders to follow it?  

It's nonsense. For the add-ons that are Open Source, the solution is clear: fork it and update it yourself, or use the fork from someone else that did it. While every License is being respected, nobody will have problems.

Some add-ons are working flawlessly since 1.2.2 (I tested ir on 1.2.2, 1.3.1, 1.4.5 and 1.5.1) without modifications or even recompilations. For at least some add-ons, this will not be a hard work.

Force?!

In Skyrim (a game I mod a lot, with a great community) AFAIK everyone uses the Unofficial Patch, an agreement so everyone have the same base game to make everything simple and compatible. Here we have similar things, like the Community Resource Pack and the Community Tech Tree. Or people are forced to use it and I'm not aware? 

About you suggestion that I should update the mods myself, I'll totally pass. If you are willing to do something similar, me and a lot of people would love to have the Tacking Station Evolved in the 1.3.1. 

[MaximumThrust]

2 hours ago, sturmhauke said:

Updates are not mandatory, nobody is forcing you to get the latest version. If you like 1.3.1 or whatever the best, just stick with that. There are plenty of mods designed to work with older versions.

I'll assume that answer was for me, if not, please disregard.

I don't know why are you talking about forcing people to update or that I should do what I moments ago said I'm doing. There are new great mods and parts that not work with older versions anymore, or require a lot of work and time. My suggestion is about making the experience of modding KSP more simple, pleasing and reasonable - not only for me.

[Lisias]

42 minutes ago, MaximumThrust said:

[…] Here we have similar things, like the Community Resource Pack and the Community Tech Tree. Or people are forced to use it and I'm not aware? 

It was the logic conclusion, as you proposed a 'democratic vote" by all users, without considering if the developers would agree with such decision. And even if only developers could vote, how you would convince the ones that voted "no" to comply? It's how life works.

CRP and CTT were things that developers agreed to use, as it would be benefcial for them. Users got beneficed too, but that decision was took individually: "should I use this or is better to add my own?". And, guess what… Some add-ons choose not to use them. Some are "compatible" in the sense they doesn't stomp on each other's toes, but one or two I had to adapt to coexist to CRP but without using it.

Things doesn't works exactly as you think.

[Superfluous J]

44 minutes ago, MaximumThrust said:

About you suggestion that I should update the mods myself, I'll totally pass. If you are willing to do something similar, me and a lot of people would love to have the Tacking Station Evolved in the 1.3.1. 

You misunderstood a statement of fact as a request for work. If you want a mod and nobody's willing to make it, you have 2 options: Make it yourself or go without. Presumably you don't want to go without.

[MaximumThrust]

18 minutes ago, Lisias said:

It was the logic conclusion, as you proposed a 'democratic vote" by all users, without considering if the developers would agree with such decision. And even if only developers could vote, how you would convince the ones that voted "no" to comply? It's how life works.

CRP and CTT were things that developers agreed to use, as it would be benefcial for them. Users got beneficed too, but that decision was took individually: "should I use this or is better to add my own?". And, guess what… Some add-ons choose not to use them. Some are "compatible" in the sense they doesn't stomp on each other's toes, but one or two I had to adapt to coexist to CRP but without using it.

Things doesn't works exactly as you think.

Would be great too if the mod developers take this decision like the community resource pack and tech tree, I would love 

[Snark]

On 11/22/2018 at 8:09 AM, MaximumThrust said:

Since was announced that the game will be updated every 3 months, would be great that the community chose (by democratic vote) one old version to support and offer backward compatibility for it, I would be fine even if it's not the one I'm using right now.

Yeah, good luck getting all the modders to follow along with that...

Modding is time-consuming.  I can't speak for other modders, but just speaking for myself:  I've got my hands full just writing my own mods and keeping them up to date with the current version of KSP.  I will never, ever spend any time at all doing extra work to "back-port" my mods to older versions of KSP, for a variety of reasons:

• I never play "old KSP" myself; I always play the most recent version.  So spending extra time to backport doesn't give me, personally, any benefit at all.
• I'd guess that a lot more KSP players will play on the most-recent version than on any particular older version.  So spending my time backporting would benefit only a small fraction of the KSP community, whereas time spent working on mods for the current version will benefit the large majority of players.  So it makes more sense to focus on the current version.
• Time is finite.  I only have just so much time to spend in implementing and updating mods.  An hour I spend backporting would be an hour I don't spend creating a new mod, or adding features to existing mods, or, well, playing KSP.  So of all the places I could put my time, backporting is the worst available option, from my perspective.

Other modders will come to their own conclusions, of course.  Certainly I expect that some will backport.  But I certainly won't, and I expect most modders won't, either.

Also, bear in mind that "democratic vote" doesn't really work when different people want different things-- and especially when not everybody can fill the same role.  For example, as a modder, I spend my time where I personally have the time and inclination to do so, no matter what anyone else thinks.  It wouldn't matter if there were literally a million people voting for me to backport my mods-- I wouldn't do it, because it doesn't make sense for me.  Other people don't get to vote what I do with my time. 

Also, speaking as a player, who likes to use other people's mods:  since I myself always play KSP on the most recent version, I'd prefer that my favorite modders spend their time working on the current KSP version rather than backports.  So a "community decision" to support some older KSP version would be doing players like me a disservice, even if the modders went along with it, because it would be taking time away from the mods we like.

I'm not saying it's a bad idea.  It's not.  It's just that I believe it's impractical, because I don't think that's how people work.

Backporting makes sense to some players.  But it doesn't make sense to lots of other players (I'm guessing the majority).  And it also doesn't make sense for a lot of modders (such as myself, and I'm guessing the majority).  Which probably explains why mod backports tend to be fairly rare.

[MaximumThrust]

15 hours ago, Snark said:

Yeah, good luck getting all the modders to follow along with that...

Modding is time-consuming.  I can't speak for other modders, but just speaking for myself:  I've got my hands full just writing my own mods and keeping them up to date with the current version of KSP.  I will never, ever spend any time at all doing extra work to "back-port" my mods to older versions of KSP, for a variety of reasons:

• I never play "old KSP" myself; I always play the most recent version.  So spending extra time to backport doesn't give me, personally, any benefit at all.
• I'd guess that a lot more KSP players will play on the most-recent version than on any particular older version.  So spending my time backporting would benefit only a small fraction of the KSP community, whereas time spent working on mods for the current version will benefit the large majority of players.  So it makes more sense to focus on the current version.
• Time is finite.  I only have just so much time to spend in implementing and updating mods.  An hour I spend backporting would be an hour I don't spend creating a new mod, or adding features to existing mods, or, well, playing KSP.  So of all the places I could put my time, backporting is the worst available option, from my perspective.

Other modders will come to their own conclusions, of course.  Certainly I expect that some will backport.  But I certainly won't, and I expect most modders won't, either.

Also, bear in mind that "democratic vote" doesn't really work when different people want different things-- and especially when not everybody can fill the same role.  For example, as a modder, I spend my time where I personally have the time and inclination to do so, no matter what anyone else thinks.  It wouldn't matter if there were literally a million people voting for me to backport my mods-- I wouldn't do it, because it doesn't make sense for me.  Other people don't get to vote what I do with my time. 

Also, speaking as a player, who likes to use other people's mods:  since I myself always play KSP on the most recent version, I'd prefer that my favorite modders spend their time working on the current KSP version rather than backports.  So a "community decision" to support some older KSP version would be doing players like me a disservice, even if the modders went along with it, because it would be taking time away from the mods we like.

I'm not saying it's a bad idea.  It's not.  It's just that I believe it's impractical, because I don't think that's how people work.

Backporting makes sense to some players.  But it doesn't make sense to lots of other players (I'm guessing the majority).  And it also doesn't make sense for a lot of modders (such as myself, and I'm guessing the majority).

I can't possible know how to running around to update to new versions (being they good or bad) can consume more time than keeping one old version, but ok. About the "human factor" I can't really tell or predict how people will act, but I have seem way bigger agreements in other games that makes modding a better experience to everyone.

[Snark]

1 hour ago, MaximumThrust said:

I can't possible know how to running around to update to new versions (being they good or bad) can consume more time than keeping one old version, but ok.

Correct. If you haven't modded,  then you can't possibly know. You'll just have to take the word of those of us who do.

Basically,  there are two possible scenarios. Either the mod works just fine across two different versions of KSP, or else it gets broken in one of them.

In the case where it works just fine in both (i.e. where the KSP update didn't break the mod)-- which, in my experience, is the overwhelming majority of the cases-- then great! Everyone wins.   The players win, because everyone on both versions of KSP can use the mod. And the modder wins, too, because zero work is required.

However, in the case where the KSP update did break the mod... well then, it's a whole different ball game, at least from the modder's perspective. To keep it working in two different places, the modder needs to maintain two different versions of the mod.  It's essentially the same as maintaining two different mods. Double the work. Double the testing.  Bearing in mind that the modder's precious free time is scarce,  and it's not as though they're getting paid for this.

Also, there's another aspect that makes it even worse, which may not be obvious to someone who has never been in the position of having to support a mod: the testing to support two versions of a mod for two versions of KSP can actually be substantially more than double the effort. Here's why:  When I write a mod, I use it myself in my own gameplay. (Of course! I mean, if I don't even want it for myself, why would I have taken the substantial trouble to create it in the first place?)  This means that I get a lot of testing "for free" just by virtue of playing KSP a lot with my mod-- if something's wrong, there's a good chance it'll turn up just in the course of normal gameplay. And it's fun, because a lot of it is just playing KSP the way I like to,  which I enjoy.

On the other hand... if I were also maintaining a version of the mod for some other version of KSP (i.e. one that I don't actually ever play, myself)... then I don't get the benefit of any such "free" testing. The only way to verify that the alternate version is working properly is to do straight up software testing-- grinding through endless, boring,  un-fun scenario variations on software that I myself have no use for or interest in. It's an incredible chore, seriously tedious and unpleasant. (If you don't believe me, ask anyone who has ever worked as a game tester. It's one of those jobs that's a whole lot harder and less fun than it sounds. I could never do that job myself.  I'd go nuts.) And doing all that work would provide zero benefit to me, because it's for a KSP version that I myself never play.  So why on earth would I do it?

1 hour ago, MaximumThrust said:

About the "human factor" I can't really tell or predict how people will act, but I have seem way bigger agreements in other games that makes modding a better experience to everyone.

Shrug. Different environments are different, so I'm having trouble seeing how that's relevant here.

A different game would have a different player culture,  different modder culture, different programming platform with different requirements,  different cadence of updates from the game developer, different size of player base, different modding interface with different churn rate, different ease of modding due to different degrees of interface documentation, and of course different mechanics in the game itself, to name just a few factors that immediately spring to mind, just off the top of my head.

In other words, just because a thing is doable in one place does not necessarily mean that it's so easy in another.  It's utterly apples and oranges.

[Kernel Kraken]

A: never.