The future of SafeBuckets on S

[TheRandomnatrix]

I noticed a couple of times where you guys said that it takes hours for your water requests to be filled, so I have a quick explanation/solution for your you. I don't log onto S except to check if a specific player is currently on that server. I also don't generally check modreqs even when I joined the because most of them seem to be grieving without fixing or land disputes and I don't deal with that. However, the modreqs of C and P are almost always empty. As placement requests are super quick, just pop on P for a moment and ask if a mod can do your water req. I'll let the other (more active) staff clarify on this point, but I'd be more than glad to switch servers (especially with the new lobby system) and finish your request.

This all coming from a guy who just instantly fills his own water/lava requests, so I can't comment on things surrounding that.

I also want to throw in my "voice of reason." Unless you're(generically speaking) making your flow reqs at 5AM out of peak hours, I find the idea of waiting hours for a flow request to be an exaggeration considering the bulk of most modreqs. And while I'm sure there are many kind moderators on PvE and Creative who enjoy filling out quick reqs to help people, I dislike the idea of pestering mods to do things for you. Making the modreq is the request for help. Asking mods to drop everything they're doing to come on another server strikes me as a bit selfish, aside from when spammers or griefing teams come on to the detriment of others and there are no mods to deal with them.

 

As for grinders and other large flows, it's general practice for staff to stick around and help the person make needed edits.

 

Another necessary point to make that nobody has bothered to say: If safebuckets is being removed it should be at the start of a rev, not the middle, lest builds relying on safebuckets be negatively affected.

 

I would like to see, if it's something the techs could feasibly implement, a command that allows moderators to check open modreqs on other servers. (A counterpart to /list-all and the server-hopping commands.) So they could see if there are waiting reqs open on another server and switch to complete them if they've been left unattended for awhile.

 

I also brought that idea up a little while ago in game chat and think that it would be an excellent feature, especially with the lobby system.

[SpiderMan]

Enters TheRandominatrix, the actual voice of reason.

 

Thanks for viewing it from a player perspective!

 

- gsand

;D

Edited January 7, 2014 by gsand

[Diznatch52]

Unless you're(generically speaking) making your flow reqs at 5AM out of peak hours, I find the idea of waiting hours for a flow request to be an exaggeration considering the bulk of most modreqs.

 

I disagree. Firstly, I would like to call attention to the fact that you are a player primarily on P and this is a discussion about S. While it may be true that there is not usually such a long wait for modreqs on P (and as someone who played on P for a decently long time I don't believe that this actually is the case. I've had to wait hours for assorted simple reqs on P.), the same cannot be said honestly about S. I've been on vacation for the past few weeks and have been playing on S during all sorts of hours. There have been long stretches of time at any given point in the day or night that no moderator has been on. S players have been saying for a long time that S needs more moderators. "the idea of waiting hours for a flow request" is most certainly not an exaggeration. Does one always have to wait an extended period of time before a modreq is filled? No. Does it happen often enough to be considered significant? Probably.

 

Also, there's the issue of having to make multiple modreqs for one build due to water not working quite the way one expected, or if one is making a build that requires one source be flowing before another, or honestly any instance where not having flowing water is a clear inconvenience. Furtheremore,

 

As for grinders and other large flows, it's general practice for staff to stick around and help the person make needed edits.

 

It's absurd to assume that players making large structures should idle around online waiting for the req to be done. When I am making a build that requires a req to continue, I don't just stand there waiting. I go do something else, generally something not on minecraft.

 

Another necessary point to make that nobody has bothered to say: If safebuckets is being removed it should be at the start of a rev, not the middle, lest builds relying on safebuckets be negatively affected.

 

Very much this.

 

EDIT: Wow such grammar

Edited January 7, 2014 by Diznatch52

[Denevien]

Firstly, I would like to call attention to the fact that you are a player primarily on P and this is a discussion about S.

I don't think this should stop anyone from commenting their opinion, considering there was much discussion about this on S in the first thread, which was about safebuckets on P.

[Diznatch52]

I don't think this should stop anyone from commenting their opinion, considering there was much discussion about this on S in the first thread, which was about safebuckets on P.

 

Right, but that sentence follows into the next one.

 

While it may be true that there is not usually such a long wait for modreqs on P (and as someone who played on P for a decently long time I don't believe that this actually is the case. I've had to wait hours for assorted simple reqs on P.), the same cannot be said honestly about S

 

My point was not that those who play primarily on P are not allowed opinions. It was that in this case, where TheRandomnatrix is in fact talking about S, the statement that was made was an incorrect one likely based off of his experiences on P.

[Denevien]

Ah, I see the intent now, my apologies.

[Trooprm32]

As someone who, just yesterday, had to fix grief that was a result of diverting a lava mote to a redstone circuit; it's easy to fix, BUT slightly time consuming. This was just once instance i saw this rev, and I've seen in on the previous one too. Most traps on S are easily griefable, and I think removing SB will make it easier.

 

My verdict; Removing SB will save some time for players, but increase potential workload for moderators. This is necessarily a horrible thing; just something to think of.

[barneygale]

trooprm, what if we ran roastnewt's plugin which disables fluid destruction of redstone? AFAIK there aren't any uses of that functionality since pistons were added.

[Trooprm32]

See that would be interesting. I wonder if a plugin like that could be customized to toggle what fluids are able to destroy... and maybe having regions work in a way for a user to request what it can and cant. I don't know limitations of this tech stuff' so it may be impossible.

[WickedCoolSteve]

trooprm, what if we ran roastnewt's plugin which disables fluid destruction of redstone? AFAIK there aren't any uses of that functionality since pistons were added.

 

Please forgive my lack of technical knowledge in this area. What's the impact on server load by such a plugin? IIRC the last time we tried something like this was with WG a couple years ago and it melted the server. How is this different / how have things changed since then?

[coolgamerovr90]

Please forgive my lack of technical knowledge in this area. What's the impact on server load by such a plugin? IIRC the last time we tried something like this was with WG a couple years ago and it melted the server. How is this different / how have things changed since then?

Even though we have a new box, the server's kind of already maxed out now with the lobby, and the lobby has the bare minimum needed for a user to connect, see a world and run a command. I doubt anything will be different, we'll have to run those server timings and see just like what we did with CH awhile ago

Edited January 8, 2014 by coolgamerovr90

[BeastBruiser]

Personally I've never seen that big of a problem with modreqing for fluids. Granted I've only been playing since S rev 15, but in that time I have built many grinders, auto farms, and tons of other builds that needed complex water flow.  I can never remember having to wait more than 18 hours at worst, and usually it was under 2 hours. definitely not uncommon to have them filled in a matter of minutes. I have played at all hours of the day (unemployment can really screw with my sleep schedule) and still there was usually a mod or admin who would hop on and catch the easy modreqs. If i got too impatient I was usually able to find water nearby that i could use to avoid a modreq altogether.

 

All that being said, since I was made a mod, I have loved being able to place water instantly. It does make the game-play more seamless. So from a player POV I would really like to see it place-able.

 

 

From a mod point of view, I don't mind the water flow modreqs--actually I enjoy them. It seems to me like the other staff does as well. If there is more than one staff playing when one pops up it is frequently a race to see who can claim it first. (damn you buchanman.) It's a fun interaction with the player base that I enjoy. While I don't usually stick around the whole time, if I notice the player is working on a grinder i'll usually tell em to feel free to msg me with any more so i can get on it ASAP.  (I would like to caution against suggesting that players /msg a mod to draw attention to a modreq as that sets a dangerous precedent of bugging mods to get modreq completed, it is a volunteer position and I would hate to see mods getting burned out)  

 

I would much rather spend 2 minutes completing a water flow than spend an eternity and a half trying to clean up a huge lava/cobble/water mess that a player was able to build just by alternating buckets of water and lava. (I've never had to clean that up, but Kitcatbar didn't seem to think too highly of me when she had to clean it up on a chaos server a few revs back... oops) A player only needs 6 iron for that, pretty easy to find in almost any surface cave.

I definitely think it would be worth a trial run for a week to see if it is abused especially with a game mod in place to prevent rail and redstone grief (would that mod destroy obsidian generators?) and i would agree that it shouldn't be tried mid-rev. I think the perks that the playerbase stands to gain is worth the added risk of griefing. BUT during that I would suggest that we keep a very close eye out and if it does start to become an issue, revert back to the current system.

[c45y]

Please forgive my lack of technical knowledge in this area. What's the impact on server load by such a plugin? IIRC the last time we tried something like this was with WG a couple years ago and it melted the server. How is this different / how have things changed since then?

If you simply cancel the flow event...medium?

 

If its coded by someone familiar with working with liquids...low

 

Basicly instead of just cancelling the event turn the water attempting to flow into the block into non-flowing water and as long as the blocks around it aren't frequently updated you should never have to touch it again. Keep in mind we already take effectivly the same actions using safebuckets... just the events that trigger stopping the flow of water are not buckets but water flowing into protected block space

[Synergetrick]

If its coded by someone familiar with working with liquids...

Hi.

 

I coded (and stole some old code from SafeBuckets for) this plugin which should fit the intended use-case. Relatively untested as I'm typing this at 1am, however it Worked4Me™. Has persistence in way of a standard database which stores ID / World Name / Hash. I found no reason to store x/y/z individually, however if someone can prove that it's necessary I'm happy to change this functionality.

 

I added in something that I'm unsure is present in SafeBuckets; the monitoring of the "EntityChangeBlockEvent" event, which is called, among other executors, when gravel falls onto a block.

 

The only caveat is this to the deprecation of type IDs (block material IDs), you have to use the internal names. As such, things like redstone repeaters are actually called "DIODE", and redstone itself is "REDSTONE_WIRE". There's very little I can do about that without coding in natural language processing.

 

Edit: I've been discussing this with md_5 just now and I'm going to be making a few edits to the hashing system. c45y also pointed out that it's a good idea to store x/y/z for manual correction if somehow the automation breaks. As such, if you're using this anywhere please realise that it'll break with the next few commits I make.

Edited January 13, 2014 by Synergetrick

[c45y]

Hi.

 

I coded (and stole some old code from SafeBuckets for) this plugin which should fit the intended use-case. Relatively untested as I'm typing this at 1am, however it Worked4Me™. Has persistence in way of a standard database which stores ID / World Name / Hash. I found no reason to store x/y/z individually, however if someone can prove that it's necessary I'm happy to change this functionality.

 

I added in something that I'm unsure is present in SafeBuckets; the monitoring of the "EntityChangeBlockEvent" event, which is called, among other executors, when gravel falls onto a block.

 

The only caveat is this to the deprecation of type IDs (block material IDs), you have to use the internal names. As such, things like redstone repeaters are actually called "DIODE", and redstone itself is "REDSTONE_WIRE". There's very little I can do about that without coding in natural language processing.

 

Plugin works logically, caches things so you're not hitting the database all the time. If you are going to actually implement something then this is a solid attempt at a good solution.

 

I would store x, y, z on the off chance someone one day needs to manually touch the database, but I'm just being pedantic now

[Synergetrick]

Plugin works logically, caches things so you're not hitting the database all the time. If you are going to actually implement something then this is a solid attempt at a good solution.

 

I would store x, y, z on the off chance someone one day needs to manually touch the database, but I'm just being pedantic now

 

I've updated this now. Removed the hash from the database because storing it is nonsensical if there's a chance that the values it depends on - x, y and z - can be modified.

[Four_Down]

Would it be possible for an S Admin to weigh in on this? I'd like to hear their opinion on this.

[SpiderMan]

Would it be possible for an S Admin to weigh in on this? I'd like to hear their opinion on this.

 

A week has elapsed, since there wasn't a definitive response from the actual server admins, there is no point in continuing this thread and wasting everyone's time for a feature that may or may not be implemented.

 

- gsand

;D

 

 

End of Thread, Waiting for Admin Response

[Mrloud15]

Actually it doesn't seem to be being discussed at all.

We have been discussing this and serval other things. You just don't see it because we do it in private, and I have been extremely busy this week so I have not had the time to reply to this. 

 

 I've tried to contact two of the S admins three times now to no avail, and I believe others have as well.

I can tell you that no one has tried to contact me about this, yourself included.

 

 

 

We would be fine using an alternative solution to SafeBuckets, but we are against trying it during a rev. We don't know what that alternative solution is yet, but we looking at our options, including ones you guys have said here.

[Synergetrick]

We have been discussing this and serval other things. You just don't see it because we do it in private,

 Why? This doesn't help anybody. I don't expect you to have the entire conversation in #redditmc but the least you could do is offer "we're looking into this" or something on the forums.

 

 

I can tell you that no one has tried to contact me about this, yourself included.

I asked after you specifically either in-game or in Mumble (I forget which), and redwall_hp. In redwall's case I asked another moderator to tag me when their conversation in Admin and Mod was over so I could have a quick word about it. That never happened.

 

 

We would be fine using an alternative solution to SafeBuckets, but we are against trying it during a rev. We don't know what that alternative solution is yet, but we looking at our options, including ones you guys have said here.

Moving to this or any other solution (including Glacier, but that is broken in multiple ways which I am not willing to fix anymore) mid-rev would not work. Consequently, bringing this into a new revision is the best (if not the only) way because it would break so many things and confuse players.

Edited January 17, 2014 by Synergetrick

[SpiderMan]

Hi Mrloud,

 

What is your specific opinion on the matter?

 

- gsand

;D

[redwall_hp]

Mrloud, cab, Draykhar and I have come to a consensus: we're definitely open to trying an alternative to SafeBuckets in the future. (Such as roastnewt's proposal, for example.) The caveat being that it should only be done at the start of a new revision, so we don't break things or confuse players.

 

Edit: The above mention of roastnewt's proposed solution is not an endorsement of a particular plugin. Merely an example. 

[SpiderMan]

Why not synergetricks proposal? it seems to be the least performance intensive solution whereas roasts would be called 20 times per second times the number of places water would flow for each block.

 

- gsand

;D

[Richi]

Honestly I agree with this statement and think it isn't a ridiculous idea to try this out.

In my agreeing vision, Our server is mostly and almost completely vanilla (Guard and staple required plugins don't count when I'm saying this). SafeBuckets let's you place water and lava that are stationary, blocks which are not obtainable vanilla without a valid bug or glitch.

What this does is make floating water/lava a lot more easier, for example you do not have to place ladders/signs below the block to make it floating (same with lava blocks, because of the fact that fire spreading is off. People quite take the advantage that you are able to make this float, like making lava/obsidian checkerboard walls, lava walls, etc. I personally the plugin should be removed, or either not allow taking a advantage on the plugin.

 

The only problem are for only people who are clumsy are pointed out. You can grief much easily on other people's builds because of the spread, which isn't really a big deal anyway.

 

As I said this is just my opinion and I can be wrong on some facts I said.

[roastnewt]

Why not synergetricks proposal? it seems to be the least performance intensive solution whereas roasts would be called 20 times per second times the number of places water would flow for each block.

 

- gsand

;D

 

 

I fully endorse syn's solution, it's much nicer than mine. :P