Implemented Deletion of accounts

[DeletedUser]

Anyone familiar with how databases work will agree that inactive level 1's should be removed from the game after VERY little time...

Sure, individual inactives do not contribute significantly to query clogs, etc., but every single player (active or not) requires a great number of entries in various tables on the db server. All of that clutter piles up quickly, and can lead to some very noticeable performance issues. I don't know how The West db is structured, but inactive users are always a drain, regardless of how their data is stored and accessed.

One example of a query that hits a lot of crap records is Ranking. I don't know if each page populates on the fly or what, but if the entire ranking loads at once, then you're hitting the database thousands (and thousands) of times -- and that's just one person clicking a button. Even if each ranking page queries the db for ONLY that page's results, the app still needs to sort through ALL player records to find out who is ranked where. Gah.

We may not see problems with system performance yet, but I suspect The West's population will only gain momentum as future builds are released. It would be prudent to nip the problem in the bud.

Let's not forget -- also -- that inactive level 1's contribute literally nothing to the game's community. Should they have rights similar to higher-leveled players? My opinion: no.

 

[DeletedUser]

Yes, and you completely ignored what Diggo said about 2-9 as a group.

I chose not to respond to it verbally because the chart I posted tells the whole story. Sure, when you clump 2-9 all together, it almost starts to compare to level 1 (and 10+), although it's still a lot smaller. But look at the distribution per level across 1-9 and the problem is instantly obvious.

No, it's not as bad as level 1, but it's still a sizable percentage of level 10+.

It's tiny per level, which is just what we'd expect from a constant inflow of active players.

If they're relatively active, they have nothing to worry about, but half of them probably are not.

How has this been determined? And anyway, with the numbers being so tiny, who cares? Those few dozen orphans will go after 45 days. Meantime, there just aren't enough of them to bother about, unlike the vast forests of level 1 orphan zombies.

Inactive level 1s are the real problem, but all lowbie inactives shouldn't be ignored while fixing a problem that may be hardcoded as Whistlingleaf suggested.

Thanks for recognizing the painfully obvious. I don't object to including a few more levels above 1 if everyone else wants to do that. I only object to designing a solution that doesn't match the problem, out of a desire to factor our own expert assessments of the ease of leveling up to X or Y level into the solution. The accuracy of those assessments isn't at issue; the complexity of the proposal in comparison to the simplicity of the problem is. As we've both said, and my chart (based on emperor wes' numbers) shows clearly, the level 1s are the problem. In the interest of simplicity, I say stop with the 3- or 4- tiered proposals here and make it VERY SIMPLE. Level 1 (or, if y'all insist, 1 through X) should have a short grace period; leave the rest as is.

Simple. Neat. Easy. Problem identified, problem solved.

Next....

Now let's say we do it my way and aim this just at the level 1s. A week after implementation, the zombie orphan forests are cleared and we're down to the actives plus a few inactives higher than level 1. At that point we can always reassess things and call for another adjustment if we think it's needed. Based on the numbers at hand I don't think it will be, so I say let's just do level 1 and keep it very simple. The normal 45-day reaper will keep the remaining levels thinned sufficiently IMO -- and I'm basing that opinion on the numbers we have on the table, plus my IG experience. YMMV on the latter, but the former is objective info. (I'm assuming emperor wes did an accurate job of it!)

P.S. As to the hard-coding, I don't expect that adding another tier after implementing the level 1 fix would involve more than a copy-and-paste with a quick edit on the new line to change the level number for that line. It'd be a 5-minute job. Just my guess though.

 

[DeletedUser]

Yes Luap, we should get rid of all the inactive level 1s. But as we are fixing the main problem of low level inactives we might as well talk about getting rid of the other low level inactives. I think that is the only point they were trying to get across. No one is saying you don't have a good idea. We are just trying to expand on it to make it the most effective it could be when dealing with the inactivity problem.

 

[DeletedUser]

Hmm. I was ready to rant, but I guess you may have a point. I was thinking the suggestion should be an alteration to the 45-day inactives, but I guess it would be much simpler to add a second reaper for 7-day level 1 inactives, or whatever.

 

[DeletedUser]

Yes Luap, we should get rid of all the inactive level 1s. But as we are fixing the main problem of low level inactives we might as well talk about getting rid of the other low level inactives.

My point is that there is no problem there. Look at the chart. You're wanting to solve a nonexistent problem.

Hmm. I was ready to rant, but I guess you may have a point. I was thinking the suggestion should be an alteration to the 45-day inactives, but I guess it would be much simpler to add a second reaper for 7-day level 1 inactives, or whatever.

Righto.

The-West, now with DUAL REAPER ACTION!

 

[DeletedUser]

I have level 1 players at my town and they have been there and never moved since the town was started. Would be nice to get those deleted. It would make it easier to see who is in town without having to scroll through 10 level 1 inactive players.

 

[Red Falcon]

Okay, here is what I recommend. For each player at levels 1 through 5, they have 5 days until their account gets deleted. For each player at levels 6 through 10, they have 10 days before their account gets deleted. If they can stay active long enough to get past the required level to choose a character class, then they most likely will continue playing The West for at least a few more weeks.

 

[DeletedUser]

Okay, here is what I recommend. For each player at levels 1 through 5, they have 5 days until their account gets deleted. For each player at levels 6 through 10, they have 10 days before their account gets deleted. If they can stay active long enough to get past the required level to choose a character class, then they most likely will continue playing The West for at least a few more weeks.

Why have 3 tiers, though? That's more complicated than the system needs to be to fix the actual problem.

 

[Diggo11]

Yes, and you completely ignored what Diggo said about 2-9 as a group. No, it's not as bad as level 1, but it's still a sizable percentage of level 10+. If they're relatively active, they have nothing to worry about, but half of them probably are not. Inactive level 1s are the real problem, but all lowbie inactives shouldn't be ignored while fixing a problem that may be hardcoded as Whistlingleaf suggested.

Thanks for that image, I still can't rep you though. I actually tried repping as many people as possible so I could rep you, but that still failed.

I chose not to respond to it verbally because the chart I posted tells the whole story. Sure, when you clump 2-9 all together, it almost starts to compare to level 1 (and 10+), although it's still a lot smaller. But look at the distribution per level across 1-9 and the problem is instantly obvious.

Ok Luap now you are being stupid. You are clumping all level 10+ so you can not argue against clumping level 2-9! And remember you a comparing a mere 8 level range to a 91 level range here, when accounting for that you will see my point.

Yes Luap, we should get rid of all the inactive level 1s. But as we are fixing the main problem of low level inactives we might as well talk about getting rid of the other low level inactives. I think that is the only point they were trying to get across. No one is saying you don't have a good idea. We are just trying to expand on it to make it the most effective it could be when dealing with the inactivity problem.

Exactly, thanks to everyone who said this too. Whilst we fix this problem we may as well fix this too.

The-West, now with DUAL REAPER ACTION!

If they ever implement this I want to see this line in The Western Star.

Personally, I am a big proponent of warnings. I think The West should send an email to you letting you know you have 72 hours left before account deletion, etc.

Some people just get busy with work or school and forget to check their game. BUT... they shouldn't be punished for having a life. A warning would help with active player retention, as well. Players who take the time to log in after receiving the warning are more likely to continue playing when their schedule settles down (I would suspect).

Addendum: I DO think completely, shamelessly inactive players should be deleted, and I DO think level should play a part in the timeline. A level 70 player should get a lot of time, a level 5 should get very little. I still think a warning should be issued via account email beforehand.

I think this shall be part of the final proposal.

 

[DeletedUser]

Ok Luap now you are being stupid. You are clumping all level 10+ so you can not argue against clumping level 2-9! And remember you a comparing a mere 8 level range to a 91 level range here, when accounting for that you will see my point.

Oh brother. No, you're missing the point and being stubborn. 10+ isn't the issue; forget about it. Look at 1-9 broken out and ignore 10+ altogether if it's going to distract you so much (or go get all the numbers for all those levels; I would if I wanted to spend an hour doing it). It's quite plain: the logjam is virtually ALL at level 1. Comparing 2-9 combined vs. just level 1, and then arguing there's a problem in 2-9 because the result almost compares to 1's, is silly. The breakout tells the tale clearly. It's you who are ignoring that. Apples to apples, por favor.

We could combine 2-99 and get our panties in a wad over that result too. It would be every bit as legitimate as what you and Elmyr are trying here -- all just to prove me wrong. :laugh:

And FTR, I didn't clump 10+. emperor wes compiled the numbers. I simply charted what he gave us. Had I realized it would prove to be so distracting I'd have left that last number off. Didn't figure it'd hurt anything to have some extra info. {sigh}

Exactly, thanks to everyone who said this too. Whilst we fix this problem we may as well fix this too.

There's nothing there to fix.

 

[DeletedUser]

Luap, I give up. If only to please you in this one thing so that you will stop shouting at us. I shall not stand in your way on this thread a moment longer. In 3 months we can start talking about the other inactives when people get their panties in a bunch about them.

 

[DeletedUser]

Luap, I give up. If only to please you in this one thing so that you will stop shouting at us. I shall not stand in your way on this thread a moment longer. In 3 months we can start talking about the other inactives when people get their panties in a bunch about them.

That's actually what I suggested.

And Diggo's shouting louder than I am.

 

[Diggo11]

Well Luap let me present my findings.

So we both agree that the swarm of level ones must go yeah? Well for the sake of this example say that happens. we are now left with the following stats:

Now you may not see a need yet, but that is because you are comparing apples with pears. You see you are comparing 89 levels to a mere 8! So I devised the graph below to compare (metaphorically) apples to apples.

So [in world 6] if you pick a level at random between 2 and 9 it will have approximately 962 players. But if you pick a level at random between 10 and 99 then it will only have a mere 174 players! When you look at it that way you will surely see my point.

So in conclusion, in World 6 we do need that middle bracket yet in World 2 we don't. So what is the answer? Well I'd say we state these facts in the proposal and put the middle bracket as an OPTIONAL extra to the top and bottom brackets; let the developers decide for themselves. Can we agree on this, or do my findings sway your mind?

 

[DeletedUser]

I would be happy if the game mods would just delete the inactive players after 30 days like it says in the rules.

 

[DeletedUser]

@Diggo: Nice follow-up graph. I see you used the Excel 2007 default colors as well. I was much better with 2002/2003 graphs, and I haven't used them much in 2007.

Dealing with the inactive level 1s alone would be enough to keep all worlds open for a long time. Deleting them combined with the 45 day deletions for level 2-9s might be enough. If it isn't enough, at some point the worlds will be clogged with inactive level 2-9s and they'll have to be dealt with down the road.

Deleting inactive level 1s promptly might solve the problem, but it might just be a case of future generations having to clean up our mistakes and deal with inactive 2-9s themselves.

 

[DeletedUser]

So [in world 6] if you pick a level at random between 2 and 9 it will have approximately 962 players. But if you pick a level at random between 10 and 99 then it will only have a mere 174 players! When you look at it that way you will surely see my point.

The rest of the graphs needs to be broken down per level again, then there will be a gradual decline in the number of players at higher levels. There will therefore always be a next lower level with a higher number of player, regardless of at which level you stop. If you weed it out to level 10, you will have the same picture with the next 5 levels as well - where do we stop then?

I would like to highlight Luap's point - from the graph, everybody agrees that there is a significant problem with level 1's, while there are some debate about the following levels onwards (imo not a problem currently). To keep it simple for now and present something easy to implement for now, removing a problem we all agree we have now, I would suggest we focus on removing the inactive level 1's, and include level 2's as well (there are significantly more level 2's than 3's, after which the data does not change significantly any more - probably due to most people only doing the immediate levelling up when they start). Let's work that up as a simple proposal (including the warning), and ask for a vote on that?

 

[Red Falcon]

I like those charts you drew up, Diggo! Great job!

 

[DeletedUser]

I don't think this idea is going to go anywhere. They're going to just keep opening world after world.

http://forum.the-west.net/showthread.php?p=142022#post142022

 

[Red Falcon]

You're right Elmyr. They're doing the same damn thing with Tribal Wars! Opening new worlds isn't going to get more people to buy premium! But apparently, Inno doesn't seem to get this!

 

[DeletedUser]

I was amazed to see they're opening two new worlds.

Seriously, this is starting to remind me of debt-based economies with their attendant "bubbles". I think you may be right, Elmyr.

Dig: thanks for the intelligent post and the work you put into it. I still maintain that the problem is in level 1. Consider that W6 is much newer than W2, so there aren't as many higher level chars there yet. That creates a different distribution than W2 has; it's weighted more toward the lower levels since people haven't had time to get as far yet. W2 presents a truer view than W6 of the way all worlds will tend over time, IMO. (We could do a couple others for comparison; I'm confident the pattern would hold.)

I'd say, make 2 reapers, not 3, definitely not 4. Have the low one cover level 1, and see how it goes. After a while, if we are seeing a problem with levels 2 or 3, the level 1 reaper could be extended to cover any necessary additional levels -- because, as you and others have pointed out here so vigorously, it's easy to level up past those first few levels. I'm certain it wouldn't need to extend very far at all. The distribution patterns flatten out nicely on both W2 and W6 once you get past level 2.