Catch-22 about changing the voting system

Here's a conversation I had with Live Help that illustrates some problems with the feature voting system. It's quoted as it happened, with permission from the people involved, except :
1) timestamps are removed
2) text has been hyperlinked to relavant URLs



There are some other votes aimed at reforming the voting system, such as :
* Allow an unlimited number of 1 point votes per resident
Give more votes per player
redirect people to voting system when grid is down
allow voting on acknowledged proposals
* clear up the voting system
link from feature suggestions forum to voting system
allow rating of proposals from "strongly for" to "strongly against"
let proposers cancel their proposals

These proposals basically boil down to a few things :
1) Lindens should detect and close duplicate, impossible or pointless votes in the feature voting system, to make it easier to find good, worthy proposals.
2) Players should get more votes, either a lot more votes (say, 20 or more per person), or they should be able to cast an infinite number of '+1' votes... or the voting system should switch to a system based on "strongly for" to "strongly against".
3) It should be possible for the proposer of a vote to add comments to it later (say, linking to other similar votes).... and to close/delete the vote (say, if the person regrets posting it after severe problems are pointed out in discussion). If they delete the vote, then all votes cast for it are returned, and everyone who voted for it is given at the very least a URL to the proposal's forum topic so they can find out what happened, and, preferably, also a short message from the proposer as to why it was cancelled.

These proposals for the voting system boil down to two things :
1) Getting people more involved in voting - e.g. by giving people more votes to cast and advertising the fact they can vote.
2) Reforming the voting system to make it more useful by removing duplicate proposals, and helping to focus Linden/Resident attention on the proposals that are truly useful.

Another idea :
when someone is proposing a feature in the feature voting system, their proposal is split into words, and the top 5 most popular proposals (measured by total votes) for each word (excluding words like "change", "the", "it", and so on) are displayed.

E.g. If I proposed "change the group system"
The top 5 proposals with "group" in their title would be displayed ("change", "the" and "system" should be terms ignored by the search).

This would drastically cut down on the number of duplicate proposals in the SL feature voting system.

These are all good ideas, and definitely something to look into.

A big problem for me is, that there are so many duplicate proposals that it's hard to know how to aportion your votes. If that were dealt with, and if more regular users are made aware of the voting, I think that a lot of proplosals that have few votes, but which I find a lot of general sentiment for in world would have a better chance of getting the votes they deserve.

I also think that the number of votes required to get a Linden review should be lowered. Something like 300 votes should, IMO, show sufficent interest in the topic to deserve a response. Particularly considering that each of those votes probably represents the opinion of a dozen or more people who haven't made it to the proposal voting area.

Finally, some form of indicating feeling *against* really is a good idea. For example, I don't care if 500 people *do* vote for something like requiring AVs to eat and sleep, or banning sex or the like, they're *still* terrible ideas. Giving people a chance to indicate that on the proposal itself would be a very good idea.

Advertising secondlife.com/vote in the SL client's menu
We really, really need a way to advertise "to vote for" lists like these to SL Residents, to make them aware that they can change things.
Here's another idea.... under the "world" menu, for example, there should be an option "Vote for new SL features" which opens a mozilla window, inside SL, already logged in to the SL website, and loads up secondlife.com/vote - that'd be a great way to encourage people to vote.

-------
A group to encourage voting
Maybe we should create a group, say.... "SL Voting Awareness" or something, to encourage people to *use* their votes?
If I created such a group, with a charter of something like :
"This group exists to announce and discuss popular suggested new features in SL, as detailed by topic URL ______________ . Feel free to discuss proposals on the group, but advertisements for proposals must be done by getting them on the forum topic which is linked to above, in order to cut down on the frequency of group IMs members get."

would you guys join it? It'd be a minimal-spam group aimed at encouraging everyone in SL to *use* the secondlife.com voting system, by providing a group that announced (or at least linked to) major, important and useful new ideas on the forum or voting system.

How about that?

Just making sure that my support for this is known and that no one edits it under the 'naming names' forum policy. My consent was given and i agree that voting needs improved.

/13/be/108078/1.html

My person proposal on this issue. Sad thing is there ARE others created for it and i COULDNT find them prior to creating my own because of the limited topic choices.

THESE ARE KILLER!

Okay let me reflect and let this gel and I'll comment more. Thx Angel for pointin' me to this.

I think the 60-second change is a reasonable one.

If we want to go further and suggest an overhaul of the voting system, this is what I propose:

The way I would change the voting system would be to simply not list individual proposals. Rather, I support a system like that used on KittenWar: two randomly selected proposals appear, side by side. You pick the one you prefer. You are then presented with ANOTHER pair of proposals, and so on.

Over time, this allows a clear picture of what is popular and what is not. (Note on KittenWar, that the majority of the unpopular cats are all those weird pointy-looking hairless breeds.)

I believe that properly-worded proposals would allow for even nontechnical users to understand the benefits of allowing crazy scripting whatever, and to make an informed decision.

The system KittenWar uses is very good for comparing lots of unordered and unique kittens.
Unfortunately, I worry if it would be suitable for the voting system here, because :
1) People use the voting system, usually, to put a few votes on proposals they've heard of somewhere which are relavant to them. Very few people, I would think, have the patience to sit on the voting system, sifting through all the proposals on it which they don't care about... just to eventually reach one they do care about.
2) Votes on KittenWar are not changable after they are cast. If we had a similar system here, then suppose discussion of what seems like a good proposal throws up some seemingly terrible problems.... but most people have already voted? What then? If we allow people to change their votes... then it becomes just like the current system, except that it doesn't allow searching for the most valuable proposals, and requires everyone to slog through random selections from the database.
3) I personally like the idea that everyone gets 10 votes on the current system because it means that a person doesn't get a great deal more voting power simply because they spend more time on the website. On KittenWar, a single person with lots and lots of time on their hands can just sit there, clicking away, hour after hour, to influence the results quite a lot. If we adopt a KittenWar-like system, the maximum number of votes per day per person should probably be capped.
4) If we are forced to slog through proposals we don't care about in order to vote for one we do, then campaigning to gather votes for specific good proposals is likely to be very hard, as we'd have to persuade people to sit on the voting system for hours clicking through until they got to vote for our proposal. This means that fewer people will propose new ideas, and fewer people will campaign for change.... it'll slow down the whole process, imho, because get bored and leave when they're forced to go through a tedious slog in order to get something done.
5) As voting in a KittenWar-like system system would be more tedious (due to you having less control over your voting options)... then it would lead to either less people being bothered to vote, or people casting their votes willy-nilly (imagine people always voting for the proposal on the left side of the page, to get through as many proposals as possible to find the one they want to vote for). We could stop smart people doing this by including a 'page refresh' option, but there would still be some people who skew the results of the whole thing by always voting for the proposal in a certain page position (as they don't have to move their mouse) to skip quickly through the database...
6) To alleviate the above problems, we could let people 'lock' one of the proposals (say, the proposal on the left hand side) and let them compare it to others in a KittenWar-ish way. They'd keep clicking it as more important then the ones it was compared to, until eventually a better one came up, then they could lock that one.... and keep clicking *it* until a better one came up.... this would mean the good proposals accrue points from the process, and the proposal with the most points gets implemented most urgently. This would also let people choose to vote for a proposal they like. Unfortunately it has two serious problems... #1 it would encourage people to blindly rate a certain proposal as more important then others as a way to vote for it, without reading those other proposals, and #2 a system like this is roughly equivilant to the system we have now - (you choose where to allocate your limited number of votes between proposals, and the proposals with the most get attention) - so we have no motivation to adopt this modified KittenWar-like system becuase in order to make it workable we have to modify it in ways which just make it a more complicated version of what we already have.


So, in summary, I think it would be a bad idea to implement a KittenWar-like system for secondlife.com/vote, because it would make voting tedious and therefore discourage people from voting or encouraging others to vote, and because it wouldn't be much help in dealing with the fact that there are lots of duplicate or very similar proposals in the voting system.

Put simply, I like the idea of the feature voting system, but there are two things that need addressing :

1) There are small, uncontentious changes, which would harm nobody and benefit everyone, yet which are not classed as 'important' enough to get peoples' votes on the feature voting system. There needs to be a direct way of contacting LL's feature implementation team about these ideas, bypassing the normal feature voting system.
An example of such a change is the idea that when the SL grid is closed but the SL website is open, secondlife.com/vote should display a link to secondlife.com/vote - to give people something SL-related to do and to encourage voting. Something like this could be done in a few minitues by a webmaster or someone with edit access to the SL website. It is a good idea, because :
A) it would benefit everyone by making the voting system a more useful tool by increasing participation
B) it would benefit the Lindens by encouraging people to spend their time voting for new features on secondlife.com/vote instead of endlessly complaining that the grid is down and nagging LL about when it will come back up.
C) it would benefit the Lindens by making sure the voting system better represented the whole community (by encouraging more people to vote who would not otherwise vote, or were not aware of the voting system) and thus that they could be sure the proposals with many votes on the voting system were in fact highly desirable.
D) it would benefit residents as it would leave LL free to work on the grid when it was down instead of answering resident questions, and put more pressure on them to follow the recommendations of highly voted proposals in the voting system.

In short : there are some ideas which can't pass the test of the voting system, yet which are definately good ideas. We need a direct way of telling LL about these ideas so they can be quickly implemented, bypassing the voting system. Of course, we don't want any drastic changes in SL made this way - so this channel for giving LL feature ideas should be restricted to small, uncontroversial changes which benefit everyone.


2) The voting system is not adequately managed, or advertised within SL or on the SL website. By 'managed' I mean that the tools for residents and seemingly the willingness on the part of LL to do things like:
A) stop people from proposing votes which duplicate already existing proposals, by forcing them to go through a confirmation step before their proposal is checked where they confirm that the proposal is significantly different from the other existing proposals on that topic. For example :

B) Provide the creators of proposals with the ability to mark them as 'cancelled' (which would be exactly the same as "can't do", in that it returns all votes allocated to them, stops further voting on them, but leaves them in the system's records).

C) Provide an option to filter out proposals which cannot be voted on when searching secondlife.com/vote - so that one does not waste time looking at proposals one cannot vote for or influence, and can concentrate better on finding good proposals to allocate votes to.
D) We need to advertise the feature voting system more, in the SL client. It's an important part of Second Life that residents can shape their world, both individually on their own parcels and also collectively shaping the grid via feature requests.

E) We need to create a culture whereby people are at least *encouraged* to vote - where everyone in SL knows about voting and how to do it. We need to educate people, we need to tell them that voting on secondlife.com/vote *does* make a difference to how SL develops, and thus, they should do it. I've come up with some ideas for this, such as :



In summary : I like the idea of voting for new features, but I suggest that :
1) we also have a faster, more direct way of contacting LL about specific, small, uncontroversial changes to SL that they can implement with almost no effort and which would benefit everyone in SL.
2) we reform the voting system in the ways above to cut down on duplicate proposals, make it easier to find good proposals to vote for, and to make more residents aware that they can and should vote.

Another example of a little thing that should be changed :

Another example of a tiny, easy to make change that would help everyone :
Currently, when we search proposals, we can see the "Discuss this Proposal" link on proposals that have a forum topic to discuss them on. We cannot, however, see this link when we actually view the proposal itself. Why not add a "Discuss this Proposal" link (for proposals with a valid forum URL) to the actual proposal page?
A lot of new people don't know how to get the link if they're not familiar with it... and a lot of other people are (understandably) too lazy to search for it without even knowing if the proposal concerned has one.

=== Voting
* allow anti-votes #1 (votes against proposals)
http://secondlife.com/vote/index.php?get_id=96
http://secondlife.com/vote/index.php?get_id=639
http://secondlife.com/vote/index.php?get_id=615
http://secondlife.com/vote/index.php?get_id=1282
http://secondlife.com/vote/index.php?get_id=1242
* Clear up the voting system - remove duplicate proposals
http://secondlife.com/vote/index.php?get_id=1295
* require linden approval for proposals to prevent duplicates
http://secondlife.com/vote/index.php?get_id=1296

A quick point : the proposals to reform the voting system have one major drawback - they don't get many votes - because most everyone who recognises the severe problems with the voting system won't vote on proposals to reform it because they realise that the Lindens don't seem to give much attention to the voting system. I can't remember which topic it was, but I think it was Kelly Linden (a LL developer) who I saw today admit that she basically never looks at the voting system when deciding which features to next implement in SL.

So, my point is : Lindens, pay attention to these proposals to reform the voting system. They don't have many votes, but the reason for this is that most people who see the faults with the voting system..... don't vote.

shameless bump.

The question is: if more people voted, would the Lindens pay more attention?

WTF we have web page guests?

This problem seems closely tied to the other half of a problem I see. I keep finding new proposals being made with no room for discussion being made because these people don't know how to or haven't bothered to make a corresponding discussion topic for it. As a result, votes get fractured by almost identical proposals that have no chance of being adopted alone but would get sufficient attention if they were discussed and refined properly.

The problem is, I can't get people into here to make those proposals. A link from Feature Voting to this subforum would really help my cause, and actually help your feature get some much-needed attention here too by getting them to read this topic too.

Of course, being the only one who ever voted in the proposal, that one seems like it's going nowhere fast too.

No, feature voting was added so that the Lindens didn't have to listen anymore than they were, with the ability to say "please add a proposition for..." for those who still tried to make them listen.

I'm sure there is something ironic about that.

As I have been saying, we need a DIRECT way to contact LL to get small, uncontroversial code changes, like, for example :
* all proposals on the SL voting system which DO have discussion forum topics should display a link to those topics on the page that gives detailed information about the proposal.
* the secondlife.com/status page should link to secondlife.com/vote so when the grid is down, we can get people voting instead

Also, we need :
* the ability for proposers of votes to DELETE their proposals no matter how many others have voted on them, and to send everyone who has voted an email message saying something like "I'm deleting my proposal because it duplicates proposal ____, please go to ____ and move your votes to proposal _____."
* someone with Linden access to go through the database and merge duplicate proposals, combining their votes.
* mechanisms to be put in place to stop duplicate proposals. For example, a confirmation page which takes all the words in the proposal's title, filters out common ones like "change", "the", "a" etc.... and then returns the top 5 voted proposals containing those words, asking "please check the following, to make sure that you're not creating a proposal which is very similar to any of these....".
E.g. if I put "more then officer and member in groups" it would pick up on 'groups' and show me the proposals related to group roles that have already been closed, making sure I was at least discouraged from creating a duplicate proposal on this topic.




Very funny. I just *hope* it doesn't represent how LL actually treats us.

The "Major Major" bit was stolen straight out of Heller's Catch-22.

I see that you are newish here; unfortunately, quite a few customers do get the impression that it is how they are treated.

Suggestions for the voting system:

1 - Allow someone who has submitted a proposal to recommend merging it with another proposal if they feel it is similar. This would send a message to the author of the other proposal asking if they approve the change. If both approve, the two props get merged, with both numbers referenced and links to any discussion areas are maintained. Those who have cast votes (for either of the merging proposals) should also get a notification at this point so they can check the new version and see if they want to keep their votes where they are.

2 - Automatically create a discussion thread in Feature Suggestions for a prop if none has been indicated by the creator. If a legacy prop has been created with no discussion thread, allow the original proposer to add one. Also allow other viewers to sent a message asking the original proposer to create one.

3 - One of the LL responses could be "approved for 3d party development". LL could implement an interface, if needed, and the proposer and/or other smart folks in the community could develop the feature.

4 - votes against -- I don't think this is a good idea; I think it would be better to have discussion in the linked forum, rather than just a "no" vote. A valid objection should be able to stop any proposal, whereas a couple of "no" votes would just sit there, not explaining anything.

5 - I like to advertise props in-world with prims that link to the voting site and that particular prop. I put them near things that relate, e.g. when I was selling clothing, I put notices about capes etc. in my clothing stores. We generally have links to prop Prop: 203 - Support local devices beyond mouse and keyboard near InnerLife areas and activities. It's not appropriate or easy to do for every prop, but it helps get the word out, especially to people who don't read the forums. Other good places to put links to the voting system are in personal profiles, land descriptions, etc.

Just a few thoughts,

neko

I know lots of people who've got that impression, and I think it's partially accurate. LL have a difficult job - there is too much for them to listen to all of, so they have to find the good stuff and listen to that. But they can't say what the way to make them listen is, or else everyone would use that and they'd be back to square one.

So they end up with this system where they have lots of ways of putting information to them but you're never quite sure if they're listening or not unless Torley or someone directly comments on your work and says it'll get forwarded internally to the right linden for action (which has happened a few times to me).

Nekokami Dragonfly's comments are insightful.
#1 is a good idea
#2 I'm not sure of, but I can see the argument for it being a good idea
#3 is a good idea
#4 is good in theory, but in practice I think there needs to be a counterbalance to proposals on the voting system with many votes.... and this counterbalance has to be *shown* on the voting system so time-pressed lindens don't just go by the voting system alone.
#5 is a definate good idea

I wouldn't mind if the "nays" were managed separately from the "ayes". Suppose there is an arbitrary number n of "yes" votes. Then let there also be n/m of "no" votes, where this is some fraction of the number of "yes" votes (could be the same, could be less-- I don't have a strong opinion about this). The reason is, I think people should be careful with "no" votes, and they should carry a lot of weight. I also think "no" votes really should require a comment. Not just "I like idea X better" or "This would suck," but "This is dangerous and here's why." That's why I would prefer to see concerns discussed in a forum-like setting, so Lindens as well as other users -- including those who had voted for the proposal-- can understand what the objections are. Ideally this could result in the proposal being refined to make it better, or withdrawn if it develops over discussion that it's not such a good idea after all.

In fact, I would strengthen my #2 suggestion (above) and recommend that *every* proposal which does not currently have a discussion thread should have one created (perhaps after like props are merged). This was manually done once last year, at least, but if it could be done in an automated way at the same time as the change in functionality to automatically create a feature suggestion thread for props with no discussion, that would be great.

One of the problems with the current system (yes, there are many problems, but here's one) is that votes of people who have abandoned the game still count. Even leaving out alts, this means the votes of former players carry as much weight as those who are still in the game, paying attention to the voting process and the way props evolve. I might suggest that accounts which achieve some level of inactivity (e.g. no logins for 3 months) have their votes removed or greyed out or something.

neko

I agree no votes should have to have a serious, publically displayed reason to count.
I agree that every proposal *must* have a forum topic.
I'm not sure that we should ignore the votes of players who've left SL, because if we did, then we might forget why lots of people are leaving, and thus forget about old problems that need addressing. Maybe a greater timeout, like a year or two would be reasonable... because new players would vote for the proposal in that time. Given the low level of use of the voting system, though, a 3 month timeout for votes would be bad as it would discourage, rather then encourage, use of the voting system and may result in valid concerns getting forgotten.
Proposals already struggle to get the 500 votes required for Linden action. Why make it harder?

I plan to shift this discussion to this related and larger topic to consolidate my ideas about how to improve the feature voting tool into one place.