Great Idea Ben - I was thinking the same thing.

There are some ideas I would spend all of my money on!

Another criteria would be cost.
Each user can choose between 1 and 5 how much effort he perceives the feature will cost IBM to implement. Of course we don't really know how complicated it will be to add certain feature, but maybe the average gives some clue.

It's an interesting idea Ben. What happens once you've run out of money? Or do you get an allocation weekly or monthly or something?

You seem to be coming round to Nathan's view that voting should have some cost associated with it (see his battle idea { Link } ).

I am still to be convinced. I like the simplicity of just voting up and down but I do agree we need some way of having top ideas. I think Rob's idea is probably the best way of doing this: { Link }

Why not just allow, for each vote, an 'importance' value.
e.g. 1=OK, but not that important to me, ... 5=Great, very important to me
This way you can vote, and at the same time give your appreciation.
A bit better than the binary promotion you got now, and it does not have to make the UI more complex.

Matt - I've always thought that there should be a cost, so I'm not really sure that is "coming around to Nathan's view", but in any case, I suggested the favorites idea before to Bruce, but I think it would be somewhat cumbersome to manage, and much less clear what happens when you have three favorites vs. all ten, for example.

I do like the simplicity of just voting up or down, but I have also seen the cases where somebody simply votes No for all kinds of issues, and also am concerned that Nathan is right that the top vote getter right now is a relative light weight issue that I voted for but doesn't have the importance to me of a number of other issues. I could withdraw me vote, but I think it is an OK idea, just less important and there is no current way to express that.

@5 - Sorry must have mis-remembered who said what when we've had this discussion in the past.

@4 - I do quite like the idea of scoring an idea rather than just voting yes or no for it. It would be a fairly significant change in design so we'd need to look at if/how we do that. One for Bruce to chime in on I think.

Maybe you could earn dollars by submitting ideas/comments which would encourage contributions...

I don't agree with adding more factors. The beauty of the tool is it's simplicity: it's easy to vote just by clicking a button. If we add other fields to it, it will take more time, get less voters that are interested in submitting.
Compare it to a subscription form where you only need a username and a password, and one that asks for all kinds of information. I don't like to do a lot of input for this kind of applications.

A while back I thought that if I just like an idea as opposed to like it a lot I do not have a way to show that. I sort of settled into if it at all makes sense and I care to see it done I vote for it. So I opted for a lower threshold thinking that there are others who will have a higher threshold.

That said I wonder if there is any data or voting theory that would make this decision a bit clearer. I suspect that like for myself it might be just a gut feel that things are sub-optimal and there are perhaps a few ideas that did not seem to rise to the top while they are deserving. Partly to blame might be the top thirty. The popular just get more popular since they have the lime light all the time while other ideas just continue bumping the older ones off of the treadmill.

So going back to my perceived lack of expressiveness in my binary vote. Like Theo I thought of allocating say 3 votes per idea to be used for or against. If I really like it and place a high priority on it I give all three votes. If I really like it but can live without it for now then its two votes and one vote for something that is a good idea but only a nice to have as and when. No vote still shows indifference or lack of opinion.

As a variation to this we could have one vote per three ideas so we would have to choose more carefully. This would be harder to implement as when do you decide that a person has voted enough that they need to wait for more ideas to be submitted. Just thinking out loud but I feel this is not so good.

As for the pool of money to spend on ideas as one pleases like Matt I am not sure about introducing this artificial scarcity. Yes there is a cost to ideas. However, we are trying to gauge popularity not estimate cost of implementation per se. Having a limited pool of money means there is a cost.

I think that the primary issue is that some ideas will fall through the cracks in an occasional avalanche of submissions. How do we make sure that that does not happen?

How about displaying a random idea in a fairly prominent spot on rotation over say a few hours. These ideas can be picked from the less popular ideas as a kind of reminder for people to take a second look in case they missed it.

Other than that if we take this site as place to submit and tally up the votes the blogosphere then takes on the role of marketing and spreading the word. That's where you campaign for the merits of your idea. I realise then that we have a situation that those with the loudest voices may gain the most exposure. However, I think that this is a meritocratic community so I would probably not worry about that too much.

That would be my $0.02 ...

Great idea and feedback from all. Let's keep the dialog going. Currently we are seeing lots of new ideas as we have lots of new users. The dev team is watching things very carefully and is exploring different ways to present various ideas to users for voting.

I think this is an excellent concept and discussion to have, I'm still working through the mechanics in my head (a scary picture BTW).
Nothing's going to be perfect - they're all just approximations. The idea is instead of just blanket YEAH! to everthing, to force you/us to weigh and (the point:) allocate importance to the different ideas. The 'IJ$' idea is very good, but I would say that we each user is given, say, IJ$100 per month to allocate instead of an absolute IJ$100.
Also, the I'm wondering about the conversion rate - Does this mean that Canadians/Europeans get more IJ$s than Americans? 8-)

I don't really like the "importance" approach because it doesn't force any kind of allocation - everything can be important. However I DO like it better than just the blanket up/down.

@11. I was just going to say that we should probably settle on a hard currency like the € :-)

@12 - just wait... give it a couple of years, like when we get a real president. The currency roles will reverse (as they have in the recent past...)


NOTE: That was not a partisan political comment. I'm from Louisiana - it should be easy to work out why I might not like the current administration.....

Just generally on allocating IJ $. How will you determine how many dollars to have in circulation? It is important because you run the risk of inflating or deflating your dollar supply in relation to the number of ideas in a given time. This means that you will not be able to compare the relative value allocated to ideas over time unless you track the inflation index and are then able to convert any past amount to current dollars.

This is the situation you have with the banking system right now and it does not seem all that stable.

@14 - Agreed, that's the core issue with the approach. But to carry the analogy along: If you just allow 'importance', then you're basically giving everyone as much IJ$ as they want - they can spend all of it on every idea they like.
So, the IJ$ approach is not perfect, but IMHO, it's the most adequate of the options.

You all can redefine the Idea all you like, but my goal was explicitly to have a set amount, not new amounts each month. My goal is that people would have to set priorities. I don't think the UI would be hard, just a few of all the items I have voted, sorted by how many dollars I have allocated to each perhaps. Have a total clearly visible, and an entry next to each item in which you could put a number. If the $ concept gets too messy, call them points instead, but I did want more than 100 to allow some granularity.

@16 - sorry! Not trying to redefine your idea, just think about how it work in practice.
So, if you only have IJ$1,000 - I'm assuming that you'd allow me to re-allocate my IJ$ as newer ideas come available? For instance, if I've already allocated all my IJ$ and an idea I liked was submitted, I could go to my previous 'votes', and deduct some IJ$ from them in order to scrounge up some IJ$ for the idea?
That works, but it seems really hard to program. That's why I was thinking of having a monthly allowance that I could send on ideas as they rolled in.....

@17 - Actually, I guess a different spin on the same thing that might be easier is to simply rank your own Promoted items. Assuming you could make drag and drop work in the pseudo-view, you could simply say "this wish is higher than that wish". Then, the math of the dollar/point allocation would be implicit rather than explicit. To make it even closer, you could exclude any approved items you wanted, so you could just have three "really, really want" items on your ranked list, or thirty, and the dollar/point amounts would correspond. This would be a lot more visual and interactive than adding up numeric counts and all, but would have the same general intent. Should I make that a separate idea?

@18 - How about allowing you to only put, say 10% of the ideas you voted on into your "hot list"? This would encourage participation while still letting people indicate what is most important to them. I do think the bookmark function Rob McDonagh asked for is still relevant and serves a different purpose.

Why add another voting scheme with IJ$? Why not just one vote per idea, but you only have x number of votes to use. That way, I can't just put my 1000 ij$ on one idea and give it the same importance as an idea that 100 people voted 10 ij$ on. I think that the scheme needs to keep community importance in the fore-front. It shouldn't be too hard to design so that if I want to cast my one vote for a new idea and I'm out of votes, that I get a list of currently voted ideas to choose the one I want to yank my vote from in favor of the new one. That way, items will be prioritized as the idea list grows.

I just don't see the need to have multiples of something to plunk down on any one idea.

@16 - It's just me supporting your idea with comments, some reservations.

@14, 15. - To answer my own question about inflation/deflation in the value of allocated IJ$ we could do this. Allocate an additional vote IJ$ for every created idea to everyone. So as new ideas get submitted you pool of allocatable $s grows proportionately. (There is more to it as it is still not ideal but then it gets even more complicated) Now this way it is no different to just allowing two or more votes per idea, with the exception that from the secondary pool we could allocate all to one idea.
Not perfect. I actually see more issues with trying to impose a cost on promoting. This is essentially what it is. Remember we voted against imposing a similar cost on demoting previously.

Your original idea Ben while I think not really equitable is simple enough to implement. The variations proposed here seek to make this idea more equitable but they complicate.

Seems to me like the linked idea of increasing the number of votes per idea to 3 or 5 is a good compromise out of this. :-)

I really prefer the simplicity that the current scheme has, but I also can not disagree with you that there needs to be a manner in which I can say that I prefer one idea over another. As was stated above, if I can re-allocate the points/money/whatever then I would be more in favor of this idea, but I also don't want to have keep re-allocating. I think that it may become too much work for most people, and before anyone says that that will just ensure that we have quality votes then let me say that I think the only way that IJ will really succeed is by the shear power behind a large number of people using it so it needs to remain easy.

Currently I am torn between Nathan's Idea Battle (which I have seen implemented for a different voting-type system and it worked very well at promoting the good ideas) and Ben's re-imagining of the original idea (see @18) with a top-ten list that I can use to rank ideas I have voted on. Right now I like the top-ten list idea just a bit better than the Idea Battle because I feel I have more control.

This is actually a pretty good idea, but it's got one new issue.

I register and allocate my 1000 dollars.

Then I register with a different email address, and allocate another 1000 dollars.

Now, put a credit card number behind this idea, and I'm 100% on board. But then, you'd start to undercut the commercial proposal I'm putting together for Bruce. ;-)

I like the idea of differentiating ideas that you like from ideas that you really, really like.

But I don't like the dollars idea. I'm with Martin @8: "The beauty of the tool is it's simplicity".

I would prefer a simpler system based on what we've had in the BP forum for the past 10+ years: each registered user has 10 "Top 10" votes to give. I.e., after you click "Promote" you see a "Top 10" button, which you can click. The system prevents you from giving out more than ten Top 10 votes. In your profile, you can see a list of your Top 10s. You can click a "Withdraw Top 10" button on any of your Top 10 ideas to. The system will display Total, Promotes, Demotes, and Top 10s.

Caught this just after I my 1st entry for a coporate Out of Office Calendar (see Server Ideas.) I would hope that Lotus and IBM development is looking at this daily and has the ability to tag / prioritize some of our great ideas. I might add that our ideas are pro-bono, no patent agreements and should be a great feed to the OEM. At least, I think, we are all spending our time with a vested interest.

I think you have the right idea, but the wrong execution model.

If you're dealing with actual customers, I would simply take the votes with user ID in tow and put all the information into a business intelligence tool (e.g. Cognos, Intelliprint, etc.) and put the votes up against the "loyalty" of the customer, whether that's straight dollars, number of visits, etc.; that's up to the business.

But let the customers actions be the currency to add or subtract weight to their opinion.

I'm in the keep it simple camp. Certainly we should wait to see what happens over the course of a few months before messing about with it too much.

I really don't see what the big deal about having a no-brainer simple idea at the top of the idea chart is. OK, no one is going to go nuts if it doesn't get picked up, but does it really matter? It's not like anyone is sitting out there thinking that the top X ideas are going to get implemented and the rest ignored. It's just a measure of who thinks this is a good idea, who doesn't. Nice and simple.

<plug type="blatant ">
I'd still like to see simple vote ratio somewhere. :) { Link }
</plug>

On that note, if you have an idea you want to get votes for, go out and pimp it!

Finally, I demoted this idea in favor of Theo's idea with 5 (or 3) weighted vote options (see linked idea).

Oops, in the meantime, another idea was linked to this one.

See Add a 3-level oder 5-level promotion at { Link }

I think this is over-engineering. The primary goal of the IdeaJam is to generate and INITIALLY rank ideas. If I have to start budget planning on a virtual budget to prioritize ideas, seriously, I won't do it. Despite all idealism and enthusiasm and all, I won't do it.

Let alone start to think about how much effort it may take IBM to implement it. What information shall I base this estimate on? I don't have enough insight into the inner workings and dependencies of the whole product.

What about this: each user get 1 $IJ per idea in the system. When a new idea is added, everyone get another 1 $IJ. This way one can use 1 $IJ per idea, or mark some as "no opionion" and save those $IJ for more important ones.
Anyone that create an idea that after 5 votes have a positive number of votes (more promotions than denotions) get 5 $IJ additional. This would encourage the creation of good ideas.
Of course, new ideas need to be monitored somehow, so people don't enter anything into the system (spam, racist comments, etc), the same moderation would ensure that people don't add bogus ideas just to get points.
I seen some users (I seen a few with obvious bogus names) that seem to demote everything. "Blah Blah" is a good example...
{ Link }
By the way, how is the "report user" function { Link } coming along?