Idea Exchange Priorities

[JackDunaway]

I have been throwing around an idea for a little experiment to see which Ideas on the Idea Exchange people really care about. There are close to 1000 Ideas now, and I have supported roughly 20% of those. That's 200 features that I would like to see NI implement! Obviously, that's too much for one, two, or maybe even three release cycles. Take a look at a similar situation Spolsky faced here a few years back:

"By the time we were ready to start development, we had enough ideas for improvement to occupy 1700 programmers for a few decades. Unfortunately, all we have is three programmers, and we wanted to be shipping next fall, so there had to be some prioritization."

Anyway, my experiment is such: anyone who wants to participate gets a fixed amount of "money" to "invest" on whatever idea they want. This is a classic management budgeting activity, and it hit me a few months ago while reading Spolsky's article (linked above) that it could be perfectly applied to the Idea Exchange.

The rules are as such:

1. Anyone can participate, and each participant automatically receives $100
2. If you are an NI employee, you receive an extra $50
3. For every 1 year you have been a professional LabVIEW developer (getting paid real money to program) you get an extra $20 (maximum cap at 10yr, $200)
4. You can allocate ANY amount of money to ANY idea in increments of $1.
5. You can change your mind and redistribute your investment at any time (just don't be too finicky)
6. After all participants have divvied their money, the ideas are ranked by total investment in that idea.
7. The leaderboard total will be calculated every few months
8. Once an Idea is "In Development" or "In Beta", you receive a refund for any $$ you had invested in that Idea, to be redistributed at your discretion

For instance, I am not employed by NI, and I have been getting paid to program LabVIEW for 4 years, so I would receive $180 to divvy amoung whatever Ideas I deem most worthy. Example investment with my $180 (how I actually would have divided my money for Ideas that are currently In Beta. You may notice that I Kudoed many of the 12 Ideas currently In Beta, but they don't show up as my "favorites" that I "really want" below) 

1. $100 to View Cluster Constant as Icon
2. $30 to Local Variables Redesign
3. $30 to New Boolean Diagram constant design!
4. $20 to Indicate the Display Format of Strings

My purpose for this experiment solidarity: to identify the most helpful Ideas as a group, giving slight weight toward those of you who have the most experience. Also, this reveals our personal weightings for our votes. For instance, although I voted 1 Kudos for the New Boolean Diagram constant design, I wish I could give 20 Kudos to the much more powerful idea of Option to discard stale entries in the event queue. Today, the first Idea has 288 Kudos, and the second way more powerful Idea has 24 Kudos. WHAAAAT?

Would anyone else participate in this experiment? If you would, please reply back with your tweaks on the rules. This thread will not act as the experiment thread, just for setting up the experiment. If enough people would participate, then I'll start a new thread where we each post our investments.

[JackDunaway]

I realize that the experiment is only an adaptation of Spolsky's excercise. Since we are not developing LabVIEW, we outside R&D cannot estimate the cost to implement a feature. We are simply placing our "investments" as how much we desire the feature to be implemented.

[tst]

I suggested a very similar mechanism to Laura after playing with the exchange for a while, except my suggestion had points instead of $ and the points were assigned based on number of posts (e.g. you get 1000 points after your first 30 posts and an extra 5000 points every 1000 posts). My basic reasoning was the same as yours (this would force users to actually think about what they want and will allow them to give more weight to things they want more), but I didn't push it because it was pretty obvious to me that it would not be implemented, for two reasons:

1. It would be considered to be too complicated for the users.
2. Lithium already invested in their concept and they would be unlikely to change it.

What I really don't want to do, though, is manage this manually. Unless there's an automated tool, I doubt I would do this. Maybe a VI which converts everyone's data to a standard format or uploads everything to a DB on the web?

[JackDunaway]

Could you post a link to your suggestion?

Ideally, this experiment would be implemented as an automated system, and it would actually replace the Idea Exchange, but like you said, Fat Chance. My original guess is that maybe a dozen or so people at max would participate, and I would not mind doing that by hand just based on trudging through replies in a thread. However, if there were a web front-end, this experiment would gain WAY more participation since it would be very easy to implement.

Does anyone want to even consider some web development to implement this type of a system? Are there any turn-key free or paid solutions?

[JackDunaway]

Oh, and as a reply to concern #1: I actually consider this method a bonus. Fly-by users give a quick Kudos on the Idea Exchange, which is perfect for getting a really "rough estimate" of the population desires. Here, the desires of the fly-by/casual/newbies is weighted more heavily, because there are more of them.

With our proposed new system, it would give weight to the people who cared enough to give really careful consideration. Neither system is a replacement for the other, but I think both systems are necessary to gain a better understanding of the needs and wants of the entire spectrum of users.

[tst]

---

JackDunaway wrote:

Could you post a link to your suggestion?

---

 I could, but then I'd have to kill you. It's in a private board (I was hoping it would be easier to work out the details that way), but I can post my text:

"Incidentally, I feel that the current system is actually lacking in concept. Today, I generally give kudos to any idea which I like and would like to see implemented. A better implementation was if each user had a limited number of points they could use to vote for the ideas. People could dynamically retract their points or would get them back when ideas are implmented or rejected. This way, people would be a lot more judgemental. You could probably get around the new-login-for-more-votes issue by only letting users with more than X (not a high number) posts to have points."

---

Does anyone want to even consider some web development to implement this type of a system? Are there any turn-key free or paid solutions?

---

I don't know about existing solutions, but you don't need web development. I assume there are sites which will let you host a DB (or maybe even files you can manage on your own, like Google sites), and you can simply have a VI which will contact the site and serve as the GUI.

[JackDunaway]

---

tst wrote:

...you can simply have a VI which will contact the site and serve as the GUI.

---

I know of sites that would host MySQL databases. The problem I'm more worried about is data-mining the NI Idea Exchange for an entire list of the ideas and their statuses. I'll look a little more this evening, but are there legal issues data-mining the NI forums?

[muks]

---

A better implementation was if each user had a limited number of points they could use to vote for the ideas.

---

I think this will be a real intresting way to see the priorities. Now that the idea exchange has been active long enough to get an idea and help in setting this limit..

[tst]

---

JackDunaway wrote:

but are there legal issues data-mining the NI forums?

---

Maybe there are (although I doubt it. This is a public site, after all, and you can read it any way you want), but it didn't stop people in the past, and NI does not seem to mind as long as you don't create a strain. If it's only your connection doing the requests, there shouldn't be a problem.

[JackDunaway]

---

tst wrote:

If it's only your connection doing the requests, there shouldn't be a problem.

---

I'll see what I can do. Stay tuned for a couple days...