Enthusiasts 2

[Ray.R]

---

@ChillyCharlie wrote:

number index = answer number/5000 . . . . . .(this will leave some room to Dennis...)
rate index = Average answer rates/5 . . . . (5 answers a day, on average !)
quality index = (c * ((average rating - 3.5)/1.5) + ratings/answer number)/(c + 1)

---

Excellent work CC!

As for the roof of 5000, Dennis will be there in no time. Maybe you should consider his number of answers as the roof (dynamic) or round up his number to the next 1000 so that he consatntly chases after the roof (he may not like that one).. Or a rule so that he reaches it, surpases it by 10 answers then the roof re-aligns itself to the next 1000. In any case, that roof will always be moving away from us 😞

What does "c" represent? c = constant??

Ray

[Ray.R]

---

@chilly charly wrote:
More generally, my idea would be to develop an index based on 3 parameters : answer number, anwer rate and quality. For instance :
Enthusiasm index = 100 x (a * number index + b * rate index + quality index)/(a + b + 1).

---

What about the score average given? Should that appear in some form of equation?

😄

[chilly_charly]

a, b, and c are constants that need a serious tuning. I have played with them, and with some combinations, I ended up as being n°1. I must admit that this is a very comfortable idea, however I also recognize that this could be slightly unfair for some minor members (Dennis, Ben, GreG...) 🙂

The idea of using Dennis performance as a standard is something that has haunted me for a while. May be Ben, although being a semi-god now with 3 golden bars, could be a more human reference ? 😄

CC

[chilly_charly]

---

@JoeLabView wrote:
What about the score average given ? Should that appear in some form of equation ?

---

Unfortunately, I do not have access to the given scores. That's a private information, as first and last connection dates (although for these 2 last infos, there are some work-arounds).

Anyway, I don't think that this could be taken into account, since there is no fool-proof mechanism to avoid over-use of this function. I do agree that the ability to distribute stars is a very valuable feature (without Ben around my own score would be much lower !), but I don't see how we could seriously rate it.

CC

[Ray.R]

---

@chilly charly wrote:
The idea of using Dennis performance as a standard is something that has haunted me for a while. May be Ben, although being a semi-god now with 3 golden bars, could be a more human reference ? 😄

CC

---

Yes.. it is difficult to follow into God's footsteps...

Maybe 5000 would be a good starting target. After Dennis reaches 6000 answers, you could change the target to 10,000 😉

[nnmehta]

Hi,

This is an interesting discussion! I work for the company that provides the discussion forums used on this site and we are looking into adding some leader board functionality and also looking at possible ways to compute the top experts. If you have any other suggestions, keep them coming. I'll take whatever is discussed here back to my team.

Thanks.

[Dennis_Knutson]

Not that it matters much, but my first connection (same name but different email) was 6/16/1999. My first connection with the current profile was 2/10/2000.

[altenbach]

---

@Dennis Knutson wrote:
Not that it matters much, but my first connection (same name but different email) was 6/16/1999. My first connection with the current profile was 2/10/2000.

---

Yes, it might be puzzling to some how posts can exist that have an earlier date than the registration date in the profile.

If I remember right, these forums started to exist on 1-25-2001 and "LabVIEW General" mirrored the comp.lang.labview newsgroup. (Most forums are still mirrored to this newsgroup and vice-versa. The newsgroup existed way before that). I guess if you had an NI profile setup on 1-25-2001 you were automatically in the system if newsgroup posts with the same e-mail address existed. Just guessing here!

My oldest posts here seem to be from:
11-04-1999 (as myself)
08-30-1999 (as Christian Altenbach)
They were posted via usenet. There are even threads where both my profiles contributed. 😮

I wonder if anybody from NI remembers details of the "big bang". Some pages of ni.com have been archived in the Wayback Machine, but it seems the Developer Exchange is mostly missing.

[Ben]

Since we are throwing out ideas....

The answer ratings have always been arbitrary and unless we wipe the slate clean, always will be.

Next:
I would like to go agianst me previous statement re:#of stars that can be distributed.

What if we earned stars that we could in turn share?
I am thinking maybe we get 15 stars to distribute for each posting. Novice questioners would automaticly get the possibility of assigning up to 3 five star ratings.

The rest of us earn the stars as we go.

Maybe get 10 stars to distribute every day we log in. That way the "lurkers" can still assign ratings.

I think the "extra bar for ratings given" can still slow down the "one star wonders" from trashing answers.

Next:
How about rating the difficulty of a question? I think this should be researved to the >100 contribution group because by that time they should have a feel for the range of questions posted. The difficulty rating could be used as a multiplier.

Next:
CC,
Great work and efforts on support the exchange. I am concerned though. I do not want you to burn out! Please make sure you pace yourself.

Any thoughts?

Ben

[chilly_charly]

Thanks for caring, Ben !

Just in case you missed it, LabView is doing all the work for me. I wrote a small vi that went through the NI data base (well, the part that's accessible to everybody !..). 40000 entries, and about 800 users with more than 20 answers. Updating the list takes now less than a minute.
No pain, no sweat, just fun. That was tyhe occasion for me to learn about internet connections, and to improve my skills with string manipulation.
Let's study these results carefully and see how we can derive objectively a fair method of evaluation and reward...

CC