And a reminder...send in your People's Picks!

Excerpt from original post about it:

Announcing the People's Picks!
Squidoo is launching a new series of Top 100 lists: the People's Picks. We're inviting our best lensmasters (yes, that's you) to be honorary editors and submit your own lists of Top 100 lenses. It's your turn to put a popular spin on LensRank.

How? Just make a lens linking to your favorite 100 lenses, themed or not, Plexo'ed or not.  Then email the lens to peoplespicks [at] squidoo.com for review and official posting on our Top 100 lists page. Let the Picking proceed!

HI Maurreen,

I'm curious how you came up with the formula for multiplying by the sqrt of the # of ratings on your Top Lenses lens. Would you mind explaining? We've had trouble internally with identifying the best way to balance # of ratings vs quality of ratings.

Thanks,
Gil

Thanks, Gil.

I originally was just going to list the lenses by the number of stars. But then someone, I think it was Rick Phillips, pointed out that the number of ratings can be more meaningful.

I toyed around with a few idea, and the current combination is simple, in my view, gives a fair amount of weight to both the number of stars and the number of ratings.

I had considered just adding those two numbers or just multiplying, but those seemed to me to give too much weight to the number of ratings.

A couple of ideas on possible ways to improve on what I did (I don't know how feasible these ideas are, but they would be easier for you than for me):

Possibly more weight might be given to --
*Raters who have a longer history on Squidoo
*Raters who are more active on Squidoo, in one or more ways
*Ratings that deviate more from the usual ratings that that specific person gives
*Raters who have lenses on the same or a similar topic (This could go both ways. They would be more knowledgeable, but they would also have some incentive to knock down the competition)

Possibly less weight might be given to --
*Ratings that deviate far from what is typical for a given lens (For instance, if a given lens has had some healthy number of ratings of at least four stars, and then one or two people give it one or two stars.)

While we're at it, I'm wondering about the possibility of allowing non-Squidooers to rate lenses. If some lenses get unfair ratings from their competitors, that might be balanced out by allowing non-Squidooers to rate lenses.

Maybe you could even have two sets. Just a thought.

Handbuilt - Promoted - Recommended; can't think of better attributes to focus on ---- way to go. Maurreen; it was me and thanks for giving credit --- I saw you do that before too; thanks.

KilnGoddess wrote:

Lens rank crashed. Tweeks always trash my ranking.

I just looked over a bunch of your lenses.  I too am a lensmaster who builds lens clusters, that is many lenses on different facets of a topic.  My feeling is there are certain aspects of your lens building that is very undifferentiated from lens to lens and I think that's why you get hurt when these changes happen.

Writertiff wrote:

Someone explain "hand-built"...as opposed to?

...RSS modules?

Well I guess I'm building them by hand. LOL Nothing automated going on that I know of...

Rick, glad to give credit where it's due. And I've linked to your lensmaster page from my main ratings page.

N376, thank you! How kind. I've rated a few of your lenses. But your animated avatar is a little spooky.

all this sounds really nice except for the fact that hand-built lenses have obviously been relegated into irrelevancy, as have number of ratings received, traffic, and all the other commonly assumed things that have gone into rankings so far.

there are certain 'commonalities' amongst 60% of the top 100 list

yes, of course I noticed.

Yes, tiff, sometimes it doesn't occur to people that as a groupmaster of a very large group, the  lensmasters expect me to be able to rate their lenses in some sort of fair and predictable manner.  These things are TOOLS, tiff, features of most profitable lenses on that top 100 list, for example, may assist member lensmasters to improve their own lenses and even make some money.  Giving stars based on effort, content and sales ability of the lensmaster should have some recognizable meaning and offer some predictable results.  I have suspended the rankings I give applying lensmasters until the algorithms are set and some reliability finally re-enters the system.  This much work needs good tools, bubbly charm hardly qualifies.  This way, if the lensrank algorithms take over two weeks  to work out, the rating I give the lenses won't be wasted, but have the predicted effect, which the lensmaster can see right away.  I have some other plans to increase participation and traffic, but those too will need to wait until these things have settled out.  Last December lensmasters could work together to do a few different things and observe the effect they  had on their lens rankings and visibility.  Similar programs designed to actually experience what certain efforts can do will make a marketing effort outside of squidoo to bring new traffic seem like something  attainable, with measureable results.  Seeing results encourages lensmasters.  Having all those things discombobulated discourages them.  So yes, tiff, I for one have REASONS to notice and pay close attention to what is going on with those tools.  Success as a group master is the increasing income of the lenses in the group, utilizing the tools that Squidoo itself provides.

Last edited by Margaret_Schaut (01/24/2007 10:30 pm)

last month.  it seems like a year ago, though!

Actually, I think we are missing one of the best metric, especially for noncommerce sites.

It would be good if we could get numbers on repeat visitors. In fact, I think I'll suggest that in the requested features section.