Re:The Death of the Photo Studio
by bryanrasmussen on Wednesday July 29, 2020 @07:52AM
(#23985616)
>Ok, this photo has +.8 attractiveness, but -.6 smart. Ok, not sure what to do with that information
Use that photo for dating, but not for potential employers, unless you're searching for a modeling job - just my guess.
Re:The Death of the Photo Studio
by PaywallBuster on Wednesday July 29, 2020 @09:16AM
(#23986283)
You can choose different categories (work, casual, dating) but all of them have 3 characteristic which you're voting for.
If I remember these 2 are part of dating
Re:The Death of the Photo Studio
by TheOtherHobbes on Wednesday July 29, 2020 @07:23AM
(#23985479)
Perhaps there's an opportunity for a startup to automate and simplify this.
Photofeeler will go that way: https://blog.photofeeler.com/photofeeler-d3/
You probably have already thought about this—and it is a bit of a derail—but it seems to me that a "veriface" service that provides attestations, or some combination of verifications, as to the accuracy with which a photo represents someone could be useful.
"Yeah, that's him… 10 years ago."
Just to share my 2c: I can't get over how strange the pricing structure for photofeeler is. I've come to the site a few times with my credit card ready, willing to drop a few dollars to get results.... but the abstraction of your pricing model is just confusing enough that I've dropped off the funnel each time.
Ok, so I need credits or karma to run a test. So I answer a few photos, get my karma to medium, and leave the test over night. The results are often inconclusive - my photos are consistently rating as "meh->ok" across the board. Ok, that's fine, I'll try another. Upload another, get more karma, leave that over night. Ok, this photo has +.8 attractiveness, but -.6 smart. Ok, not sure what to do with that information, but it's intriguing. The results are close enough, that I'm curious what the margin of error is if each is getting ~10 votes.
Ok, well maybe this is just the downside of the freemium model. Let's take a look and see what I can buy. 40 credits for $10. Ok, so 1 credit is .25c. Cool. What's a credit get me. A CREDIT GETS ME ONE VOTE! Ok, hold on, that can't be right.
Let's think about this backwards. How many credits do I need? So I probly want 15-20 votes on any one photo to get enough data to be conclusive about the results. Sure. And I'll want to run at least 3 photos. So, I'm looking at $15 per test to know which of 3 photos is preferred.... but realistically if I want to get rid of my own biases, I'd be testing 5-10 photos... Hmmmmm.
At this point I do some work to get some free karma (note: if there were hotkeys for voting it would make a WORLD of difference).... leave the test over night. Look at it the next day, say "meh, results are inconclusive based on sample size" and forget about the site for 8 months.