This could lead to some delightful or awkward realizations. π
Here are my results:
P.S. Likecreeper is still one of my favorite Instagram apps. So troll.
@tylerpearson what was the inspiration here?
I know lots of us are nosey and want to know who our #instafans are haha...
Did Best Nine of 2015 have a part to play in the development of this?
@bentossell Actually about a year or so ago a friend commented to me that it would be interesting to see who liked her photos the most... so I wrote a quick script that found it using Instagram's API. A few more people asked about it, so I figured it might be worth spinning into a little microsite to make it easier for other people to use.
The timing was due to me having some free time around the holidays to finally finish it -- the first commit was back during the summer, so not related to the Top Nine.
Interesting thought: does the favorite measuring start all of your friends with your 1st photo?
It'd be more interesting if each friend's first like was the start of their best buddies test.
If you've been on Instagram for years and a good friend recently joined and has liked all your photos since they joined, their statistic should be 100%, not some small number.
@rometty_ It goes back up to 60 photos, which was kind of the sweet spot between taking too long to process and having enough data to do a decent-enough analysis.
That setup would be interesting -- main things that would make it a bit more difficult to do is:
1. I don't think Instagram's API provides info on when you connected with another user
2. It would require some sort of weighting for likes because one new friend liking one photo would be 100%, but that shouldn't be valued the same as a friend that's seen 50 photos and only liked 45 of them (100% vs 90% but a big difference in photos viewed).
@tylerpearson@rometty_ totally understandable. I feel like every project has a perfect version that requires ridiculous diminishing returns. You really hit the sweet spot of complexity and app speed! :)
Great idea! Processing my account now, though I think your server might be overloaded (or my account too big, doubt it). It keeps hanging in processing!
@dshaw_ Sorry about that, I was busy with the day job and didn't realize how much traffic this was getting. Cranked up the servers and changed a few things, so the queue should be cleared out soon.
"Share or add your email to get the full list of results!"
That's such a bait and switch. I already gave you my trust by handing over OAuth access, now you want more of my data instead of holding up your end of the deal. This is not ok and we should discourage this kind of behavior. It's a marketing website for your main product and you already had my attention, but then you forfeited it.
Nice and simple UI. I noticed this is similar to the "Admirers" feature on Crowdfire, which provides a whole suite of Instagram analytics. Any plans to expand the data you provide?
@adammash At this point, no, I don't have any other plans for building more with this particular microsite. I'll likely open source it on my Github page some point soon. Also may do some analysis on the data that's gone through the app for a blog post.
It sounds like a great idea but I'm getting the following:
{"code": 403, "error_type": "OAuthForbiddenException", "error_message": "You are not a sandbox user of this client"}
Keep getting this error when I attempt to authenticate - "{"code": 403, "error_type": "OAuthForbiddenException", "error_message": "You are not a sandbox user of this client"}"
@moravtchik Agreed, I think it's fascinating to see for Instagram and elsewhere. I've written scripts for myself before just to see more stats than what the networks normally provide.
Product Hunt