I've done this exercise and generally what you want is a combination of your suggestion (which is elaborated by 'polyphora[0]) and 'p1necone's suggestion [1]
Polyphora mentioned CIELAB as just one example, and it's a good example. I believe state of the art these days is Oklab[2], talked about here[3]. I'd like to pull out a comment from 'jiggawatts in that discussion:
> This is a tour de force of colour theory, and should be mandatory reading for anyone serious about computer colour!
I completely agree.
With regards to 'p1necone's suggestion, k-nearest neighbors is one simple and relatively easy way to separate the colors into bins. I've only done this on a single image, but with multiple images maybe you could also k-nn bin the resulting colors from each image and only return bins which have multiple members.
Some really important things that are built into it are:
- Everything is given in "today's dollars" and that's done by reference to a salary index, NOT a price index, to reflect increases in community living standards.
- It shows some indication of the income level you can expect during retirement given your lump-sum retirement benefit.
[1] I was one of the actuarial staff that contributed to it
> They are a public traded company, and for us casual users of Twitter the only change that has been in the talk the last couple of years has been to increase the character limit.
I think that's a bit hyperbolic.
But, that being said, it annoys me that they haven't even put enough effort into this change to update the documentation.
> Note that the current best definition of the algorithm for counting characters on Twitter is described in the page on the twitter-text Tweet parsing library. This documentation will be refreshed shortly to reflect the most recent changes.
This is not a product that I'm interested in, but may I provide some feedback on your website?
It took me a little while to understand what Clickrouter does. The very first line, "Monetize Outbound Clicks", contributed a lot to my confusion. It made me think that maybe you were just a frontend to e.g. Amazon's affiliate program.
It was only when I read Step 4 of "How's it work" that I understood what it did: "Watch ClickRouter route each click to your best (i.e. most profitable) merchant network every time !"
I would've been much happier if the very first line was something like "Maximise your revenue by automatically using the best affiliate links."
Straight away I know _why_ I should be interested in your product, and how it works.
I don't (yet) have an opinion on this, but it's important to note the other side of the coin...
> Hope everyone's paying attention to the identity of these investors behind this horrid behavior. Travis is not without his flaws but he created an awful lot of value for them.
He's been very well compensated for the value he's created. You don't keep a job because of what you've done in the past, you keep a job based on what you can do in the future.
Firing founders was tried in the nineties and it always resulted in failure. Guess the Uber VC's need to painfully relearn that lesson?
If Travis had been smart enough to create two classes of stock like Mark Zuckerberg and the Google founders he'd be calling the shots, instead he's out of a job.
Travis was calling the shots. That's why he's getting blamed for the alleged toxic culture (which includes not just sexism, but mistreatment/underpayment of drivers, among other things).
There are many things in life which technically give a perversive incentive to murder or commit other crimes, yet the perverse incentives aren't a show stopper.
One example of this is "I want your Mercedes, I will murder you and steal it."
Another example is a will and testament. Sure, sometimes people murder because of these contracts, but as a society we've done a good job of mitigating this problem.
The most naive of interpretations (e.g. averaging each of the R/G/B values) is generally not what we want.
Would be super cool to add some different averaging schemes to this.