I've gone through this. Here's how you do it.

Start gathering a few popular science books on subjects directly on and also near to your goal. Some people reject popular science books as too light weight, but it does have value. This exposes you to the variety of subjects in and around your interest. You might not have been aware of some aspects of your topic and you are introduced to them here without too much effort. You also learn to associate detailed technical topics to the wider areas where they are used.

Read the whole book. Books are better than random google searches and videos because they will guide you into areas you might not have considered relevant. Broadening your base knowledge will allow you to make a more informed decision about your favorite topics. Once you have a broader and more informed understanding of the topics and areas involved, you are better able to identify your interests, or even switch interests.

That is when you start going into a more detailed dive into your target topic. Follow through and read the whole thing. Again, pick one or more text books or deeper science books. The purpose again is to guide you into areas you might not have considered before.

This time, you will hit lots of technical subjects that you might not know. That is when you go searching for online information, wikipedia, online course videos, Youtube content or other textbooks. For these, you will only need to cover enough to support your primary interest, and you will have a fairly good idea how much that is.

You are not going to go professional with this, but it will be more than enough to keep your interest up and curiosity satisfied.

So, installing malicious software means your information can be accessed? SHOCKING.

Man, here I was hoping helm's deep would reinvigorate LOTRO. And yes, I had even pre-ordered it.

Are the pulses from pulsars visible from all directions, or just from the plane of rotation? If you move far enough, will some disappear and others appear?

I've put together a page showing results as they come in, grouped by domain, sorted by money "earned". You can find it here: http://penny.1889.ca/results

There have been a few quirks along the way. First, most people who hit the site seem to have clicked the links on the penny site while moving them to their bookmarks bar, so I'm just filtering anything at penny.1889.ca entirely. Probable noise, I think.

I'm not merging domains into bundles, because there are some cases (code.google.com/mail.google.com or anything blogspot) where the subdomains are vital information. So in some cases, a site owner would actually be collecting from multiple entries at once.

The default behaviour of the bookmarklets appears to have confused some people... after the site is tagged in the system, I do an immediate redirect back, which may be too quick to see, so the buttons get hit multiple times in quick succession until the user notices a flicker. I've left all those "duplicate" tags in, only because I can't absolutely be sure that the user wasn't doing it on purpose, to show heavy support for the page they were tagging.

Other general stats:
- 1600 entries recorded in 24 hours
- 356 users participating so far
- The heaviest user recorded 105 donations, 177 users recorded just 1
- 594 users paid $0.01, 256 paid $0.02, 751 paid $0.01
- If the first 24 hours could be extrapolated over a full month, Slashdot would earn $260. At this stage, that doesn't seem to be a viable alternative to paywalls and/or ads. But it's early yet.

I also just realized I was calculating these stats off the live dataset over the course of an hour, so numbers may not actually add up. Oops!

Ideally, you'd pretend it was real money, if only for calibration purposes, and act accordingly. This experiment is gathering information broadly, but also for you, specifically. So you can see what you'd spend, if you were spending.

Wow, that pretty much articulated everything I'd have wanted. The only thing I might change is the notion that you can still "like" a page on a non-participating website, so latecomers aren't punished. The tip distributor could hold the funds until they're claimed, meaning you're supporting the content you like, just maybe not as immediately as you might have liked.

My thinking on the subject has evolved since my comment a bit further up, but I'm still not sure that's something we necessarily want to track. I mean, I definitely see your point, but in the end it would be like "5,000 people hate digg.com." I guess it shows that 5,000 people were there, COULD have paid but didn't, but it's a bit like throwing a piece of chewed-up gum into a street performer's hat. It says "you suck", but it's not entirely productive.

On the other hand, if we were tracking individual pages instead of domains, that's something... but then this system becomes a focus testing tool, which is a bit sideways of where I wanted to be. It might be more useful to the site owners, but then you'd have obvious pandering all the time.

Hmm.

I'm approaching it more like: slashdot.org got $3.44. 60% of those were 1 cent, 30% were 2 cents, 10% were 3 cents. The people who didn't pay didn't find anything worth seeing, or they forgot they had the bookmarklets, or they hated it, or... lots of things. I suppose it could be interesting to track explicit 0-cent transactions, but that just feels a bit negative for what I'm trying to do, rather than constructive. Though I definitely take your point.

I sat in on a meeting last year where a company was trying to convince a potential advertiser (hand-picked, too) to put ads on their site. The cost was probably about 10x the norm, for the audience. The advertiser said: "Why would I pay that much when I can blast the world for less?" To which the site owner said: "But this is a PREMIUM ad. Premium!" Said the advertiser: "How is it premium, aside from costing me a lot of money?" And that was pretty much that.

Still, companies will gladly spend ten times their online budget, putting an ad in a newspaper for a single day, where you have no idea of impressions, no way to measure follow-through, and really so much friction it's a fair bet less than 5% of people who saw it, acted on it... I still can't quite grasp why print advertising hasn't crashed and burned in the last decade. Leprechauns or something.

Damn, I think I even read that, way back when. I built a lot of my brainspace around his notions of the double standard, valuable and value-less content. I shall drop 3 cents on that now! :)

Still, I take your point. At the same time, I had this really distressing moment a few weeks ago with some junior developers who were working on this-and-that, and I said to them with my wise, old experienced voice: "Oh you kids, that'll never work. We tried that back in '99 and it fell flat." And then they showed me that no, really, it does work these days... they just needed reality to catch up with our dreams.

So yeah. I guess that may have stirred up some of my old "let's fix the world!" idealism again. Probably a bad idea. It can only end in tears.

No, it's entirely a tip jar approach, so if you read it and like it, you can opt to spend 3 cents to show your gratitude. Tip jars exist, of course, but because of fees etc, their definition of "microtransaction" is usually at least $1, which almost defeats the purpose. So what this is doing is saying: "Did you like what you just read/saw/heard? If so, how much?" And that's it. Down the line, someone else can figure out how to turn that into countless riches, but right now, I'm just interested to see how the impression of the ideal shakes out.

I'm so scientific sometimes.

Indeed, Flattr is a really great idea. I kinda wish there was a way to, as you say, support A more than B, but the predictability of the billing would make me sleep better at night :)

Actually, maybe I just didn't explain it properly. There is a zero-cent option, which is to just not click anything at all. This isn't something mandated or integrated into the sites, it's client side, so it's 100% voluntary, and only worthy content stands a chance of getting rewarded. "Worthy" being a very subjective concept in this case.

Yeah, I admittedly only put minimal effort into building a bare-bones tool. Not entirely sure how to make it work in the Android browser, actually. Interesting predicament, especially since I'm guessing lots of people do a lot of their reading on mobile devices...