I think this would boil down to rules interpretation in the end, and I would hypothesize that your interpretation is incorrect, thanks to the blockchain. Now unless I missed something, @bycoleman has not specifically confirmed or denied anything that could be even remotely construed as an attempt at a rules interpretation. But looking at the blockchain, one can see that he found and determined the resteem winners through position 50, about 19.5 hours before you posted your claim that there were only 48 "valid resteems" -- a claim that I certainly dispute.
I could speculate much further and also go into detail on how I interpreted the rules, but I'm not sure that would be such a good idea until after the contest has officially ended.
Feel free to check it out for yourself. You can verify this yourself on steemd.com and by checking users and following lists manually, but it would take a lot of time to confirm that SQL query is returning the correct results. Check it out - here are the two lists side by side - on the left are "valid" resteems (resteems by users who are also following @bycoleman). On the right are all resteems, including those where the user did NOT follow @bycoleman.
I checked the first name that appeared in the total resteem column but did not appear in the valid resteems, and it is correct - he is not following @bycoleman:
Ok, if you couldn't infer what I was trying to get at then I'll have to provide a detailed rundown. I didn't want to do this right away because there are probably lurkers who are going to get some ideas out of this and try things before the contest ends. Before I get into that... I checked manually from geeyang15 down through hasnain06 and have confirmed those timestamps; that took me about a minute per person on average. I don't need to check further because the question of what happens with resteem position #5 will cascade down through the rest of the lists.
My position is that because the blockchain has recorded it, all resteems are valid and all resteemers' positions are locked, regardless of whether or not they have followed, i.e. the list on the right is the correct list to use, and the only list necessary for identifying potential winners. The rule "You must follow me [@bycoleman] to be able to win" is an eligibility requirement, and if someone ends up in a winning position but has not followed by the end of the contest, they would become ineligible and forfeit their prize. It would then be left up to @bycoleman to decide what to do with the forfeited prize money.
The most similar real-world scenario to my position would be that of U.S. TV game shows, in which at the end of every episode viewers can see a disclaimer stating "Contestants must meet eligibility requirements to receive cash or prizes" or other similar language. If a game show contestant were found to be ineligible, you can't undo the fact that said person appeared on and participated in the show, but you can refuse to pay any of their winnings. Likewise, the fact that someone resteemed can't be undone, but if they would land in a winning position and ended up ineligible, @bycoleman can absolutely refuse to pay.
Obviously dante01 didn't follow, and he should be a winner if he remembers to follow before the contest ends -- that much we can agree on. Now if he were to be DQed for having a follow timestamp later than his resteem timestamp, under my interpretation of the rules that would be an eligibility issue with a single person. Under your rules interpretation, there would be people who could benefit from moving into one of the winning positions by making the argument that cryptoeater, arctan, cgrave, and others also resteemed before following, and therefore should also be DQed alongside dante01 and removed from your list on the left. It's an argument that certainly sounds absurd, but would be technically correct.
Your rules interpretation also presents another big problem -- if you're going to add "invalid" resteems back to the left list because they followed several days after resteeming, then it should also work the other way around -- a person could voluntarily DQ themselves by unfollowing @bycoleman and then they would need to come off the left list. Now I'm not saying that this would happen, but it's theoretically possible that a group of friends (or a single person with multiple accounts) would manipulate their follows to ensure that whomever they favored to win would end up in a prize-paying position. Not only would the manipulation of follows have the potential to reflect poorly on @bycoleman (to say nothing of how it could affect public perception of a much larger company), but in a higher-stakes situation (which I'll get to next), I'm pretty sure that this would enter legal territory involving tampering with the operation of a contest.
So the next thing I thought about was a higher-stakes scenario: suppose that instead of @bycoleman running this contest as an individual with a 150 SBD total prize pool, a major corporation ran (or more likely sponsored) an identical contest but with a 15,000 SBD (or USD equivalent) prize pool. That's one big prize of 10,000 SBD and five smaller prizes of 1000 SBD -- 100 times the current stakes. Now, there will almost certainly be a major compliance company like Enteractive Solutions Group (ESG) involved, and they would be poring over every little detail (timestamps, follows, resteems, etc.) in addition to having a set of legally binding official rules. And given two possible ways to administer the contest -- my rules interpretation or yours -- which one would ESG go with? Would it be with: (1) a set of rules that is straightforward, has a clear-cut method of determining winners, and is as tamperproof as can be reasonably expected, or (2) a set of rules that exposes the contest to possible
manipulationtampering and/or coordination by the entrants, with said tampering/coordination being implicitly encouraged? I'm confident that ESG wouldn't consider administering such a high-stakes contest using your interpretation of the rules -- there's just too much potential for bad publicity and legal actions to arise, both for ESG and the sponsoring or co-running company involved in the contest.At the end of the day though, @bycoleman is running this contest and whatever rules interpretation he has -- whether that's yours, mine, or something else different -- is the rules interpretation we both have to accept and respect. The reason I've gone into this level of detail is because as the title of the contest post states, the contest is an experiment for the business sector, and the opinions I've stated here reflect what I believe would happen with companies being involved in a contest like this. Hopefully @bycoleman gets a lot out of the feedback that we've both been providing.
Final formulas have been posted.
Thanks for all the help @doughtaker
You know what is funny this hadn't even occurred to me, though of course it makes a lot of sense. I choose the alien eyeball as my profile picture in no small part because my thought processes are very often alien, as in they are totally outside of the "obvious" interpretation. Often this actually is a good thing (in creative endeavors) but in this case I think your logic is both solid and more appropriate to the situation. Obviously @bycoleman agrees :) I think part of why I didn't even see this as a possibility was your interpretation meant the original rules were incomplete. There was no provision for what to do if one of the winning resteem count spots was a user who did not follow @bycoleman. He has updated the post with these rules now :) I am actually glad you are in the #25 spot, it was not my intention to actually claim a prize spot when I activated a few of my alt accounts and followed/resteemed - I was just trying to show @bycoleman why a contest of this nature was kind of a fool's errand on Steem blockchain :)
Cheers - Carl
Indeed -- there is a lot of feedback here. I was unaware of it entirely since it did not hit my replies inbox.
Thank you VERY much (both of your) for taking the time to analyze this so deeply.
Please see my initial response addressed to Carl's reply
Per the contest language I posted.
I am looking into the "un-follow" circumstance, but my current belief is the only way to ensure fairness is that an "un-follow" will not change the re-steemer that has won the slot but only if they are eligible to collect the prize. Any prize positions that cannot be claimed will either be given to another contestant randomly or possibly distributed to all who had no other winning slot.
With that said, the initial statement on the 51 restreemers still holds true, but we do not know yet whether they can collect the prize until they meet all the criteria.
Your thoughts are very welcome. And yes, this was certainly an experiment.
I am not entirely sure what you are saying here. There are two requirements to participate in the contest - a valid participant has to follow @bycoleman and resteem the post. I shared the SQL query above, you are welcome to check my work. I returned all users that had resteemed the post AND were following @bycoleman at the time I ran the query, and I returned the timestamp (UTC time) of the resteem. This is a matter of public record. The only thing that would seem to be open for interpretation is whether or not someone who initially failed to follow @bycoleman (but did resteem the post) would become a valid entry if they did realize their error and followed @bycoleman before the post pays out. I would assume that yes, they would be a valid entry then - and if that assumption is correct, the payout positions that I listed above would change. There were several individuals who were among the first 25 to resteem, who did not follow @bycoleman. Cheers - Carl
Your assumption is correct as long as they meet all criteria at the time of the contest close, then they are potential winners.
See the other replies made in the last few minutes also.
At the time I posted the 51 resteems, I was looking only at the number of resteems and not as to whether they met all other criteria. This was a premature announcement indeed.
My goal is to make this a fair as possible and minimize manipulation potential, which is huge due to people having access to multiple accounts. I am in the progress of running a query directly (thank you @carlgnash for the head start on this.)
Also, I expect to release the final random formula on the 6th day of the contest.
Thank you everyone for the great help on this.