[Transparency request] Publish the RNG engine of the lottery system

[dutiona]

In computer science pure randomness does not exist. It is emulated with random number generator (RNG) engines whose performances can be very good or very bad, depending on how it is initialized, implemented, etc.
For instance, a badly initialized RNG engine will draw numbers that are NOT random at ALL and can even be PREDICTED.

After this MASSIVE screw up that we are witnessing on the housing system, I believe the player base have the right to know HOW the "fairness" of the system was actually enforced by Square Enix in FFXIV.

We want to know :
- what RNG engine did SE choose (at the very least Mersenne Twister??)
- how the engine was seeded
- a graph of the distribution of the engine actually picking 1B numbers : show the variance and the standard deviation.
- we also want the actual result of the lottery (all the tickets that were drawn) to prove the lottery was fair (graph of distribution, var and std)

As far as I'm concerned, I do not believe the system was fair seeing how it was buggy. I think I'm not alone in this case.

tl;dr: I want PROOF OF FAIRNESS OR ROLLBACK.

[Olivia0x25]

They did a random number between 0 and list length :^)

That way when someone bid with zero competition, list length says 1, but the random number is between 0 and 1. 50/50 chance it's 0 or 1. :^)

[Shibi]

...what RNG engine did SE choose...

I'm going to guess...

Eeny, meeny, miny, moe
Catch a tiger by the toe
If he hollers let him go
Eeny, meeny, miny, moe
Yoshi told me
To pick the very best one
And you are not it
so... 0

Disclaimer, I got a 0 as the only entry. Yoshi hates me.

[Nythia]

Ironic that in most programming languages a single line of code would have prevented what has happened.

while result is equal to zero, re-run lottery roll.

[dutiona]

Yes, but what tells you the number 0 is not more likely to be generated than the number 1 or 2 ?
That is what I'm asking here ^^ . Proof of that fairness.

[AnnaUK]

I guess it's to slow down those who use alts to bid all over the place?

[Iscah]

Yes, but what tells you the number 0 is not more likely to be generated than the number 1 or 2 ?
That is what I'm asking here ^^ . Proof of that fairness.

Do we need proof of that fairness?

Admittedly I don't know what the numbers are on how many successes there were and how many got the zero glitch, but the number of complaints with the system is naturally skewed to be 100% made up of people who got zero. The people whose lottery result was a real number have no reason to speak up.

The prevalence of the glitch would probably (I speculate, not a maths expert) be best worked out if you could gather up a list of all the instances where a zero occurred, map it against the number of people bidding and see if it matches the random-chance expectation of happening mostly to single bids (50/50 chance of occurring) and not so much to contested lots where the dud outcome only has one chance out of many.

[Moonlite]

Do we need proof of that fairness?

Admittedly I don't know what the numbers are on how many successes there were and how many got the zero glitch, but the number of complaints with the system is naturally skewed to be 100% made up of people who got zero. The people whose lottery result was a real number have no reason to speak up.

There is also the question of chance though. They didn't know they were bidding against the system as well. Maybe a FC bids multiple times on one house only then instead of spreading out bids, since they were bidding against +1 on every plot.

[dutiona]

Do we need proof of that fairness?

Admittedly I don't know what the numbers are on how many successes there were and how many got the zero glitch, but the number of complaints with the system is naturally skewed to be 100% made up of people who got zero. The people whose lottery result was a real number have no reason to speak up.

As far as I'm concerned, if they can prove that the system did fair draws then I'll consider every plot winner legitimate. They won the lottery against N+1 bidder, N being the actual number of other tickets than theirs and +1 for the 0 ticket.
If they cannot prove this fairness, and the graph shows for instance that the tickets 2 or 0 was drawn a lot more that the other tickets then I (and we should all) demand a rollback because the system was not fair at all.

[Menriq]

As far as I'm concerned, if they can prove that the system did fair draws then I'll consider every plot winner legitimate. They won the lottery against N+1 bidder, N being the actual number of other tickets than theirs and +1 for the 0 ticket.
If they cannot prove this fairness, and the graph shows for instance that the tickets 2 or 0 was drawn a lot more that the other tickets then I (and we should all) demand a rollback because the system was not fair at all.

I didn't win what I bid for, and I was #1. I doubt your "demand" is going to happen. Assuming you have some development experience, I'll let you figure out why you aren't going to get your rollback.