Jacob
Level: Smitemaster
Rank Points: 3746
Registered: 10-01-2004
IP: Logged

Long time no see!
Hope everyone is doing well.

I have a quick question.
I don't recall who wrote this (I assume mrimer?) but in TSS: Bonus: Developer Commentary: 1N there is a discussion of TSS/DROD's themes and there is reference to removing noise vs cleanroom analysis of gaining knowledge.

Does this refer to a real-life theoretical dichotomy (I can't find any such thing as cleanroom analysis) or this

____________________________
New to DROD? You may want to read this.
My Holds and Levels:

Pearls
Level: Smitemaster

Rank Points: 767
Registered: 12-21-2009
IP: Logged

Hi Jacob,

I don't have a full recollection of this conversation, but I can add my two cents, as this is a topic that's part of my job! I don't have any memory of using the word "cleanroom" so I'm gonna point at mrimer or something, but I'll make a guess regardless, or take a stab at what he (or whomever) might've been suggesting.

The dichotomy here (I think), as a real world analogy has to do with experimentation and the math/science behind it. At a high level, obviously, you try to control for unknowns as much as possible, but this comes in some flavors that parallel what you're asking about.

Primarily, do we consider noise information or not? In one scenario, we may do what we can to excise it from any sort of method of describing phenomena, since it muddies up our models, makes certainty harder, etc. We can approach this from a physical perspective by putting on extremely tight controls on an unwanted variable, or, if possible, remove it completely, even if it's expensive to do so. And from a math/modeling one, we might "block" for it, then cast it out of any resulting prediction/inference. Noise is not information.

If, however, we do consider noise information, in the sense that understanding that noise gives us better ideas on how reliability our prediction is, how influential noise is, that kind of thing, then we definitely go about it differently. Certain things are allowed to vary naturally on purpose, and the kind of math/modeling stuff we use to fit them are different kinds of mechanisms. As long as you have a way to measure/keep an eye on the source of the noise, you can do lots of things with that. . I think this is what was meant by cleanroom. Controlled/monitored natural state. Obviously this has some room for debate.

There are benefits and applications and appropriate contexts for both, and, for what it's worth, I was really happy to see how the obsession of information acquisition came in a couple flavors, and not necessarily great ones, particularly when applied to people, and called into question what it even meant to acquire knowledge, which, well, yeah, neat stuff to talk about and chew on.

Hope you are well!


Post Thought Edit:



____________________________
Hey, w-wait, that's the guy who shamelessly promotes his own...

---

[Last edited by Pearls at 08-31-2021 08:13 PM]

Illusionist
Level: Master Delver

Rank Points: 165
Registered: 05-27-2020
IP: Logged

I would like to use the opportunity and thank Jacob for Advanced Concepts, Uncle Arable's Rooms, Missing Pieces, The Goblin.

Also, Beethro and the Secret Society is one I had fun replaying at least two times.

Beethro's Bad Dream is the one I have trouble with beating, though. Maybe someday I will.

And the Grinfidel saga is my next hold to beat. Wish me luck.

---

[Last edited by Illusionist at 09-06-2021 10:01 AM]