View Full Version : An idea... an experiment, if you will

03-10-2011, 22:34
Apologies if this has been posted many times, but it is in regards to a certain bit of information retrieved from the GDUK 2011 design studio seminar. During the seminars, the many people who went were informed that the GW (apparently) listen to information from the many people who give feedback.

Now this got me thinking... i think it is time we find out whether this is true or not. This will be treated as a proper experiment.

The aim is as follows: To determine whether the dependant variable (GW) can be affected by a independant variable (feedback)

The method is (possibly) as follows:
1) We decide on a simple request (No S-word or new armies, maybe a model for a army that needs updating)
2)We get as many people as we can to pass on the message to their friends in the hobby, and everyone informs GW of the same request (not mass spamming mind you! Just giving feedback. This is an experiment, not a chance to whine to GW)
3) We wait and see (I have no idea how long it might take, and i appreciate that it does take a long while for a model to go from the drawing board to production)

For this, the following Hypotheses are given
Hypothesis: GW does react to the feedback, resulting in the measurable alteration to it's product line or service
Null Hypothesis: GW does not react to the feedback, resulting in no measurable alteration to it's product line or service

Now, at this moment at time, i can see the following outcomes:
1) GW does not listen, in which case the null hypothesis is proven
2) GW does listen, but not to a measurable extent, in which case the null hypothesis is proven
3) GW does listen to a measurable extent, in which case the hypothesis is proven

It should also be noted that there is a potential extraneous variable, in which the request of the independent variable is already in production.

I am not saying this is what we 100% must do, all i am saying is that it is an interesting chance for an experiment to determine whether we can actually have a say on what goes on at head office, in particular the design team. If no one agrees, then let this message fade to memory.

As an aside to any forum administrators, i give consent for the locking and/or deletion of the message if it is found to be inflammatory, offensive or in violation of the rules and regulations of the forum

03-10-2011, 22:41
For the sake of science, I'd do it. What did you have in mind?

03-10-2011, 22:49
I'll be honest, i don't really have many ideas what to request. I'm open to any ideas for the independent variable, as long as they are realistic, as put forth in part 1 of the method. If we start off with something that is known to be rejected (a certain S-word springs to mind), then there is not much point to this, but if we start small, say with something for one of the armies of the major three systems and see what happens after a set time period, with any evidence in the form of an actual reply from head office, or the design studio, being posted to the forum.

Of course, if anyone could post a email/ postal address for contact, i would be grateful.

03-10-2011, 22:51
There are a couple of issues that make this sort of experiment flawed in its design and aim.

Provided this experiment does happen and feedback arrives at GW HQ in a timely manner, the result could be seriously affected or obscured by numerous factors. Say GW hear you and want to respond to the feedback, they might not have the opportunity in their schedule to fit this in for a few years or so. Sure, we don't know their release schedule but I'm sure they've got the next 2-3 years planned out.

You would also need to pick something so obscure and improbable that there would be next to zero chance of it already being in the pipeline and therefore appearing coincidentally (ie. you wouldn't ask the community to request new Necron Immortals or a new Chaos Space Marines Codex as there's a good chance they're already in development).

Finally with the long development process that GW practices, you need to have the patience to see this experiment drifting silently over a good few years and not forget about it/lose interest.

What this experiment needs for a positive outcome, I believe, is GW to be candid and say 'yes, we heard you wanted this and we are now making plans to make it happen with immediate effect', and if I know GW, that ain't happening any time soon.


03-10-2011, 23:10
@Trick: I can understand where you are coming from.

It would be interesting just for an insight into whether or not the information in regards to whether or not GW respond to fan feedback, as the original idea was in relation to the release window for the many products produced by GW, in that a consensus, also gleamed from the studio seminar, was that the first information of an upcoming release, which has now been decided by GW that it will be one week before the actual release date. This may be felt by some that it is insufficient, in that people are unable to plan their purchases, which is necessary in the current economic state.

Anyways, that is my personal opinion which lead to the creation of the hypothesis behind this experiment.

In regards to the 3rd point, i was going to propose that a 2-year time span for the experiment, with the data gathered being any actual response from GW, and if after 1 year, 11 months there was no data gathered, then the null hypothesis would be accepted, and the experiment concluded

grey knights rock
03-10-2011, 23:14
I would suggest a tervigon. I have heard a lot of people upset that a model for one of those isn't out there

04-10-2011, 00:49
Genestealer Cult Rules and Models!

Come on, most of the models are already there, we just need som hybrid models and a workable list.

Even more off the wall, how about Tau Genestealer list? (I can dream...)

Although I doubt GW would do it, the impression I get is they treat the design team as the Golden Goose, and don't want any external influence to influence what goes on in there.

04-10-2011, 01:00
you could just send a polite letter to their headquarters asking if they take notice of consumer requests and ask them a hyperthetical question such as if 100,000 people requested that they make a 40k siege supplement(picked out of thin air) to see how they reply.

it may not be a genuine or straightforward reply,
they may not even reply

but it is a lot quicker that the afformentioned idea.(if a bit less reliable)

04-10-2011, 01:08
As a scientist I find this very interesting and would be willing to help, but I think your study has one major flaw: There is no possible way to link the outcome to the cause in the case that the change is made. ie - correlation does not equal causation.

Lets say we all request a new tervigon, and in 2 years a new tervigon comes out. How can we prove which of the following happened:

a) This model was already slated for design and release in this schedule as a tyranid wave
b) A tyranid wave was planned, but GW listened to the feedback and included a tervigon
c) The model has already been designed and has been sitting around waiting for other stuff to finish to release a wave of mini's.
d) The requests get filed\thrown out, GW continues with their normal plan which just coincidentally includes a tervigon because it needs a model
e) GW completed listens and releases the model

Without having a way to determine if the cause affects the outcome, or if it is coincidence or correlated in another, albeit unrelated, way - I don't think any useful data can come of this.

Edit: For those not familiar with correlation not equalling causation, a typical example is ice cream sales and shark attacks. While they correlate very well with each other, they do not impact each other whatsoever.

04-10-2011, 04:15
Tervigon is supposedly in the works.

Something small, relatively easy to do, and maybe already in a codex.
Kosarro Khan Finecast to put on a bike?
Plastic Librarian?
Inquisitor Valeria?

04-10-2011, 04:29
The best thing to do is write them a physical letter.

04-10-2011, 05:38
Increase the cost of the vendetta through an Errata (or remove twin linked). They've already done the FAQ, and rarely do an errata to actually change rules unless it's updating an army to current wargear (ie the Dark Angel and Black Templar update to storm shields and power of the machine spirit), much less on points. Also not entirely specific so gives them freedom to feel in control.

Not hard to do, probably not in the works, and would be sweet.

correlation does not equal causation.

Though yes, that is true. Which is why it has to be obscure (not that mine really is, but GW isn't known for doing it for anything).

Delta Echo
04-10-2011, 05:39
I agree w/ linuvian on this one. A Tervigon model request or any other models missing from other armies would be a poor choice as test example for reasons already stated.

I would propose something smaller, maybe something not even physical. How about a rule change for one particular model from one particular army? Its a small thing, doesn't involve any investment on GW's part such as designers, molds, inventory, etc. A mere paragraph in the form of an updated Errata/FAQ would be all they have to do.

Lets keep it simple and avoid the highly controversial rules or anything that would drastically change the game, or anything that has already been covered by existing Errata / FAQ. How about a special rule addition or option to the Tyranid Lictor (and Death Leaper) allowing it to assault the turn it deepstrikes and a point increase of 10 points pre model over its existing cost?

A) It fits the fluff. This brings the model back in line with what it once could do (back in 2nd or 3rd edition I believe).
B) Its not overpowered (especially for the points). Compare to Marbo or the Mawloc's Terror from the Deep special rules.
C) A lot of controversy surrounds much of the current Nid dex, but relatively very little of it involves the Lictor, therefore even if GW was considering redoing the Nid Errata back from 2010, the likelyhood of them unilaterally adding a special rule for the Lictor is extremely unlikely IMO.

As clearly evident on this forum and other, the Nid community is
not overtly happy and are a very expressive about their feeling with regards to current dex. Its my guess they would be strong proponents for you with this experiment.

04-10-2011, 08:06
@ linuvian, i agree with what you are saying, in that there is no way to determine whether a model release, using as an example the tervigon, would have actually been caused by the feedback given (the independent variable in this case), as it may have actually already been in the pipeline.

I think that sticking with something non-physical may actually be a better idea, as there are too many potential problems with requesting a physical model, due to the long design period, and the aforementioned pipeline issue. This would also be the case for conversion of a model from metal to finecast, as that would also fall under the previous.

04-10-2011, 12:41
Just because GW (or any other company) listen to feedback doesn't mean they're obligated to act on it, they're subject to factors other than 'a bunch of people off Warseer said to do XYZ'.

I mean, what are you expecting to happen here, People off Warseer tell Games Workshop to make New Model XYZ, Games Workshop don't release New Model XYZ, people on Warseer foam at the mouth and yell that Games Workshop is LYING TO THEM LIKE THE EVIL EMPIRE THEY ARE

40 klicks below
04-10-2011, 13:07
I mean, what are you expecting to happen here, People off Warseer tell Games Workshop to make New Model XYZ, Games Workshop don't release New Model XYZ, people on Warseer foam at the mouth and yell that Games Workshop is LYING TO THEM LIKE THE EVIL EMPIRE THEY ARE

Couldn't have put it any better. Pretty much this.

But yeah.. for simple "scientific" value, the OP's proposition lacks a control group. To truly *cough* "prove" *cough* your "null hypothesis", you'd need to test the exact same method of "feedback input" among a few other companies (of similar size). Only if other companies respond to the same independent variable as predicted while GW does not, would you even approach something possibly, remotely resembling a reasonable case for your "null hypothesis".