Return to Unfiction unforum
 a.r.g.b.b 
FAQ FAQ   Search Search 
 
Welcome!
New users, PLEASE read these forum guidelines. New posters, SEARCH before posting and read these rules before posting your killer new campaign. New players may also wish to peruse the ARG Player Tutorial.

All users must abide by the Terms of Service.
Website Restoration Project
This archiving project is a collaboration between Unfiction and Sean Stacey (SpaceBass), Brian Enigma (BrianEnigma), and Laura E. Hall (lehall) with
the Center for Immersive Arts.
Announcements
This is a static snapshot of the
Unfiction forums, as of
July 23, 2017.
This site is intended as an archive to chronicle the history of Alternate Reality Games.
  

The time now is Fri Nov 15, 2024 6:19 pm
All times are UTC - 4 (DST in action)		 View posts in this forum since last visit
View unanswered posts in this forum
Calendar
 Forum index » Meta » Puppetmaster Help
Email Trailheads Viability
Moderators: imbri
View previous topicView next topic
Page 1 of 1 [11 Posts]  
Author Message
Rogi Ocnorb
I Have 100 Cats and Smell of Wee


Joined: 01 Sep 2005
Posts: 4266
Location: Where the cheese is free.

 		Email Trailheads Viability
An experiment to determine whether they can still work
I'm starting this thread to report on the outcome of an experiment I performed to assess the efficacy of trailheading a game using email. Questions about the puzzles, themselves should be addressed in the original thread.

It has been slow in coming as one of the rules I set for myself was to allow 5 days to elapse without a reply from any of the players queried in that step before going on to the next level of inquiry.

I will attempt to report on the general reply rates and demographics of the respondents for each phase.

Required Elements
The original email should attempt to present several common "flags" to SPAM-checking routines. The five primary components of this were:
An email alias that is obviously not a name
Code:
"clue less"

A Subject line that is both an incomplete sentence and includes "SPAMmy" terms
Code:
Warmest Hello

An irresolvable return path (gmail works for this)
Code:
warmesthiSPLATgmail.com

A message in the body that is mostly gibberish
Code:
*Th%i%s! i!+s^ w*a!y& mo&!r!e t%ha%!n* I c$an& ha%nd!!le^` on my own.
Won't you please help?

A binary attachment of a type commonly flagged by SPAM routines
Code:
Diversion.pdf


It should also provide clues that it may be something more if looked at further:
The fairly obvious text buried in the gibberish
Code:
This is way more than I can handle on my own.

The obligatory request for assistance
Code:
Won't you please help?

The naming of the attached file
Code:
Diversion
(Though this was also semi-supportive of this being only a puzzle and not a game.)
The use of nine different versions of the file was an element that didn't address the initial "SPAM/Not SPAM" question, but was done to help those "on the fence" determine that it was something other than random SPAM, once the fora was consulted, encouraging them to post about their "part" of whatever it was.

Procedure
Send the email out as a Blind Carbon Copy to a reasonably sized audience in nine parts with each of the nine groups having a similar mix of experience ranging from newbies to old-timers. My population was made up of folks I have had email communications with about game related issues in the past who probably wouldn't mind receiving something like this and the entire list of people in the "Contact The Interaction Thread" who gave an email address for contact. In all that worked out to 61 people. The basic demographics of each group were:
Code:

      Number Avg.  Avg.  Avg.
      of     years Total Posts
Group people on uF Posts /day
1     7      2.24  620   0.46
2     7      2.48  674   0.85
3     7      1.98  415   0.63
4     7      1.88  151   0.30
5     7      2.13  725   0.71
6     7      1.96  344   0.59
7     7      1.81  388   0.45
8     6      1.95  525   0.60
9     6      2.06  179   0.44
10    5      0.64  339   1.83  (These are what I termed "Joiners". People
                                who weren't included in the original email.)

I let the "game" proceed for 30 days before starting to inquire about what happened to the unknown quantity, simply answering questions, providing clues as appropriate to the emails I received in the interim. In my attempt to be as responsive as possible, which was another constraint I set to minimize player apathy due to poor communication, I set-up the IG gmail account to forward all email my primary gmail account to serve as a notice to log into the IG account and answer the inquiry. In my haste, on two occasions, I inadvertently forgot to switch accounts and answered the email(s), giving myself away. I did the same damn thing on celina's Secret Santa quest and thought I had learned not to do it in the future. Apparently not. So there's my first "Don't" to prospective PMs. Idea
The inquiry process involved sending out "tracers" in three steps:
1) A 2nd email to anyone who didn't email me or post about the original email on uF from the same email account with no gibberish or attachments, Subject of "I Need Your Help!" and an assurance this isn't some kind of scam and asking what had happened to the original email sent. (Another "Don't" comes out of this step: Idea "Don't forget to use the BCC: option when addressing a bulk email. I fully meant to, but pasted the list into the wrong address box.)
2) A copy of the same request for status to the remaining people, this time from my regular gmail account
3) A PM on unFiction with basically the same request to anyone whose original email fate was still unknown.

Results
As the intent was to see what the original fate of the email was, I'll concentrate on the overall picture of what happened and not the number of replies received in any one step. I will say, there were some interesting replies on those requests for status, though.

I guess the bottom line for the analysis would be to say that of 61 original emails sent:
12 remain unaccounted for (19.67%) after 3 attempts to determine their original fate
21 recipients posted about it to the forum (34.43%) and 16 of those (26.23%) posted the attachment needed to collaborate.
6 recipients emailed me upon receipt of the initial email (9.84%). Those with a lot of experience asked the "What is this?/Why should I help you? questions, while the newer players asked "How can I help?" or stated, "I'll do what I can."

Breaking down the known fates shows the following:
3 of the initial emails (4.92%) bounced due to bad addresses (PMs sent to those folks, today).
2 of them (3.28%) were confirmed as being caught by SPAM handlers and deleted, automatically. (These were both in private domains, not like hotmail, MSN, gmail, etc.)
7 (11.48%) were deleted, manually by the recipient (It seems about even for those held in a SPAM filter for deletion and those just deleted outright by the recipient.
8 (13.11%) of the recipients saw the email and ignored it (most of them found it in their inbox and didn't recognize it when they found it based on the request for status.)
7 (11.48%) have no record of ever receiving the original email( This one is interesting as most of the replies of this type were received in the first request for status, which came from the same email address. But leaves us unable to tell whether they were automatic deletions by SPAM routines for content or just that the recipient deleted the original and didn't remember having done so.)

I can provide more granular information about the domains involved in certain fates or player experience rolled-up differently (Always anonymously), if that is of interest to anyone.
_________________
I'm telling you now, so you can't say, "Oh, I didn't know...Nobody told me!"


PostPosted: Wed Jun 13, 2007 6:18 pm
 View user's profile AIM Address Yahoo Messenger MSN Messenger
 Back to top 
konamouse
Official uF Dietitian


Joined: 02 Dec 2002
Posts: 8010
Location: My own alternate reality

 		Shocked

Wow! That's an amazing analysis. Conclusion? Is email viable for establishing the rabbithole?
_________________
'squeek'
r u a Sammeeeee? I am Forever!

PostPosted: Wed Jun 13, 2007 7:42 pm
 View user's profile
 Back to top 
Sin Vraal
Decorated


Joined: 28 Nov 2002
Posts: 219
Location: NJ

 		hm
hm
What I find fascinating is you didn't post how many *solved* your puzzle(s), or the solution paths. As the forum seems to have failed it, especially.

And if you're eager to report from a PM perspective, I challenge you to statisticify the reasons how much effort was put in, how far individuals got, and so forth. =)

PostPosted: Wed Jun 13, 2007 8:35 pm
 View user's profile AIM Address Yahoo Messenger
 ICQ Number 
 Back to top 
Rogi Ocnorb
I Have 100 Cats and Smell of Wee


Joined: 01 Sep 2005
Posts: 4266
Location: Where the cheese is free.
konamouse wrote:
Shocked

Wow! That's an amazing analysis. Conclusion? Is email viable for establishing the rabbithole?


I kinda figured I'd leave it there and that we could open it up for discussion.

I think that, due to the very small number of confirmed SPAM auto-rejections, and the general overall lean of the comments I received, that people's inboxes keep them busier than I might have thought and not as careful as they'd like to be in scrutinizing everything they receive in them.

In this case, the whole of the original content had pretty easily human-readable subtext, so I must assume when they were manually deleted, it was due only to the information available in the email client's inbox summary view. Only a couple of the respondents specifically said they looked it over and decided that while they knew what it was, in theory, they didn't want to pursue it.

In reply to Sin Vraal:
I don't think anyone arrived at any of the three "solves" available. I left the solution paths out of it in case anyone was interested in working on them further. I'm not sure how much effort people put into the text puzzle beyond what was discussed on the forum, but it didn't seem that many put a lot of effort into it. On the origami, I did make one error in reversing the valley/mountain indicators. I didn't know there was a specification for that and went by memory of some of the origami I have seen in print form. Also, if folded the correct way, the partial lines on various pages combine to form complete global folds across the entire sheet. I thought that would be easily caught by the players, but nobody commented on it. Also, an assumption was made about "fronts" vs: "backs" somewhere along the way that became kind of locked-in. I attempted to hint the proper orientation in two of the replies I made to questions about that aspect of the puzzles. I believe everyone was stymied at the same point. That being determination of a proper order to put the squares in. All I can say to that, is that the proper order was hinted at by a player, very early on and is really the only available information that could be used in that fashion.

Can you further explain the "statisticify the reasons how much effort was put in" part of your challenge? I'll gladly comment on it, if possible.
_________________
I'm telling you now, so you can't say, "Oh, I didn't know...Nobody told me!"


PostPosted: Wed Jun 13, 2007 10:34 pm
 View user's profile AIM Address Yahoo Messenger MSN Messenger
 Back to top 
gazdemon
Veteran


Joined: 28 Nov 2005
Posts: 132
Location: Norway

 		Ahh I had wondered what happened to that 'game', I feel very honoured to have recieved an email off you, and I am happy to say that I was one of the people who posted it all to the forum...unfortunately for you I am not that good with puzzles.
It was a very interesting experiment, and thank you for putting in the effort to conduct it.
I think that the results you have produced show that email rabbitholes can work, if you send them to enough people, although if the emails each hold a unique part of a puzzle, and the recipient doesn't post it for whatever reason, then that creates a problem. Pretty much common sense, but interesting none the less!
_________________
***GAZDEMON***

PostPosted: Thu Jun 14, 2007 7:43 am
 View user's profile
 Back to top 
Rogi Ocnorb
I Have 100 Cats and Smell of Wee


Joined: 01 Sep 2005
Posts: 4266
Location: Where the cheese is free.

 		I'm not sure this is part of what Sin Vraal was looking for but I doubt anyone is interested enough in the text puzzle to really try figuring it out, so here's a logical walkthrough. The puzzles were secondary to the overall effort and I wanted them to be just hard enough to foster communication on the forum and bring out as many people who looked at it as possible, right up front.
You might wanna top off your coffee. What follows includes a primer in basic substitution, where it matters not what you're substituting. Single characters, multiple characters, symbols, whatever. This one did have a twist in that a single symbol could relate to more than one text character.
As presented, that puzzle looked like:
Code:
*Th%i%s! i!+s^ w*a!y& mo&!r!e t%ha%!n* I c$an& ha%nd!!le^` on my own.
Won't you please help?

There are eight symbols that stand out: *%!+^&$`.
They occur between letters and don't replace letters.
The number of times that each symbol occurs are: *=3, %=5, !=8, +=1, ^=2, &=3, $=1, `=1.
The total is 24 symbols. Maybe the hidden message contains 24 letters?

Here are some possibilities concerning these eight symbols:

Maybe they modify the letter they follow.
Maybe they modify the letter they precede.
Maybe different symbols modify different letters.
Maybe their positions are a factor, regardless of the letters surrounding them.

For the first three scenarios, we have a problem in that the hidden message appears to start and end with a symbol and there are a couple of places where symbols appear with no letter between them. So none of these three scenarios would seem to be a workable method.
What about the physical position of the symbols?
The number of letters between symbols are 2,1,1,1,0,1,1,1,1,2,0,1,2,2,0,1,2,2,2,2,0,2,0.
That is fairly consistent. If we have 8 symbols, and 3 spacings between symbols (0-2), then we can encrypt 24 letters.
Here, then, are the letters of our message:
Code:
0*, 2%, 1%, 1!, 1!, 0+, 1^, 1*, 1!, 1&, 2&, 0!, 1!, 2%, 2%, 0!, 1*, 2$, 2&, 2%, 2!, 0!, 2^, 0`
There are 14 unique paired values, in all, so this might be a simple substitution cipher, in which the number/symbol combination resolves to a single letter.
We can guess that any given symbol can then be associated with three specific letters, but how to determine which ones?
Frequency analysis should help with this, but we still need someplace to start.
How about experimenting with just assigning the symbols in order starting with "A" = 0 and some symbol, then "B" = 1 and the same symbol, "C" = 2 and that symbol, followed by "D" = 0 and the next symbol.
That would produce groups where:
Code:
    a b c d e f g h i
0 = A,D,G,J,M,P,S,V,Y
1 = B,E,H,K,N,Q,T,W,Z
2 = C,F,I,L,O,R,U,X (One short of the other groups as we ran out of letters)
Or a 9x3 grid where there are 9 total symbols(represented using lower-case letters a-i, above).
Still very possible, given the 8 unique symbols seen in the text.
We could also have stopped at eight and just assumed that Y and Z aren't going to be used, but it'll get messy if either of them are.
The message is rather short; so it may be difficult to solve. Let's start with the highest frequencies in the text.
"2%" and "1!" both appear 4 times.
If we assume that "2%" = "I" (because "I" is the most used letter in English found in the "2" row), our "c" column becomes "%" and the message becomes:
Code:
0*,I,1%,1!,1!,0+,1^,1*,1!,1&,2&,0!,1!,I,I,0!,1*,2$,2&,I,2!,0!,2^,0`
Or:
Code:
0*,I,H,1!,1!,0+,1^,1*,1!,1&,2&,0!,1!,I,I,0!,1*,2$,2&,I,2!,0!,2^,0`
if we convert the other two values from the "c" column to create a "running table" version.
I'm not fond of the "II" occurrence, though. Not very many words with that pair or instances where it would likely appear in english (Like, "SKI IS").
How about we try the next most common english letter in the "2" row which is "O"? Now our "e" column is "%" and the message becomes:
Code:
0*,O,1%,1!,1!,0+,1^,1*,1!,1&,2&,0!,1!,O,O,0!,1*,2$,2&,O,2!,0!,2^,0`
Or:
Code:
0*,O,N,1!,1!,0+,1^,1*,1!,1&,2&,0!,1!,O,O,0!,1*,2$,2&,O,2!,0!,2^,0`
if we convert the other values from the "e" column.
That seems more common. On to the next.
"1!" could equal "E", but that would give us a "EOO" in the middle of our message. I'm thinking, no.
How about we try the next most common letter, "T"? That give us a "!" in our "g" column and would make our message:
Code:
0*,O,1%,T,T,0+,1^,1*,T,1&,2&,0!,T,O,O,0!,1*,2$,2&,O,2!,0!,2^,0`.
Or:
Code:
0*,O,N,T,T,0+,1^,1*,T,1&,2&,S,T,O,O,S,1*,2$,2&,O,U,S,2^,0`
in our running table version
Still seems reasonable. Let's go with the next most common. "0!" occurs 3 times. "A" is the most used english letter on the "0" row, so we get:
Code:
0*,O,1%,T,T,0+,1^,1*,T,1&,2&,A,T,O,O,A,1*,2$,2&,O,2!,A,2^,0`

Here, we run into a problem on our "running table" version as the "1!" and "2!" substitutions already occurred. So nothing additional to do. We simply have:
Code:
0*,O,N,T,T,0+,1^,1*,T,1&,2&,S,T,O,O,S,1*,2$,2&,O,U,S,2^,0`
At this point when comparing the single substitutions with our running table version, I'm liking the look of the latter. The "ATOOA" string just looks wrong.

Okay. Let's go on to "1*" and "2&", which both occur two times.
We didn't use our most common letter for the "1" row, yet, so let's replace "1*" with "E", and assign "*" to the "b" column to get:
Code:
0*,O,1%,T,T,0+,1^,E,T,1&,2&,A,T,O,O,A,E,2$,2&,O,2!,A,2^,0`
and our running version becomes:
Code:
D,O,N,T,T,0+,1^,E,T,1&,2&,S,T,O,O,S,E,2$,2&,O,U,S,2^,0`
The running version is looking better and better. Let's do the "2&".
We used the "E" and "T" from the "1" row so the next most common english letter is "I". Switching out the "2&" with it gives us a "& in our "c" column and:
Code:
0*,O,1%,T,T,0+,1^,E,T,1&,I,A,T,O,O,A,E,2$,I,O,2!,A,2^,0`
and for the running version:
Code:
D,O,N,T,T,0+,1^,E,T,H,I,S,T,O,O,S,E,2$,I,O,U,S,2^,0`
At this point I'm all about forgetting the single substitutions and going with the table.
It's pretty evident that the last three words are going to be "THIS TOO SERIUOSLY", so let's just work with those 3 subs and their associated table columns.
Now we have "2$" = "R", "2^" = "L" and "O`" = "Y" with "$" in our f" column, "^" in our "d" column and "`" in our "i" column. Or:
Code:
D,O,N,T,T,0+,K,E,T,H,I,S,T,O,O,S,E,R,I,O,U,S,L,Y
It's pretty obvious the "0+" is going to be an "A", but going back to frequency analysis would have locked it in as "A" is the most common english letter in the "0" row.
And our final table looks like:
Code:
    + * & ^ % $ ! h `
0 = A,D,G,J,M,P,S,V,Y
1 = B,E,H,K,N,Q,T,W,Z
2 = C,F,I,L,O,R,U,X
Column "h" never got defined as none of its characters appeared in the text.

Make sense?
_________________
I'm telling you now, so you can't say, "Oh, I didn't know...Nobody told me!"


PostPosted: Thu Jun 14, 2007 6:33 pm
 View user's profile AIM Address Yahoo Messenger MSN Messenger
 Back to top 
konamouse
Official uF Dietitian


Joined: 02 Dec 2002
Posts: 8010
Location: My own alternate reality

 		Shocked wimper

Bet ya grumpboy would have figured it out (given enough time). Wink
_________________
'squeek'
r u a Sammeeeee? I am Forever!

PostPosted: Thu Jun 14, 2007 7:29 pm
 View user's profile
 Back to top 
Rekidk
Entrenched


Joined: 19 Dec 2006
Posts: 992
Location: Indiana, USA

 		Shocked is right.
_________________
iTube - iTweet - iNetwork

PostPosted: Thu Jun 14, 2007 7:49 pm
 View user's profile Visit poster's website
 Back to top 
Rogi Ocnorb
I Have 100 Cats and Smell of Wee


Joined: 01 Sep 2005
Posts: 4266
Location: Where the cheese is free.

 		It's not nearly as complicated if you add a little "pencil and paper" to the mix.
_________________
I'm telling you now, so you can't say, "Oh, I didn't know...Nobody told me!"


PostPosted: Thu Jun 14, 2007 7:59 pm
 View user's profile AIM Address Yahoo Messenger MSN Messenger
 Back to top 
sixsidedsquare
Unfettered

Joined: 24 Mar 2005
Posts: 409
Location: 60E

 		Gah, I got part way there, just didn't think of bringing number of letters preceeding the symbols into the mix, and gave up when I wasn't seeing any patterens.
Nice one.

PostPosted: Thu Jun 14, 2007 8:37 pm
 View user's profile
 Back to top 
gazdemon
Veteran


Joined: 28 Nov 2005
Posts: 132
Location: Norway

 		Shocked Congrats Rogi!
_________________
***GAZDEMON***

PostPosted: Sat Jun 16, 2007 4:19 am
 View user's profile
 Back to top 
Display posts from previous:   Sort by:   
Page 1 of 1 [11 Posts]  
View previous topicView next topic
 Forum index » Meta » Puppetmaster Help
Jump to:  


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum
You cannot post calendar events in this forum



Powered by phpBB © 2001, 2005 phpBB Group