Saying No to Google Docs

Looks like sharing files isn’t gonna work. Bart is now complaining of having three different google accounts for different purposes and he isn’t the only one logged in. Ergo, I will not play around with google-docs any further.  If I had a thousand readers following this blog then maybe I can afford to throw Bart under the bus … okay maybe I could have phrased that better ☹

Instead, I have decided to have a spreadsheet summarising the state of play. There are 50 rows. Each row corresponds to an “action” which I define as a sequence of moves that reveals at least one more card (by turning over a face-down card or dealing from the stock). Note that clearing an empty column or removing a suit is not an action (unless you also happen to turnover a card). This is desirable because we know the maximum number of actions in a game is 50 – the game starts with 44 face-down cards and we deal six rounds, counting the initial ten face-up cards as our first action. If we win the game, then we must have exactly 50 actions. The converse is false since it is theoretically possible to reveal every card and still lose. But that’s a pathological situation most unlikely to occur. If you’re playing half-decent (and not a stooge), you can consider exposing all cards tantamount to winning.

Apologies for the lousy formatting but this will have to do for now.

 Actual playYelPurBlaWhiBluBroPinRedGre DGrOraDBl
1Deal 3c 5s 2h 5d 9s 6c 6s As 2s Kh            
2hi -> 7d            
3bg -> 7c            
4fb -> Kc            
5db -> 9c  other passpassotherotherotherother other
6ia -> 9s            
7gh -> Jd            
8Deal 2s 6d 4c 3d 2c Td Js 7s 8h 9c            
9fg,ij,jg,jg -> Tcotherotherother?other?other?otherpass?other?passillegal?? pass

The current spreadsheet has only ten rows filled. Obviously when we get to the middlegame I would have a similar spreadsheet but with less empty rows. That is assuming the card gods are kind enough to actually let us reach 30 actions! Therefore no “indexing” is necessary. If I update the spreadsheet after (say) every fifth action then you only need to read the latest blog post to say up to date.

For each action I have recorded whether each blob played the chosen action (blank), suggested a different action (other) or offered no action (pass). A question mark or indicates at least one of the readers flagged the blob as sus. A double question mark indicates very sus (since not all mistakes are created equal!). Note that most of the actions so far are trivial, so many cells will be blank.

I have not recorded (1) if a blob accuses another blob of being sus or (2) the moves of alternative actions. There is only so much information the spreadsheet can have without becoming too cluttered. Recording any extra information is the responsibility of my readers. For instance, you might wanna keep notes on which players are experts, novices or somewhere in between.

In other words, I will be responsible for updating the basic spreadsheet and my readers will keep private records on whatever additional information they think will be useful.

Note that no action does not imply failure to contribute. For instance, Red may (1) give a specific reason why Blue’s action is inferior or (2) say “our long term goal is XYZ”. Either of these can be a valuable contribution, even if Red fails to give a specific sequence of moves. Of course, it could also be a not-so-valuable contribution if Red is a stooge 😊

Please let me know if this is a viable plan. If yes, then now is a good time to point out any errors in the spreadsheet 😊

8 thoughts on “Saying No to Google Docs

  1. Well, I’ve got my doubts as to whether this will fly, but I’ll give it a shot. Based on experience with sparse-data spreadsheets, having “blank” have a meaning like “made play” is asking for trouble. Because it’s going to get confused with the blanks meaning “no information”. I also have no way of finding the blog entry corresponding to “row 5”. A column in the spreadsheet or a separate list might help with that.


    1. Hi Bart,

      I think it is best if we abort and start a new hand, seeing we have only invested 2 or 3 non-trivial decisions.
      I wanna start a “Mashup-Lite” version instead with the following rules

      (1) only three coloured blobs instead of twelve. All blobs are experts

      (2) You can only vote someone off after the game has ended

      (3) 1 VP for removing all eight suits, 1 VP for ejecting the right blob.

      (4) On each decision, all blobs must give an action (sequence of moves) which is recorded in the spreadsheet.

      (5) Blobs can also give short explanations supporting their action, but this is not recorded in the spreadsheet

      (6) I will aim for two (game-related) posts per week. This is only a rough guide and obviously I have no control over the number of trivial decisions or whether the game ends prematurely and catastrophically with half the cards unseen.

      What do you think?


      1. That sounds like a pretty good idea. Too bad the other one didn’t work out, but when a person tries to be creative, something often goes wrong the first ten times. So we try again.

        I’m still not seeing any way I’ll be able to navigate to the post where I can see the screen shot. I saw you put in text notations of changes, deals, etc., but that’s of no direct use to me. Screen shots, screen shots, ra ra ra! We’ve got to have a link in each post to the previous post, at the very least.

        My GoogleDocs idea was also a spectacular failure, but here’s an idea… Can you make a single blog post but then keep editing it? Like adding the screen shots, text, and the updated spreadsheet at the end? Treat it a bit like a block chain in that you promise never to rewrite history from the first part, please. A quick search suggests there is no upper limit on WordPress post sizes. Replies would form one big long list, but I figure that’s OK.

        With only three blobs, and easy access to the text of previous posts/positions, we might not even need a spreadsheet.

        So blob conviction is a relative matter? If someone looks “sus” we withhold judgment to see if maybe someone will be even MORE “sus” on a later move? I can handle that.

        It could be argued that you owe me and the ‘hopper your view on what you would have done in a couple situations he and I disagreed about. Does your vague promise to address them when the game is over apply if the game is aborted? Of course you might just want to move on, and I couldn’t blame you for that. (But didn’t your mother remind you that if you have a job to do you should usually do it right away so you don’t forget to do it later? 🙂 )

        I wouldn’t work for a single VP. I need motivation. Let’s make that 100 VPs for the 8 suits and 100 VPs for nabbing the bad guy. 🙂


  2. Ahhhh yes, the ol’ KISS System, I should think this works much better, start with something manageable and then complicate it beyond reason as soon as we have mastered it.

    Twice a week should be fine in the early stages. If I think I will probably be tardy, I will just do a quickie post and request a little more time. Or youse guys can just run off and leave me to play catsup.

    Master Chi-Yuen, I humbly second Esteemed Scholar Bart’s request for your insight into the two instances when we went on diffident paths. This is why we are here, to learn from the Master. And if we have a bit of fun along the way, OK by me.


    1. Hi Bart and SA,

      Yes, I am happy to abort this game and try the lite version

      Re the situations where you disagreed: (1) I like fg,ij,ig,jg since this avoids having two jacks and no tens in Column 7. If a King later appears in column 7 then at least we bury a Ten as well as two Jacks, so a shortage of jacks will be less problematic (2) in the latest position I would take my three turnovers and not mess with column 8. We have a shortage of Eights and there is a fair chance the shortage remains a problem after the next deal. Keeping two Sevens and no Eights in column 8 increases our chances of handling a shortage of Eights in the next few rounds.

      Re the format, I think it is wise to combine the ideas of linking to previous posts and continually editing. More specifically we know there are 6 rounds in Spider Solitaire. Assuming we don’t concede before the last deal, we will always play exactly 6 rounds, win or lose. Ergo, if I devote one blog post per round and each post links to another post corresponding to the previous round then we are quids in.

      Being a research scientist, I fully agree with new ideas going wrong for the first 10 times. That’s just the nature of research. The real skill is recognising when things go wrong and accepting small losses. Every poker player knows you save money by folding sooner rather than later, and the final result matters more than individual hands (though I admit this analogy is a bit clumsy).

      Hope that makes sense


      1. Fascinating to hear your takes on the two situations we asked about… not so much the decision but the reasons for them. Noting and concern for shortages of certain numbers is familiar to me, but the way you apply it to distinguish between plans interests me a lot. Thinking that burying 2 jacks with a 10 is better than burying 2 jacks alone… my thinking hadn’t gone that way. And in the other case, noting a shortage of 8s as a reason to NOT clear out a column that uses up a couple 8s.

        I can certainly do twice a week. I could do more often.

        I am concerned about this “6 rounds” limitation. As you well know, in a key round you might have half a dozen places where you have really hard decisions to make, each dependent on the previous one. So I figure if we constrain ourselves to only 6 rounds we’ll likely miss some things. One idea would be to let yourself pause more than once in a round, but don’t do it too often…

        “Folding early” as a model for dealing with research errors captures a bit of it, of course… but sticking with it can also pay off. (And of course you knew that too.) There was an article once about some key pieces of technology that nearly got canceled because they were late — the IBM-360 was one, and the Boeing 747 was another… Those are engineering, not basic science, but…


    1. I have now made our first move and edited the latest post accordingly. If nothing else, the failed attempt with 12 blobs will give me a story to tell if I ever get interviewed about “giving an example where you had to make a difficult decision when leading a team project” 🙂


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s