Reading through the past posts, it seems that there was a pretty vigorous debate regarding the new symbols to be used for :A and :B (I guess you can tell by that which I prefer, huh?). It seems that at one point a vote, or at least a poll, was going to be taken, including both the DSA and the DSGB. Did anything ever come of that? If so, what?

The DSA's Bulletin, which has just come out, says they're going to publish a summary of symbols.
As for the vote - it was tried by the DSGB many years ago and the present choice of symbols was confirmed until such time as someone could design ones acceptable to all ...
When the DSGB started Brian Bishop used the DSA symbols; the change to the pitman symbols was largely my doing; I'm quite happy with the symbol for eleven but the one for ten can get confused with other symbols when written badly, hastily or sloppily. I would like to see a suggestion that cannot possibly be mistaken for another number however bad the handwriting!

Well, I appreciate your effort in getting the Pitman symbols adopted. I don't personally have any trouble with :A; some of our current numbers are easily mistaken for others. For example, 4 often (and notoriously) looks like 9, and 1 like 7 (except when written with the European slash, which I adopted myself, though an American, in order to avoid the confusion). Still, I'd be quite glad to see some consensus, and would gladly write new Metafont fonts if a different symbol for :A were devised.

I'm sending my 10 dollar check to the DSA tomorrow, so hopefully I'll be on track to get the issues in which this occurs. Thanks for the update.

The current issue of The Duodecimal Bulletin has a short article with tables on my proposal for dozenal nomenclature (4X.09-X). My dozenal numeral system is similar to that of Don Hammond, and is readily adaptable to a 7-bar LED/LCD grid.

I do like Hammond's proposals, though I still prefer Pitman's. I'm still looking forward to receiving issues of the Bulletin to see what comes of these efforts, and to contribute where I can.

And just how would you represent Pittman's "ten" symbol on a 7-bar LED/LCD grid so as not to confuse it with the letter S?

Who cares if it looks like an "S"? We don't (can't) put our alphabet into a seven-segment display. You mean so it doesn't look like a "5," right?

In any case, it strikes me as a non-issue. First, seven-segment displays are rarely seen anywhere other than gas pumps, cheap alarm clocks, and 90-cent pocket calculators, and they'd all have to be replaced anyway because most of them simply can't be set to use dozenal numeration. Even most microwaves I see these days have significantly more versatile displays. Seven-segment displays are on the way out, and I see no need to hand them an extra crutch by changing our symbols to suit a dying technology.

Second, many of our seven-segment digits don't look a whole lot like the numbers we write. E.g., I don't write "9" as

Mine has a straight tail, no curve to the left. I also don't write "7" as

Mine has a horizontal slash across the middle, something wholly ignored by the seven-segment representation. My "4" actually does look like a seven-segment four, and always has. But I don't find it particularly tragic to have the seven-segment display look different from the written digit, which itself often looks different from the printed one.

For myself, I'd represent the Pitman :A the same way Hammond did his ten:

It's only missing one segment from being an accurate representation of :A, but it's still easily distinguishable from 5.

I know some say that this looks like a broken 5, but I don't think that matters much. Not only do we have to change our symbols to suit a dying technology, but we also have to change them to suit broken examples of a dying technology? In any case, it's no different from many of our current numbers in that a single broken segment creates ambiguity; that's the problem with having so few segments.

After all, only one broken segment could make an 8 into a 9, a 6, or a 0.

Still, like I said, while I'm perfectly happy with the Pitman characters, a consensus on the Hammond characters would make me quite happy, as well. I'm looking forward to being a part of helping build a consensus on new digits, whatever they end up being.

Don,

The Duodecimal Bulletin is considering "theming" certain issues. The theme of interest to start the June discussion is symbology. I suggest symbology because it is a highly accessible subject; just about anyone can submit a set of symbols they like. It is also of interest to most dozenalists. I've built font extensions in the style of the Bulletin font (Adobe Arno Pro) so that most of the symbologies encountered in the Duodecimal Bulletin's first three and a half dozen volumes can be typed on the computer (including Pitman, P.D. Thomas' "modulor", et al.) These font extensions are structured to operate with Mathematica's "BaseForm" command, so that we can produce Mathematica output and readily convert it to any supported symbology. We can convert any chart or text we write (sometimes with a little work) to any symbology - provided it is a traditional positional notation. Conversion to the Halcro reverse notation requires a formal change in the numerals, so it is not interconvertible.
I've written a synopsis of the dozenal symbologies, organizing them in an extension of Ralph Beard's 1940s article "The Opposed Principles". I'd like to be as "complete" as possible, so am awaiting material from the DSGB to ensure we have as much of the discussion as possible. It's unfinished, merely an exposition for the June discussion, but is extensive. Would send it to you if you like (perhaps you already wrote to us; I am writing this while on vacation/holiday). The article may be finished if found acceptable to those gathered at the 27 June meeting. A by product of this article is the DSA's ability to publish using all those symbologies from this point on.

Personally, I am not as concerned regarding symbols with "confusion" (because the current digits can present confusion, people tend to figure things out, when one needs certainty there are means to convey it, and if bad writing is a potential culprit for misunderstanding, well, write slower!) I think garnering an absolutely rock solid set of numerals that is not "confusing" and is absolutely rational straitjackets us into forms that are very arbitrated and difficult to write. Am not concerned with "efficiency" - the ease of distinguishing forms from one another is more important. To me the most important consideration is whether or not the forms can be written by human hand, used everyday, and that all other technical considerations (LCDs, pixels, etc,) are fleeting and temporary, thus needless constraints. Like you, I think the LCD/LED constraint is an invalid one that seems to function in people's minds far more strongly than is merited. Already the gas stations, a place where most see these seven segment numbers, are converting to "pixel" array LEDs (because the price of gas surged last year). The LED seven segment constraint is perhaps as useful as constraining oneself to the ability to write numerals in cuneiform. To me the decision about symbols must be based from the fundamentals; that is always going to be human execution of the forms. Anything else will be smoothed by human execution into some other form the designer did not intend. But these are my personal thoughts on symbology. Others will have their thoughts, and it is important to hear these in a discussion of symbols, and let you, the reader, decide for yourself.

I agree with most of what you said, Icarus.

It also occurs to me that all this fashion with digital is making us neglect the still very important analog. Take the TGM ruler I'm currently making. I intend to "brand" the symbols into the wood with normal, metal brands heated up, to char the numerals on. (I have no way of etching it otherwise, and pen/marker/etc. would wipe off). With the Pitman numerals, this is easy: just flip the 2 and the 3 around to make :A and :B. However, with any new numeral, I'd have to

1.) Cut new numerals myself, something completely outside my competence, or
2.) Commission another to cut new numerals, something completely outside my budget.

This can easily continue to be a problem; e.g., with rabbit-raising, another hobby of mine, breeders often mark the ears of the animals with information, including numbers. Do we really want to require people to produce new presses for doing this?

I'm glad to hear that someone else has been doing specific font design for dozenal numerals, rather than merely flipping the current numerals. I'd greatly appreciate receiving any font designs you might have. With your permission, I may see about converting these to Metafont, or using them as-is in Postscript, and including them with the dozenal package, since Postscript fonts are easy to use with LaTeX. Would you be interested in that? All proper credit would naturally be given to you, as their designer.

Icarus, are you the one to talk to about the Duodecimal Bulletin? As I mentioned in my "Getting to Know You" post, I'm very dedicated to LaTeX, an excellent typesetting system based on TeX, which is designed specifically to produce professional typeset results, especially mathematics. I've designed fonts to work with that system based on the Pitman numerals, and because TeX and progeny is pretty widespread among mathematicians, I was hoping to put something in the Bulletin later (once my dues are received, of course; I sent them in yesterday). Is that something the Bulletin would be interested in publishing?

Apologies if this response posts twice...wrote one via phone but it seems not to have taken.

The Duodecimal Bulletin welcomes article submissions from Members and dozenalists alike. The Bulletin is a peer-reviewed mathematics journal. Articles submitted will be examined by mathematicians (I myself am not a mathematician, to my own chagrin, but Prof. Zirkel and Prof. Schiffman are indeed professionals). I can produce virtually any symbol you can devise, but for the purposes of the Duodecimal Bulletin, the symbols will be crafted to resemble the typeface of the Duodecimal Bulletin (Adobe Arno Pro). That typeface is a serif font. Currently I've focused on the regular font, not the italic nor bold, etc. Many of the posts on this forum are suitable for publication; we have yet to decide regarding this; I think we'll be pointing to these rather than printing them. I would like to attribute the individual behind the post. I would be happy to work with you to get your article published in the Bulletin. I am on vacation/holiday so can't definitely recall my mail address for DSA, will post that later.

Once a symbology is covered by a DSA typeface extension (technically, these fonts are "extensions", that is, they contain the numerals, certain symbols, and transdecimal (=greater than symbols used for base ten) numerals), the symbology is "supported". All DSA font extensions are produced in the same way, i.e. so that Mathematica's BaseForm[] function can be used to generate output which can "convert" to the symbology merely by switching the typeface. This allows us the ability to convert any material I / DSA produced since about mid-2008 to any supported symbology which employs traditional positional notation. Many of the symbols are either altered from existing parts, but many are entirely produced. They are visually altered, which means that I haven't spent a lot of time correcting nuance, so some symbols may, if typed at a large point size, seem a little hurdy gurdy (I call it "energy" or "buzz" in the forms). I was and still am after mass quantity and coverage rather than super-fine form, but I think the forms are covered well enough. I have a set of 360+ numerals I use for very large bases, many of which correspond closely with glyphs others may have proposed for their duodecimal numerals; this becomes a kit of parts from which I may select and bash to fit the symbology in question. (I now use base 60 regularly, almost as much as dozenal at work, so needed a whole lot of symbols - this lead to a fluid ability to produce sets of symbols.)

I plan to ask users of this forum to submit their symbologies if they desire, along with attribution information. Have not done that to date because the DSA has yet to discuss the symbology theme and themes in general.

The article I wrote is a massive compendium that may not fit in the Bulletin, but may be made available as a stand alone PDF. It also includes octal, hex, and sexagesimal symbologies, focusing on "constructed numerals" rather than historical (Mayan, Babylonian, etc.) symbologies. There is a side-by-side comparison spread which puts all the symbologies in one place so you may compare them.

About tex/latex, I suppose we can chat about it off the forum. I am away from office till 8 June, then on a hot deadline till about 18 June. Seems exciting, Don!

dgoodmaniii, I have to disagree with your argument for the choice of new digit symbols!

If you were to simply flip the numeral 4 you would have a perfectly good numeral for either 10 0r 11 (I prefer it for use as decimal 10 as it would occupy the position directly opposite the numeral 4 on a clock). This gives a numeral that can be easily written by hand without lifting the pen, is clearly distinguishable from all other numerals, is unlikely to be misinterpreted as another digit and can be created simply by inverting an existing symbol 4 in existing computer fonts.

It would still require that you cut a new symbol to burn it on your TGM ruler as it is not a numeral 4 that is rotated 180 degrees in the plane of the page, but the image of the numeral 4 rotated 180 degrees out of the plane of the page about a horizontal line in the centre of the numeral height.

However, this is an insignificant reason for the adoption of these symbols as it is more important that printed symbols are easily created than a hard cut branding iron for a wooden ruler. The fact you can create branding irons for the two new digits by using the Pitman versions should not be the criteria for the choice of symbols.

The true criteria for any new symbols should be practicality, usability, uniqueness, minimization of misinterpretation, ease of creation and only lastly, artistic harmony with existing numerals. These are just some of my personal suggestions of criteria. The choice of symbols should not be made without first establishing the criteria for their choice. Establish a set of criteria separate and divorced from any actual symbols, then evaluate all candidates against the agreed criteria.

So far all that seems to have happened is differing examples have been proposed complete with personal preferences. We all want to arrive at a consensus on the shape of the new symbols but this will never happen unless we agree a binding set of characteristics that these symbols must possess. Forget the d**n symbols, agree the criteria FIRST!

my opinion on symbols/names is and will always be: let the best (= most widespread usage) one win. De facto usage cannot be trumped by any set of grand viziers on high (even if its us regular dozensonline posters). For better or worse that's gotta be dwiggins or pitman at this point. What makes us THE people to design and issue an edict on high to force the world (if that be possible) to use upside flipped 4s, swash Es, or any other glyph for numerals? If we dare feel strongly bout our figures, then go use them. I don't think the dudette (probably a 13 year old kid fooling around in the bazaar one slow Tuesday) thought bout criteria when she came up with 012345... Those symbols got smashed round plenty to become what we are using at the moment. Recent events have illustrated what a bunch of (financial, in our current case) no-wit-alls can concoct "this time its different". Bottom line, we dozen dozen people can agree on a symbology, butwe're probably leaving out the gal in dharmsala that actually invents a compelling superapplication for dozenal she's based on devanagari numerals, which eventually renders our agreement irrelevant. Wanna make a compelling proposal for transdecimal digits? Use 'em, get zyriads of others to follow suit. Exercise dem fat little symbols, smooth out all the kinks in that unwieldy flipped 4 with all its angles! I wanna see it scrawled on a steel i beam getting hoisted to the dekth level of a skyscraper, in a commercial for a dek dollar meter long subwayvsandwich. Lets see it in graffiti, on shopping lists, in a sign over the five and dime. Till its there we'll all have issues with one another's numerals. Its gotta come out of the math whiz notebooks and get to meet people in the whatsit2me street, where it will get twisted round and shown the real world. Even dwiggins and pitman are coy little symbols yet, let's see how that upside down two looks written in a hurry taking an order or written in a stock trade. The pitman eleven seems to hold well under duress. Nuances like Hammond would get lost and fly over the heads of valley girls and lumberjacks. Anything else is like a UN mandate (useless and meaningless) or pronunciation of Pluto as not a planet (Pluto is what it is, what collected eggheads call it is trivial, actually). Btw hi ruthe! Nothin against you personally jus my two cents (1.4 eurocents or whatever). :)

I think the problem with our Pitman ten is the way it can be mistaken for other numbers when scribbled by hand. I've tried out all sorts of variations on that ten to find an unambiguous version - but without any real success; it can still end up like a two, a seven and a six. I don't like the Dwiggins X, but it can't be messed up as much by handwriting.
What I actually use now when writing dozenal is a version of a character that started life as the Chinese symbol for 5, but without the stroke at the bottom. See the third character in the attached image. Vaxan suggested a version of the first character in the illustration, but facing the other way, and not as easy to write.

NO! :-) No, that's your privilege, obviously. Still, I disagree with what you say here.

In what way is that better than the Pitman :A? I write :A without ever lifting the pen (though I do lift it for "4", for where I come from we were always taught to write it like a seven-segment "4" for fear it would otherwise look like a "9", which it would). :A doesn't really look like any other number; it's at least as different from all others are "4" is from "9" and "9" is from "6," and we don't complain about confusion with those numbers. Nor have I had any trouble distinguishing :A from any other digit. Furthermore, passable characters for :A can be created simply by inverting the 2, though a better one would be specifically designed that way (as I did for the dozenal package in LaTeX, and as I'm sure others have done in other contexts).

So what's the advantage of inverting a "4" as opposed to the already quite common :A?

I expect that anyone involved in animal husbandry, woodworking, signmaking, and the like would not consider this an "insignificant reason." Computers are cool and all, but there is still a real world apart from them, and that real world often involves using numbers in ways that are not easily produced.

I think you're thinking too little of your fellow dozenalists here. People have personal preferences for their choice of symbols because those symbols meet their criteria for being good symbols, not because they irrationally pointed at two and said, "Hey, I like those."

All in all, I think Icarus is right; we ought to just keep using the symbols we like, and the ones that are acceptable to most people will be the ones that come out on top.

Shaun, what numbers do your :A look like when you write it by hand? That's not a problem that I've ever had, personally.

Cuneiform would at least make it easy to design new digits ;)

Ten = [DoHTML]<pre> Y Y Y
Y Y Y Y
Y Y Y</pre>[/DoHTML]
Eleven = [DoHTML]<pre>Y Y Y Y
Y Y Y
Y Y Y Y</pre>[/DoHTML]

dgoodmaniii, icarus.

Whatever, do as you like. I don't have any personal ill feeling to either of you, but just thought a modicum of logic could help to resolve the issue of trans-decimal digits. but let's just leave it to evolution.

Let me know when the accepted versions have become the de-facto standard.

Was this really necessary? As though the opinion of Icarus or myself were not based on logic!

The point is that there are lots of considerations to take into account, and many of those considerations depend on variable prudential judgment. Since we don't all have the same judgment, it's better to allow things to go as they will. How is that not logical?