It is currently Wed Dec 02, 2020 12:27 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 11 posts ] 
Author Message
Offline
 Post subject: On "How evaluate double sente moves?"
Post #1 Posted: Sun Oct 18, 2020 11:14 pm 
Tengen

Posts: 5202
Liked others: 0
Was liked: 720
Citation reference https://www.lifein19x19.com/viewtopic.p ... 91#p260691

John Fairbairn wrote:
There is no good text, in proper English, that tells ordinary players what comes after our Stage 1 equivalent of TTLS: Ogawa/Davies.
[...]from the perspective of non-mathematicians.


Since you ask for something from the perspective of non-mathematicians, you might mean "without mathematics". Then O Meien's and Antti Törmänen's endgame books fail because both heavily rely on mathematics, and more of that than needed to proceed with endgame understanding beyond Ogawa/Davies. My Endgame 1 - Fundamentals book might fit because it works without mathematics and is also useful for 4 - 1 kyu, where Ogawa/Davies lacks.

Or you might mean something explaining the "simple" mathematics of endgame theory with a scope beyond Ogawa/Davies and better explained for mathematical dummies than Antti's Rational Endgame, my Endgame 2 - Values and O's Yose - Absolute Counting. Each of these books makes some attempt to introduce endgame evaluation but also enters this topic more deeply and faster than a "dummy" desires. Surely, there is still scope for an "Endgame Calculation for Dummies".

However, do not eternally ride on the idea of such ever doing so from the perspective of non-mathematicians. Calculations do require maths. There is no explanation of calculations, for which you could then not complain "but it is not from the perspective of non-mathematicians".

One can bury calculations in informal descriptions. "Compare five pieces of coal to 4 pieces of sugar. Black has more pieces." Ugh. Similarly, Antti has tried hard to bury much and the result is an alleged layman's language that is a new burden.

For what purpose? To make a 10k at endgame a 9k?

Endgame does profit extraordinarily much from mathematics that does not hide itself! Mathematics like in Endgame 2 - Values. Even the mathematics in Endgame 3 - Accurate Local Evaluation is fairly simple: compare two values, calculate averages, apply formulas simpler than E = mc^2 or Euklid.

Maybe you have been burnt by the mathematics of combinatorial game theory a la Mathematical Go Endgames or thermography a la Spight? That mathematics is an overkill by far indeed for the purpose of practical application as a player. See Endgame 2 - Values for doing it with practically applicable mathematics. Instead of infinitesimals, I have simplified the maths to, e.g., distinguishing even from odd. You still won't be happy because you can never accept reading a formula C = S ("the count is that of the sente follower") or C = (B + W) / 2 ("the count is the average"), as might occur in 7th year at school.

You never rest to play the "sympathise with the haters of all numbers" advocate while you have been a programmer doing much greater abstraction than found in my endgame calculations. You do so while praising O's endgame book, which already enters numbers deeply. Become reasonable - admit that numbers and calculations are useful!

Quote:
I'm not aware of any good Stage 3 material


Because you have not read it. Or maybe because you would want a book explaining such material split into 10 books to meet your imagined, desired speed of learning at the price of EUR 250 instead of 25. Possible, but do you think that people want to spend that much? Or do you want people to stop after volume 1 stuck at 10 kyu level? Like chapter 4.1 Simple Gote without Follow-up of the book Endgame Problems 1 stretched to fill a whole book, endlessly drilling the most basic case while replacing explicit calculations with apples and oranges tales? Possible, but, again, why would you want to force everybody to spend EUR 250 to learn just the basics of endgame calculations or only read the first such book and be stuck at 10 kyu level of endgame understanding, without ever approaching "Stage 3 material"?

Face it - endgame does require calculations!

Quite like life and death does require tactical reading and its decision-making.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #2 Posted: Mon Oct 19, 2020 2:26 am 
Oza

Posts: 2759
Liked others: 16
Was liked: 3867
Quote:
O Meien's and Antti Törmänen's endgame books fail because both heavily rely on mathematics


O Meien's book does NOT rely heavily on mathematicw, or even slightly. He mentions a little bit of arithmetic (which is usually distinguished from mathematics in normal English in the same way that running for the bus is not the same as what Usain Bolt does). I'm guessing you haven't read the book and just looked at the pictures, and are relying on spotting trivia such as (5 + {-1}) ÷ 2 = 2. But in the actual text he laughs this away: "What I am saying is ending up sounding like an elementary-school arithmetic lesson!"

And O Meien's book certainly does not fail: it taught me a LOT, and it has the Bill Spight seal of approval. What higher praise can there be? (However, for me Antti's book did fail miserably. Not because of mathematics (was there any?). I just found he lacked enough experience of teaching or writing.)

But your reply just confirms my point. While I can sincerely thank you for taking the time to reply, I can't say that I see any unbending to accommodate yourself to a potential audience. The mathematical techniques are not the problem. I was actually very good at maths at school, but only in the sense of passing exams. I even tried to take an interest to the extent of studying engineering and advanced maths privately and passing exams privately. I have had to deal with VERY advanced maths in foreign languages, as a technical translator for 50 years. By the very nature of technical papers and patents which peers in other countries are interested in, they are at the leading edge of their technologies. I have read Berlekamp's book. I have bought and tried to read yours.

Maths itself is not the problem. Mindset is. I do not think like a mathematician. And I do not want to, even though I bought a thick book called "How to think like a mathematician" and it is still in my tsundoku pile."

I think like a native English speaker. There are millions (billions?) of us. And what I say here applies to all other human languages. We all share the characteristic that we are intimately familiar with our own language, with its ambiguities and other foibles, and with other people's linguistic foibles. What we are not familiar with, unless we have the mindset of a mathematician, is English distorted into a parody of itself by misguided attempts to turn our supremely useful, beautiful, varied, rich and entertaining native tongues into mathemicalese. Or logicese. (See how easily I can invent two daft words with the total confidence that everyone will understand me.) Where you see 'count' as a problem because it has two meanings, we can retort that it actually has many more meanings and we don't have a problem with any of them. Where you see a hiccup because I, say, temporarily take a word in a meaning you didn't intend, I see a learning opportunity, because in resolving that hiccup Instead of restricting myself, effectively and unnaturally forcing myself to think of only one definition, I am making associations and stretching boundaries, broadening my total knowledge. I want a high R number for the knowledge virus in my brain. That's normal. It's not just me. It's not just go. We all drive cars. It would be a real mess if we all restricted ourselves to staring fixedly at the steering wheel all the time, just because steering is so important. Driving is actually a much better and safer experience if we also change gears, look out the window, wave to a friend, take a slurp of pop, turn on the radio, remember to put petrol in t he tank, etc etc, using the full panoply of our neurons. After all, the human brain is still capable of much, much more than AlphaGo. Your readers are capable of much more.

Mindset, perspective. Call it what you will. But just listen to the MAJORITY of your potential readership. Engage with them in their own language, where ambiguities are actually very useful.

Allow me to give a specific example that occurred a few days ago. I was talking with a daughter about her son applying for university. To get into the best nowadays you need a strategy. She sent a text saying "Yeah, I agree, you can’t overthink the college selection." What she was agreeing to was me saying "You can overthink the college selection." My instant reaction was that she had made a mistake. But I very soon realised that was going on, in posh linguistics terms, was a coincidentally different surface manifestation of two valid underlying transformations with the same semantic origin. In normal English, I was stressing the OVERthink, she was stressing the MUSTN’T. We were actually, as near as dammit, in complete synch. If she had said "can overthink" like me, I wouldn't have hiccupped and that conversation would have probably stopped right there and been forgotten. But because of the hiccup, and resolving it, we talked in more depth, saw additional perspectives, kept the topic interesting and alive, and as a result talked about it again yesterday, with more people present and so eliciting even more knowledge and awareness of selection strategy.

Ogawa/Davies and O Meien work beautifully at that level of c=engaging creatively through normal language. No other book I know of does.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #3 Posted: Mon Oct 19, 2020 3:18 am 
Honinbo

Posts: 10554
Liked others: 3501
Was liked: 3313
John Fairbairn wrote:
I was talking with a daughter about her son applying for university. To get into the best nowadays you need a strategy. She sent a text saying "Yeah, I agree, you can’t overthink the college selection." What she was agreeing to was me saying "You can overthink the college selection."


This kind of thing is not uncommon in American English. I have heard that it actually comes from speakers of Yiddish. Interesting if true. :)

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

My two main guides in life:
My mother and my wife. :)

Everything with love. Stay safe.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #4 Posted: Mon Oct 19, 2020 3:31 am 
Tengen

Posts: 5202
Liked others: 0
Was liked: 720
We know that you have read some of my books, such as Joseki 1, but which of Endgame 1 - Fundamentals, Endgame 2 - Values, Endgame 3 - Accurate Local Evaluation, Endgame Problems 1 have you read?

You describe the contents of these books as being for mathematicians. They are not. You can recognise a text by and for mathematicians when ca. 95% is proofs, such as https://senseis.xmp.net/?CycleLaw . I have a raw draft of an endgame book with theorems and their proofs, which, when eventually published, you may then rightfully characterise as being for mathematicians.

Applied mathematics is very different: usually, the proofs are omitted and theorems might be stated in less strict language, such as a formula or a value condition. Or buried in examples with numbers already inserted in what would be a formula if still applicable to all examples that could ever occur.

You claim a majority of go players to have a preference for a writing style embedding calculations in sort of natural language. I do not make such a claim or the contrary because we do not know and have not asked thousands of players for their preference. Among go players, there are both math-type and non-math-type preferences so surely there is scope for texts of both kinds.

You try to convince people that the language of mathematics was our enemy. I say: mathematics is the natural language for calculations.

While you praise ambiguity of natural language as an advantage when doing calculations, I praise the unquivocality of mathematical annotation of calculations because it allows to produce the same result in correct repeated application and clearly identify any accidental errors.

Having just looked into Endgame 2 - Values again, if I had wanted to teach as little theory as O Meien, I could have included only one fifth of the contents of my book and replaced the currently short natural language desriptions (such as "[...] the average of two numbers is their sum divided by two [...]") by heaps of long natural language desriptions filling four fifths of the book, burying and hiding the important contents - the calculations themselves. To teach the contents of Endgame 2 - Values, I would have to split it into five books. Doing alike for all endgame books, we would get dozens of them. Such would be the right method for contents to never appear because writing would take too long.

Or have a look into Endgame Problems 1. While you describe my writing as if it were nothing but maths, what the book actually offers is 1) formula with variables, 2) actual numbers inserted and calculated, 3) natural language describing the calculation, 4) footnote characterising the kind of calculation, 5) initial diagram position, 6) Black's move sequence and marks of counted intersections in the resulting position, 7) White's move sequence and marks of counted intersections in the resulting position, 8) more diagrams etc. if there are deeper follow-ups. You will have a hard time finding more detailed calculations elsewhere but, of course, you will continue to criticise me for emphasising the scheme F calculation while missing much larger heaps of natural language because you reject emphasis of the calculations themselves. I emphasise what is important and necessary: the calculations themselves.

You put Ogawa/Davies and O in one basket but they are very different: the former does much wrong while the latter does almost everything right concerning calculations. Ogawa/Davies has great hurdles: ambiguous language including confusing usage of different phrases for the same, awkward annotation of calculation and hiding them within too much natural language and irrelevant side topics.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #5 Posted: Mon Oct 19, 2020 4:19 am 
Oza

Posts: 2759
Liked others: 16
Was liked: 3867
Yu are as unbending as ever, so this is my last post on the topic.

Quote:
You describe the contents of these books as being for mathematicians.


No, I don't. That may be the result but I am talking about books BY mathematicians who insist on retaining a mathematical MINDSET when talking to people who don't usually think that way.

And to claim that "we do not know and have not asked thousands of players for their preference" is an example of that mindset, defying common sense.

"Among go players, there are both math-type and non-math-type preferences so surely there is scope for texts of both kinds." Yes. But I'm bemoaning the lack of material for the latter type (the vast majority), and also the rubbishing of books like Davies/Ogawa, Triumphal gloating over alleged mistakes in such books is distasteful. They have more than compensated for any mistakes through their educational and entertainment, and thus motivational, value (is that value-1 or value-2?), and as I said above, readers are very often capable of coping with and benefiting from mistakes themselves.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #6 Posted: Mon Oct 19, 2020 4:20 am 
Gosei
User avatar

Posts: 1731
Location: Ghent, Belgium
Liked others: 268
Was liked: 798
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
Interesting thoughts, John. Myself I'm a native Dutch speaker, as you know, and I'm an avid speaker of other languages (English, French, Portuguese, Spanish, German, Polish and lately Welsh). When choosing a major in university, I hesitated between languages and math. I chose math because I thought it would be easier to accommodate for my linguistic desires outside the academic environment than my love for math. I have never seen, for myself but also in general, math and languages as two distinct worlds, or ways of thinking, two paradigms to choose between. Yes, you go about mathematical proof in a more formalized way than when interpreting a piece of writing. But the curiosity, the creativity and the rigor of research in either area, is rather similar. So, "I don't think like a mathematician, nor do I want to" is of course a statement no one can deny you making for yourself, nor can anyone deny your approach to matters from a linguistic standpoint. For me it's a false dichotomy. My mathematical background doesn't prevent me from understanding the subtlety of "you can overthink (it is possible to)" vs "you can't overthink (and you must avoid)", let alone that it forces me into a rigid position of stating they MUST be opposite statements. In fact, any rigorous mathematician will argue that "can" is ambiguous and so the statements are not necessarily opposite.

Incidentally, I had to look up "foibles". No problem, I enjoy enlarging my vocabulary. Your writings offer ample opportunity to do so but that begs the question if you are tuned to the MAJORITY of your readership.

That teaser being given, I do agree that when you have to make a choice between mathematical rigor and readable text, the latter will reach a wider audience. I'm not always fond of Robert's terminology either and I believe there must be a way to be precise AND clear. 1-noun and 2-noun constructs are allegedly precise AND clear to a small audience. For wider audiences one needs terminology that resonates with their inner world, for which plain language is better than math. So, it's not the thinking that obstructs clarity, but the interface.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #7 Posted: Mon Oct 19, 2020 5:08 am 
Tengen

Posts: 5202
Liked others: 0
Was liked: 720
John Fairbairn wrote:
Yu are as unbending as ever


You aren't?:)

Quote:
books BY mathematicians


Just because somebody is also a mathematician does not imply that all his writing would be alike.

Quote:
retaining a mathematical MINDSET when talking to people who don't usually think that way.


When I write about calculations, usually, I maintain some mathematical mindset indeed:
- correctness
- appreciating the tremendous value of what is general so can be applied in general
I do so not just for myself but for the readers to profit from both.

O Meien might write:
"
[...text...] (3 - 1) / 2 = 1 [...text...]
[...text...] (8 - 4) / 2 = 2 [...text...]
"

I might write:
"
[...text...] M = (B - W) / 2 = (3 - 1) / 2 = 1 [...text...]
[...text...] M = (B - W) / 2 = (8 - 4) / 2 = 2 [...text...]
"

I prefer to also state the general theory, which here is the formula for a gote move value, so that the reader can understand at a glance that the calculation always follows the same procedure despite different numbers in different examples.

Quote:
I'm bemoaning the lack of material for the latter type (the vast majority),


Sorry but the vast majority of go books (including endgame books with tactical reading) is for the non-maths type, is in fact dull teaching by example. You might be right as far as material for endgame calculation is concerned, unless I should have overlooked lots of old Asian books on that.

Quote:
alleged mistakes


The shame is rather to continue excusing them and pretending their non-existence.

Quote:
coping with and benefiting from mistakes themselves.


Like me. I needed many years to overcome the delay in my understanding of reverse sente caused by such books and many more years to overcome the restricted scope of their traditional endgame theory. So much about "benefiting" from mistakes. Everybody around me learned from those sources so could not help me, either.


Last edited by RobertJasiek on Tue Oct 20, 2020 12:06 am, edited 2 times in total.
Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #8 Posted: Mon Oct 19, 2020 6:21 am 
Oza

Posts: 2759
Liked others: 16
Was liked: 3867
Quote:
Interesting thoughts, John. Myself I'm a native Dutch speaker, as you know, and I'm an avid speaker of other languages (English, French, Portuguese, Spanish, German, Polish and lately Welsh).


Llongyfarchiadau mawr!

Quote:
I have never seen, for myself but also in general, math and languages as two distinct worlds, or ways of thinking, two paradigms to choose between.


You see it, for example, in the way people write technical papers. One reason people in the social sciences (i.e. the arts side) get scoffed at quite a bit is because they try to write like scientists and can't. But in general I rely on normal ambiguity, so when I say 'thinking' I leave it up to you whether to interpret that as, say, thinking process, or attitude, or even prejudices, etc. As I say, that's the beauty of human language.

Quote:
For me it's a false dichotomy.


Of course all mathematicians are capable of thinking like every one else. They were trained that way from toddlerhood. But not everyone (e.g. social scientists) is capable of accurately thinking or writing like mathematicians. That requires special training. I have been heavily involved, through technical translation, in dealing with mathematicians (inc. scientists; e.g. oceanographers modelling wave motions) and have perforce had to try to adapt to their way of talking and thinking sometimes. The fact that there is a difference has been brought to my attention by occasional remarks from them, i.e. when I make a mistake. And whenever that has happened, they have invariably been able to explain the confusion in normal English. QED.

Quote:
Incidentally, I had to look up "foibles". No problem, I enjoy enlarging my vocabulary. Your writings offer ample opportunity to do so but that begs the question if you are tuned to the MAJORITY of your readership.


I'm pretty well attuned to that because of my formation professionelle. And when I do use strange words, usually for entertainment, I tend to do it in a way that the meaning is obvious from the context, or where it hardly natters if it's not clear. That, too, is part of formation professionelle as a journalist (there's an example right there).

As to foible (same root as feeble and French faible, of course), my feeling is that it's one of those words that nearly all English speakers would recognise and understand, usually without difficulty, through reading. Not many would use it in daily speech because it has a literary flavour, but you will certainly hear it in pubs and so on in jokey phrases like "his little foibles" or in barrack-room wisdom phrases such as "human foibles."

And if you like linguistic research, you may wish to look up why many people (not me - and I also split my infinitives) say you have used "beg the question" incorrectly.

Further, if you want to look out for some examples of mathematical speech that mean much less to the average man in the street than foible, I'll start you off with a couple: "it's a function of ...", "of the order of...". It's not just maths. E.g. most people think a quantum leap is a huge leap.

On ambiguity (I used the word earlier on mainly because Bill brought up in a go context) vs precision. I once had to deal with a staff member who was apoplectic on return from a training course where she had been told not to end e-mails with "Don't hesitate to get in touch if you need more help." She had been told that recipients would be offended at the implication they might be hesitant. This was at the start of the PC craze and the big buzzword at the time was 'assertiveness'. I told her not hesitate ending her e-mails in her usual way. But it is a fact that deliberate ambiguity and euphemisms play a huge role in human communications, and precision often makes people bridle (or bristle :)).

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #9 Posted: Mon Oct 19, 2020 7:52 am 
Honinbo

Posts: 10554
Liked others: 3501
Was liked: 3313
John Fairbairn wrote:
I'm bemoaning the lack of material for the latter type (the vast majority), and also the rubbishing of books like Davies/Ogawa, Triumphal gloating over alleged mistakes in such books is distasteful.


Speaking for myself, the feeling is not one of triumph, but of dismay. Back in the '70s I discovered that yose books were riddled with calculation errors and classification errors. {sigh} In professional games, top players rarely made errors that cost a point below temperature 3, where plays gain around 3 points. The textbooks were very good for endgame tesuji, but sometimes correct play was not found because of calculation errors. We're not talking about 1/16 of a point here. I had expected better.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

My two main guides in life:
My mother and my wife. :)

Everything with love. Stay safe.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #10 Posted: Mon Oct 19, 2020 8:05 am 
Honinbo

Posts: 10554
Liked others: 3501
Was liked: 3313
John Fairbairn wrote:
I once had to deal with a staff member who was apoplectic on return from a training course where she had been told not to end e-mails with "Don't hesitate to get in touch if you need more help." She had been told that recipients would be offended at the implication they might be hesitant. This was at the start of the PC craze and the big buzzword at the time was 'assertiveness'. I told her not hesitate ending her e-mails in her usual way. But it is a fact that deliberate ambiguity and euphemisms play a huge role in human communications, and precision often makes people bridle (or bristle :)).


I am no PC fan. Sure, people have a right to take offense, but it might be best if they hesitate to do so. Perhaps as part of her assertiveness training she might have told her critics to sod off. (I might have used a more assertive expression, myself. ;)) Still, there is something to the advice. "Please get in touch if you need more help," might have been both more polite and more effective. :)

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

My two main guides in life:
My mother and my wife. :)

Everything with love. Stay safe.

Top
 Profile  
 
Offline
 Post subject: Re: On "How evaluate double sente moves?"
Post #11 Posted: Tue Oct 20, 2020 12:04 am 
Tengen

Posts: 5202
Liked others: 0
Was liked: 720
Quote:
When I write about calculations, usually, I maintain some mathematical mindset indeed:
- correctness
- appreciating the tremendous value of what is general so can be applied in general


I forgot to mention another aspect, which does not just apply to calculations but which I apply throughout my books consistently:

- For each term, the word or phrase is always the same with the same meaning and, as far as reasonably possible, only used for the term.

This avoids ambiguity, achieves clarity and unequivocality, and prevents the reader from having to a) identify concepts at all and b) distinguish more from less correct or relevant descriptions of concepts. Clear terminology allows generally applicable theory where research has already made such available.

Ambiguous natural language should only be used for optional additional explanations or where go theory has not evolved far enough to only describe it by generally applicable statements.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 11 posts ] 

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


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 post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group