Forum for Dead ko

why would this be so important? [#2708]

Back to forum     Back to page

New reply

Dieter: why would this be so important? (2011-12-12 19:00) [#8990]

Come on Robert. How can this be relevant for strategy or tactics? It's a rules issue! It's about removal of stones and nothing else. This is not beginner's material. I see the reason of existence for this page but aren't you inflating its importance?

RobertJasiek: ((no subject)) (2011-12-12 23:03) [#8993]

It is as important as distinguishing all other dead from alive stones!

It is much more important than all those ko shapes and types that occur only every 100th to 100,000th game such as double ko death (like in a nakade joseki) or like triple ko. If you doubt the importance of dead kos (as "rare" as occurring once every 3rd game or so) then treat as expert freak pages all those hundreds of pages about shapes and aspects rarer than 1:3...!

BTW, most kyus I have seen misunderstood dead kos because they missed this concept as a concept. Strategic mistakes and wasted thinking time are the consequence.

Slarty: ((no subject)) (2011-12-12 23:40) [#8994]

It seems exactly as important and as easy as distinguishing dead stones from live stones... The current page doesn't concisely convey to me the point, and seems to tie it to rare examples of double and triple kos.

RobertJasiek: Re: ((no subject)) (2011-12-13 07:59) [#8996]

Feel free to explain this aspect in greater detail or with more examples! - None of the examples shows a double ko or a triple ko. All of the examples show dead kos. That you still confuse them with double or triple kos shows how important the topic is; dead kos are played differently from double or triple kos! Also note that the examples are not rare but occur regularly. Only your confusion with double or triple kos lets you think that they would be rare.

Slarty: Re: belated, brief, annoyed (2011-12-14 19:09) [#9068]

It would make sense to call them dead double kos and dead triple kos, as long as we are calling dead repetitive shapes that will never be played kos. I would never write more than one hundred words without considering replacing them with a picture or silence.

RobertJasiek: Re: belated, brief, annoyed (2011-12-14 19:41) [#9069]

Two "dead kos" don't make an interesting double ko. There is double disturbing death though (recall: it is not a "dead ko"); the name evolved from Ing's terminology. - For "dead triple ko", I do not see a need; it just consists of three "dead kos". Re: belated, brief, annoyed (2011-12-14 19:47) [#9071]

Isn't the irony that the dead ko is not really a ko either? Or more to the point, it does not make "an interesting ko". --Hyperpape

RobertJasiek: Re: belated, brief, annoyed (2011-12-14 20:01) [#9074]

Life, unsettled and death status, ko status and ko types are dynamic: they can change during the game. That something is currently uninteresting (not fitting perfect play) during the game is the normal case: most parts of a current position are not hot enough. Some of them can have unsettled groups or still too small kos; others can have live groups, dead groups, live ko stones or dead ko stones while opposing multiple threats could change their status. Such kos consist of "basic ko intersections" or "local ko intersections", see Ko in general?. When later it becomes interesting to play in such kos, then (supposing correct play by both players) their intersections become also "global ko intersections). Hence I do not see any irony but what I (and your problem books) see is kos all the time. What does change is their urgency.

Slarty: Re: belated, brief, annoyed (2011-12-14 22:48) [#9081]

You should have seen some irony. Your terminology will only become more and more tortured if you try to assign it any meaning other than the intuitive one.

RobertJasiek: Re: belated, brief, annoyed (2011-12-14 23:22) [#9083]

I saw your bad attempt of creating irony...

My terminology is not "tortured". It is as consistent as the history of its invention allows.

My terminology does not become more "tortured". What you witness is your too low degree of willingness to acquire a proper understanding of it. Have you not noticed how exceptionally powerful my terminology is that I could give you such a profound answer, which explains a) why we call many things "kos" although they are currently not worth playing and b) in general with which major condition kos are worth fighting? From my terminology and reading its definitions, you could see the answer: A ko is worth fighting because otherwise winning the game is impossible. In retrospect, I have been extraordinarily astonished how close my intuition had been to the global ko intersection definition. Everybody should have had an intuitive understanding of its central condition. Surprisingly, the most fundamental aspects are those that are the hardest to define in words. Now the definition's words confirm go players' intuition and vice versa.

You could hardly have found a concept of mine closer to our intuition. Now that is ironic;)

Slarty: ((no subject)) (2011-12-15 06:04) [#9085]

It is precisely because fundamentals defy definition in words that terminology is doomed to be inconsistent. Yes, I am a window shopper, and since "dead ko" doesn't do it for me, and neither do the first few paragraphs of reasoning, I would of course in a second ditch the theory in favor of a little diagram, caption "Dead, not ko."

For you, perhaps another suggestion: "Go, not ko"

RobertJasiek: Re: ((no subject)) (2011-12-15 08:36) [#9086]

Fundamentals do not defy definition but can be related to hard-to-find definitions.

Dead ko is a ko. Therefore a caption "Dead, not ko." would be wrong.

The name dead ko shall express that, regardless of which player can first make a ko capture in the ko, one side is dead while the other side is alive. For dead non-ko strings, one side is dead while the other side is alive. Due to this relation, the name "dead ko" is chosen well.

The name is also chosen well in the contexts of all types of basic ko and of the earlier Ing-ko-rules-related history of such terms. Re: ((no subject)) (2011-12-15 08:59) [#9087]

The name is also chosen well

The main objection is that no name needs to be chosen at all. This has already been suggested several times, but I'll have another go just because I think it is great fun.

We already have the noun "ko", which takes its rule technical (as opposed to life-and-death status) sense in this situation, and we have the adjective "dead", which exactly describes all the interesting properties of the situation.

Combining an adjective with a noun is one of the more fundamental properties of the English grammar, so should we ever be interested in speaking of a "dead ko", we can (present company excluded) do so without feeling a pressing need to coin a new technical term.

RobertJasiek: Re: ((no subject)) (2011-12-15 10:57) [#9088]

As already explained on this and the parent page, "dead" + "ko" <> "dead ko". Also note that "dead" used as an adjective is a property of a string while a ko consists of occupied and empty intersections.

Dieter: Re: ((no subject)) (2011-12-15 11:16) [#9089]

I think such violations of basic language should indicate you're on the wrong track

RobertJasiek: Re: ((no subject)) (2011-12-15 11:40) [#9090]

No; it simply means that a practical term is used instead of a long, precise phrase nobody wants to use: "(basic) ko so that exactly one side is dead and it is neither a double disturbing death, a closed death nor a 1-eye-flaw".

BTW, wrong track, what a suggestion! After reading my paper, tell me whether you find any missing disturbing ko type! The basic case structure is complete, so all you could find is something exceptional and spectacular like 1-eye-flaw. A theory with such a degree of completeness is evidence for being on the right track or possibly having already the final description of classification.

Dieter: Re: ((no subject)) (2011-12-15 11:56) [#9091]

If a "dead ko" is not a "dead" "ko", there is something wrong, period. Your theory may be very valuable, I will not contest that until reading your paper, but coining a term that is either trivial or misleading, is not a good idea. Re: ((no subject)) (2011-12-15 12:02) [#9092]

If Dead Ko, doesn't refer to a 'ko shape' that is already dead, then it is a ridiculous name. It's in the same realm of stupidity as natural situational superko. However, this is no cause for alarm, as there is plenty of time to change the name to something more sensible. How about Robert Ko?

RobertJasiek: Re: ((no subject)) (2011-12-15 12:13) [#9093]

We can discuss "natural situational superko" at an appropriate page. Hints: "natural" modifies "situational", which is a term with a specific go rules meaning.

Of course, dead ko does not refer to a ko shape because my research has improved on Ing's mistake of shape dependency. My ko type terms refer to move-sequences!

Dead ko does express that something is dead: one side's strings. Since this is correct, the name is not ridiculous but chosen well. Re: ((no subject)) (2011-12-15 12:31) [#9094]

Robert, natural situational superko does not occur in nature. Therefore the term is a misnomer.

RobertJasiek: Re: ((no subject)) (2011-12-15 12:29) [#9096]

"natural" can, like many words, be used to describe something either real or figurative. Here the meaning is figurative. Now that was easy! You also don't criticise all those professionals for figuratively speaking of the natural flow of a game, do you? Re: ((no subject)) (2011-12-15 12:40) [#9097]

Natural flow is also a bit of a misnomer. It's basically saying, "oh look how nice this seems, isn't that great. It all looks so easy doesn't it. You see how easy this game is listening players? Buy my book and you can play like this too." However, we can emphaise with what the pro is saying there, because it does look so simple, and yes, it does look like a correct way of playing. Once they put down the sequence of moves on the goban, you agree with them.

Now turn back to natural situational superko and (unnatural) situational superko. The problem there, and this is rather a fundamental problem, is that one superko type just doesn't feel any more natural than the other superko type. If you had a pro lecturing you on formulating superko rules, and suddenly he started talking about natural flow... well that would just be ridiculous. :)

RobertJasiek: Re: ((no subject)) (2011-12-15 15:03) [#9098]

If you want to discuss superko, on some appropriate page.

RobertJasiek: Re: ((no subject)) (2011-12-15 12:26) [#9095]

If you want to express the difference between dead applied to string and dead applied to ko (where dead refers to strings related to the ko: one player's adjcent strings or his, if any, ko stone) linguistically, you might say "ko-dead ko" instead of "dead ko", "disturbing ko-death" instead of "disturbing death", "closed ko-death" instead of "closed death", "disturbing ko-life" instead of "disturbing life" etc. Possible but then a) terminology of the past 3 decades would be void and b) people would start complaining about unreadable because too long phrases. I prefer the current, short forms also because the different nature of application (either to only strings or to both strings and empty intersections) indicates that there is a difference in meaning. Re: ((no subject)) (2011-12-15 15:25) [#9099]

RobertJasiek wrote:

As already explained on this and the parent page, "dead" + "ko" <> "dead ko".

If "dead ko" does not mean "dead ko", then you must have an ill-defined term there somewhere.

Also note that "dead" used as an adjective is a property of a string while a ko consists of occupied and empty intersections.

No, you are wrong on a very fundamental level. Adjectives are not properties of nouns, they _describe_ properties of (for example) nouns. For example, the adjective dead typically describes the property deadness that is common to dead stones, dead strings and dead groups. From a strategic point of view, all of those should be treated in much the same way, so it makes sense to treat "deadness" as general concept that is common to any clump of stones, no matter what the shape, as long as they satisfy the basic condition of being dead.

Now, if one claims that such a useful general concept as "dead" should be partitioned so that it means different things when applied to certain nouns, there must be an exceptionally good reason to justify such an argument.

Whether such a reason exists or not is quite clear: it does not.

Instead, the whole commotion is caused by the simple fact that there is an author who has taken leave of his usual practise of using hyphens to indicate that he is redefining common words in a way that can only be understood by reading the glossary very carefully.

There would be no possibility of confusing dead-ko (something rj-specific) with a dead ko (a ko that is dead), if it is so very important that the two meanings be different.

RobertJasiek: Re: ((no subject)) (2011-12-15 20:39) [#9100]

Historically it might have been better if Ing or his translator(s) had not invented the term "disturbing ko". Once invented, Ing noticed that there are "disturbing life" and "disturbing death". One could imagine that these two terms could instead have been: "live ko" and "dead ko". Then "dead ko" could have been split into "double dead ko", "closed dead ko" and, let's say e.g., "ordinary dead ko" (the thing currently called "dead ko") but "ordinary" is not particularly cute.

Things are not that straightforward though because also "fighting kos" or "semi-stable kos" have properties of life and death.

History of those terms has been different though. Now it is rather too late to change all the terms because they have been used dozens of thousands of times. The effort of implementing such a change would be tremendous. E.g., I have been using "dead ko" (German: "totes Ko") for several years now.

A strong reason for speaking of "dead" or "death" for a ko (in contrast to usage of stones / strings / groups) is the historical usage originating from Ing in his phrases mentioned above. One can argue whether his (or his translator(s')) choice of words for the intended concepts was doubtful. We have to live with it though. Well, ok, unless you want to create a major revolution of words and phrases for possibly all the ko types related ko terms. If so, suggest and we can compare and weigh also the impact of effort implied by the history.

I use hyphens in strict rules research context, where unambiguity is a first command, while I tend to use ordinary spelling in regular contexts.

Actually dead-ko is a nice suggestion. I will reflect this spelling but I need quite some time to think about it because the question arises whether also other ko types need hyphens for clarification of the intended meaning. In German I would have no hesitation but ordinary English is much less fond of creating hyphen terms.

A very important consideration is whether everybody would immediately understand that dead-ko does not equate dead + ko or whether not potentially many would start to edit and "correct" dead-ko into dead ko, creating much more confusion than ever. Re: ((no subject)) (2011-12-16 05:36) [#9101]

Bill: If we agree that "dead" is a property of strings, the fact that the strings in question form a ko (or superko) shape is no bar to ordinary usage.

It is true that whether some ko shapes are alive or dead depends upon which rules you are using. But the meaning of "dead ko" is derivable from the meanings of "dead" and "ko". It is true that it is not derivable from some meanings of ko, but that does not matter. If I say that a player is green, I am not referring to his color, but am using a different meaning of "green". Language works that way. :)

hyperpapeterie: ((no subject)) (2011-12-13 04:50) [#8995]

Compromise: it's tremendously important for life and death, but it's also trivial to recognize in most cases.

RobertJasiek: Re: ((no subject)) (2011-12-13 08:02) [#8997]

It is "trivial" as soon as one does not confuse dead kos with double or triple kos any longer. In (fake) capturing races, it can sometimes be difficult to distinguish dead ko from other ko types.

tapir: ((no subject)) (2011-12-13 11:05) [#9002]

Some beginners are pretty definitionist in their approach to the game. They learned sth. then they apply regardless of context. This leads to the placement plays in 3 pt. & 4 pt. eyes even when the group has half a dozen of other eyes, who recognize a ko-esque shape and start playing it regardless of context... the usual way to learn to correct this is to give more context and learning when to apply what. Robert's approach is - as far as I can tell - the opposite, he tries to provide complete definitions instead of limiting the field of application. All his terms are general terms, they are not about certain distinctions but about a complete mapping of the phenomenon. (This effort is ultimately misguided and a waste of Robert's time in my personal opinion, but still he has results.)

All the other terms he gives depend on the basic terms, which might seem useless to you (or me), as they don't come along with their own context. I don't know about dead ko, but I had a discussion with Robert reg. string connection, which by itself is a trivial (and pretty useless) term. However, if we remove it all his other connection terms become incomprehensible.

This badly connects to other terminology though, and we sometimes pushed it aside by adding an "expert" keyword on those pages. In the long run this won't work, I believe a Robert Jasiek terminology page is better suited. And we already have one Go Terms By Robert Jasiek, ko terminology can get a section there or an independent page.

Dieter: Re: ((no subject)) (2011-12-13 11:14) [#9003]
Dead connection  

My hobby horse when teaching go is connection. I stress the long term benefits of connection. Hence techniques for cutting and connecting are discussed. I never felt the need though to define a dead connection, since a situation like this one is not about connection but about life and death. Indeed, beginners who are indoctrinated by my urge to connect may play such a move "because connecting is so important" and then I should intervene and point out that connecting dead stones creates one big dead group.

I totally agree with tapir that providing the bigger picture (or "context") is best and that combining the right concept with the wrong context into one definition is misguided.

That said, we can show hospitality to such concepts and give them a chance to survive and even prosper (since we can be wrong in our judgment) but I'm not inclined to host them at the front door.

RobertJasiek: Re: ((no subject)) (2011-12-13 12:05) [#9009]

Citation from the Types of Basic Ko paper:

"The roughly estimated frequency of ko types in actual games is worth remarking:

  • 1:1 - fighting ko
  • 1:3 - dead ko
  • 1:300 - double disturbing death
  • 1:10,000 - active disturbing life
  • 1:50,000 - inactive disturbing life
  • never - semi-stable ko
  • never - 1-eye-flaw
  • never - closed death"

Hence dead ko is the second most important ko type! It is about a 100 times as frequent as double kos, 3,000 ~ 20,000 times as frequent as double ko seki, triple ko, quadruple ko, quintuple ko, eternal life, moonshine life etc. and millions of times as frequent as molasses ko or 1-eye-flaw. Why are there so many SL pages about such rarities, except for expert fun? Knowing the frequent basics is much more important!

HermanHiddema: Re: ((no subject)) (2011-12-13 12:15) [#9010]

The frequency is irrelevant, because the concept is nonsense. We can prefix pretty much any go concept with "dead" to create a similarly nonsensical concept. Dieter has already given the example of "dead connection". We might then create concepts for "dead ponnuki", "dead ladder", "dead hane", "dead ikkentobi", "dead keima", "dead bamboo", etc, etc, etc.

Understanding the concept of "dead stones" is sufficient for all of these.

RobertJasiek: Re: ((no subject)) (2011-12-13 12:44) [#9015]

Calling frequency irrelevant or the concept "nonsense" is so stupid that I do not waste time to discuss such an opinion.

"dead connection", "dead ponnuki", "dead ladder", "dead hane", "dead ikkentobi", "dead keima", "dead bamboo" etc. do not need own terms because a combination of the life+death status "dead" with one of the shape etc. terms explains the same as a combined term would explain.

With "dead ko", there is no such 1:1 mapping between "a ko that is dead" and "dead ko". Hopefully a lemma applying a particular ruleset could prove that each dead ko is dead. The converse is not true: Not each ko that is dead is a dead ko! For example, a double ko death (also called perpetual ko) can also be dead. Its strategic behaviour differs very much from that of a dead ko: a double ko death can be an supply of "arbitrarily many" ko threats - a dead ko does not provide ko threats during the game until incl. the regular endgame.

RobertJasiek: Re: ((no subject)) (2011-12-13 11:50) [#9007]

1) Learning by considering more context and timing of application and 2) complete and the most generally applicable definitions are not contradictory. I as a player use both for a better understanding. It is not necessary to understand the most generally applicable definition of dead ko for understanding dead ko as a player in most practical cases, unless you are tderz and want to overinterpret every informal description as a most generally applicable definition. IOW, it is possible to improve on an informal description in the direction "usually not worth playing at all until the post-endgame phases".

The power of really basic terms is that they do not need context! Don't miss the context but appreciate its possible absence!

Why are you worried that "dead ko" badly connects to other terminology? It is a newly invented term. So no connection to earlier usage is a necessity! Traditional go theory failed to provide any description for the concept at all. About that you should be worried!

"Waste" of research time and "string connection" are off-topic.

Dieter: Re: ((no subject)) (2011-12-13 12:25) [#9013]

I'm sure that both the alive 5-17 point and the dead 5-17 point frequently occur in games, much more often than most concepts that get a definition here. That doesn't justify them being a concept though.

As said, I don't mind about the page's existence. What bothers me is its pretention to convey something important. Yes it is important to understand when stones are dead. Yes it is important to understand ko. But a "dead ko" is merely a statement that "being dead" overrules all other concepts. Like the "dead connection", or the "dead bamboo joint" or the "dead table shape" ... The list is infinite. Therefore it's not sensible to combine context with shape in separate definitions. You suggest there is something specific to a "dead ko", that it is some kind of a ko type, like a "two stage ko". That's not true, it's a whole other dimension. You'll need to create separate pages for any ko related concept, adding "dead" to it. Dead two stage ko, dead approach ko, dead direct ko ...

I'll admit there are other pages like this on SL, like the "false ko-threat" I remember, or even "false eye". There may be instances where it is useful to point out that things are often not what they appear to be. And "dead ko" may be such a case. But I wouldn't make it seem like it's a separate concept.

RobertJasiek: Re: ((no subject)) (2011-12-13 12:47) [#9016]

See my other related reply and read my paper.

Dieter: Re: ((no subject)) (2011-12-13 13:18) [#9018]

That's not how it works: either you create a concept here and convince the rest of us that it is not nonsense, or you keep that concept to your papers. It's not for us to prove it is nonsense and not for the SL readers to understand its importance by reading your paper.

You must be the only person I know quoting yourself as a reference and being more convinced of being right as the crowds gather to argue you are wrong. Democracy is not truth, but SL is a collaborative website.

Anyhow, there is enough of your work I'm enthusiastic about so I'll waste no more energy in this particular disagreement.

RobertJasiek: Re: ((no subject)) (2011-12-13 14:07) [#9021]

Usually concepts are not created on a wiki but created elsewhere, e.g., where the research is being done.

When somebody claims "nonsense" and somebody else claims "sense", then either side has the same burden of proof.

tapir: Re: ((no subject)) (2011-12-13 14:45) [#9023]

It isn't democracy either. We are not voting on terminology.

I personally tend to dismiss this page and or the term "string connection" to give another example, but as Robert strives for context-free definitions, at the cost of dependence on terms as this one, it might well be necessary to understand the other terms he has to offer and which I personally like to have on SL (but not where beginners hit them unplanned when taking their first or second tour). In fact, I believe he tries hard not to produce unnecessary terms, without a notion of pragmatics many terms are necessary, however.

I believe he would be a better go researcher if he could change his style from completely mapping a phenomenon to providing useful distinction, where it matters. And he would have more time for doing other things. However, this is not in my hands, I just would say we can't discuss the merits of a single of his terms without the other terms.

@Robert: Traditional go theory has no concept about it because it doesn't need a concept "dead ko". For me it is enough to know, how your theory needs this term, and you don't have to claim traditional knowledge was deficient for a lack of this concept.

@Dieter: Robert has no "dead connection" term, I don't believe he would make up such unnecessary term. He has string connection, however, what other people would call at times connection without attributes and at times solid connection. There it is pretty clear how the thought process goes, without pragmatics you need a term that always fits, hence string connection. Using sometimes this term, and sometimes another would make Robert uneasy, I believe.


For Robert:

Del rigor en la ciencia

En aquel Imperio, el Arte de la Cartografía logró tal Perfección que el Mapa de una sola Provincia ocupaba toda una Ciudad, y el Mapa del Imperio, toda una Provincia. Con el tiempo, estos Mapas Desmesurados no satisficieron y los Colegios de Cartógrafos levantaron un Mapa del Imperio, que tenía el Tamaño del Imperio y coincidía puntualmente con él. Menos Adictas al Estudio de la Cartografía, las Generaciones Siguientes entendieron que ese dilatado Mapa era Inútil y no sin Impiedad lo entregaron a las Inclemencias del Sol y los Inviernos. En los Desiertos del Oeste perduran despedazadas Ruinas del Mapa, habitadas por Animales y por Mendigos; en todo el País no hay otra reliquia de las Disciplinas Geográficas.

RobertJasiek: Re: ((no subject)) (2011-12-13 15:21) [#9024]

Please discuss string connection at its page. I do not reply to it here because otherwise this thread would become endless. For the same reason I do not discuss research style here; if you want to discuss it, do it at an appropriate page, forum or email me.

Traditional go theory does not know dead ko because it overlooked it conceptually.

Looks like a Roman language, which I do not know. Do you want to demonstrate that natural languages I do not know exist? Trivially. Is learning such a language relevant for go? No. Is learning my go terminology relevant for go? Sure.

tapir: Re: ((no subject)) (2011-12-13 16:34) [#9025]

I don't discuss anything, it was an example. I just voiced my opinion on your thought style, which makes terms others might see as doubtful, necessary. This isn't meant as a judgement on your results, but a tentative defense of articles such as this one - as they might be necessary for understanding pages on other of your terms. You might want to recognize the fact, that there are people who never felt the necessity to invent a term like dead ko, not because of a mistake, but because they think differently. They simply don't see "dead ko" as ko, but as dead stones.

PS The short story posted can be found by searching for "On Rigor in Science".

RobertJasiek: Re: ((no subject)) (2011-12-13 17:11) [#9028]

It is a mistake not to see the stones in a dead ko as being in a ko because the ko rule(s) do apply.

From my player perspective, my thought style underlying string connection and dead ko is: These concepts enable me to understand much more clearly and better aspects I had great difficulties with because traditional go theory does not explain them well, if at all. With consequences like for a long time overlooking the possibility for such moves; I had to discover 1 myself:

string connection  

For the scepticists: A 9p confirmed to me that locally 1, a and b are equal options and that 1 was a good choice in a game of mine.

People who "think differently" would not easily discover 1 at all because they have not been taught the underlying concept of the move type string connection. Traditional go theory (AFA I have seen it in literature or from professionals) had a gap there!

HermanHiddema: Re: ((no subject)) (2011-12-13 21:39) [#9053]

Congratulations. This is, without a doubt, by far the stupidest thing I have ever heard you say.

RobertJasiek: Re: ((no subject)) (2011-12-14 07:00) [#9056]

Do you have other references for that move, especially available already in the mid 90s? Don't you recognise its value?

HermanHiddema: Re: ((no subject)) (2011-12-14 14:02) [#9059]

Learn to Play Go (volume I, 2nd ed.) by Janice Kim and Jeong Soo-hyun, pages 20-23 and 104-114.

RobertJasiek: Re: ((no subject)) (2011-12-14 14:22) [#9061]

Interesting; the first book I see or hear about with this move. E.g., the Nihon Kiin Tesuji Dictionary, 1986 does not have it. Roughly what does Learn to Play Go say about the move?

HermanHiddema: Re: ((no subject)) (2011-12-14 15:33) [#9063]

Stones are solidly connected if they are on adjacent points connected by a line. Other types of connections exist (tiger's mouth, bamboo, knight's move), which may be better depending on the circumstances. If you do not connect, the opponent can cut. Connected stones are strong, cut stones are weak. Plus lots of examples and exercises. Re: ((no subject)) (2011-12-14 16:52) [#9065]

So it is a chapter or two about connection / cut rather than 4-4 josekis? (The phrase "solid connection" we discussed earlier:) )

HermanHiddema: Re: ((no subject)) (2011-12-14 16:57) [#9066]

Yes. It is a beginners book, starting with the rules of play.

RobertJasiek: Re: ((no subject)) (2012-01-30 19:00) [#9244]

Your claim that Learn to Play Go 1, 2nd ed., p. 20-23 and 104-114 explained

string connection  

is a totally misleading description of that book's contents, which merely lists a few shape types like bamboo, solid (in that book's terminology, meaning string connection there), knight move and gives the roughest possible advice "connect your stones". This is about as useless as suggesting the following move because it applies the advice by creating a "solid" connection.

string connection  

HermanHiddema: Re: ((no subject)) (2012-01-30 20:10) [#9245]

That is not what I claimed.

You claimed: From my player perspective, my thought style underlying string connection and dead ko is: These concepts enable me to understand much more clearly and better aspects I had great difficulties with because traditional go theory does not explain them well, if at all. With consequences like for a long time overlooking the possibility for such moves; (my emphasis)

You only listed the discovery of that specific move in that situation as a consequence of your understanding of "string connection". Your more general claim was that string connection (a.k.a. solid connection) is not explained well, if at all, in traditional go theory. I merely gave an example of literature that explains the concept of string connection.

EDIT: Also, further up, you claim: People who "think differently" would not easily discover 1 at all because they have not been taught the underlying concept of the move type string connection. Traditional go theory (AFA I have seen it in literature or from professionals) had a gap there!

Another claim that previous literature does not contain the concept of string connection.

RobertJasiek: Re: ((no subject)) (2012-01-30 23:26) [#9246]

Now I feel better that part of our misunderstanding is resolved (related to the 4-4 joseki pattern move discovery).

I maintain that prior literature has not explained string connection (under whichever name) well. Learn to Play Go 1 is no exception to that. It explains differently from more frequent ambiguity elsewhere but its explanation is not good: there one does not get an understanding beyond "connect your stones at all" of why / when one should play a string connection or one of the other mentioned connection shape kinds.

That book introduces string connection as if it were just one of several (possibly many) shapes. Given that book, now I am convinced though that it preceded me (with year of publication) with the introduction of string connection as a concept (although I am not happy with the description there). The shape context is a weak approach though. My three and only three major types of connection nature (string, direct, indirect) put string connection in the context of high importance it deserves and allows further important definitions such as n-connected and then influence / thickness. Learn to Play Go 1 has noticed the existence of string connection but overlooked the central importance.

HermanHiddema: Re: ((no subject)) (2012-01-31 10:28) [#9247]

My quoting of LTPGv1 was only in the context of your claim that there was no prior literature, I did not intend to make any claim as to whether it explains the concept well, or gives it enough attention.

Of course, one can argue about the relative importance of concepts almost endlessly. Every author and teacher has their own preferences, both in subject matter and in teaching method. I know, for example, that you dislike the popular method of teaching by examples only, instead of attempting to derive basic concepts from the examples and teaching those. Others disagree, and I am not going to claim either camp is absolutely right or absolutely wrong.

RobertJasiek: Re: ((no subject)) (2012-01-31 11:14) [#9248]

A good timing for EOD before we enter the more fundamental, general discussions? :)

HermanHiddema: Re: ((no subject)) (2012-02-07 22:42) [#9257]

Yes. This would certainly be the wrong place for such discussion. :-)

BTW, the concept of "string connection" is also specifically introduced (again as "solid connection") in Iwamoto Kaoru's Go For Beginners, first published in 1972, on pages 13-14 (pages 23-24 in the 1976 edition). Re: ((no subject)) (2011-12-14 14:19) [#9060]

nice to see a pro with a sense of humour

RobertJasiek: Re: ((no subject)) (2011-12-14 14:31) [#9062]

During the several European congresses when Muraoka Shigeyuki 9p commented about my games, he was serious. I have no reason to assume he would have been making a joke and his comment style was essentially the same as during similar comments in earlier years. In case of non-standard moves, I asked him several questions to be sure to have understood him correctly. Re: ((no subject)) (2011-12-13 16:42) [#9026]

It is misleading to say traditional go theory overlooked dead kos. Traditional go theory may have overlooked the value of delineating the concept of dead kos as such. Or perhaps they overlooked certain general principles or algorithms for recognizing dead kos (etc etc). Certainly traditional theorists knew about such shapes and that they need not be fought as ko.

Nothing is gained by exaggerating the differences between traditional go theory and your own approach. --Hyperpape

RobertJasiek: Re: ((no subject)) (2011-12-13 17:14) [#9029]

Thx for the correction.

reply ((no subject)) (2011-12-13 11:55) [#9008]

isd: I had a quick look at this page. It looks like a unnecessary term to me. It's dead shape, fine. It looks like a ko, fine. Does it need its own page, or even its own name.... no. I've seen this called 'fake ko'. Page looks basically uneducational to me, can't say I'd bother with it.

RobertJasiek: Re: ((no subject)) (2011-12-13 12:26) [#9011]

The term is necessary. Apart from expert application, the term is necessary because it guides kyu players to an understanding, which most of them do not have. Instead many confuse dead kos with kos worth fighting or with double or triple kos magically regaining life at the game end. They do not have an easy "is obviously dead" understanding. More likely, they try to find an understanding with complicated, arcane rules-related explanations such as pass for ko rules. I have seen all this and tested dozens of kyus. All had a wrong understanding of the life and death status or could not explain why an adjacent group was dead. Dan players at least know the death status but most still cannot explain correctly why an adjacent group is dead during the post-endgame phases.

Do not mix your understanding of the death status with ordinary players' missing understanding. The most active SL users are an exception because they are skilled in such questions because there is quite some expert information here.

Whether "dead ko" needs its own term is a matter of usage. A player could speak of dead strings / groups. As soon as one studies go or rules theory on an expert level, the term becomes necessary because principles etc. for dead kos versus for other ko types versus for non-ko dead strings can differ.

Dead ko needs its own page because at other places references can want to refere to dead ko while not also referring to non-ko dead strings etc.

"fake ko" is imprecise. It could mean dead ko or mean something else (e.g., crushing the ko). With such ambiguous terms, everybody is confused and experts cannot work.

If the page looks uneducational, feel free to provide more information. Re: ((no subject)) (2011-12-13 12:19) [#9012]

isd: As mentioned above, the correct point to explain this mistake/situation seems to be upon a page on death.

RobertJasiek: Re: ((no subject)) (2011-12-13 12:30) [#9014]

Apart from reasons stated above, a general page on death would be overloaded if there were more than a sentence that also kos can be dead and currently not worth playing.

reply ((no subject)) (2011-12-13 17:19) [#9030]

Bill: Why is this discussion so important?

I would not call dead ko a go term. Its meaning follows from the meaning of dead and one meaning of ko. One may regard the page as superfluous, but Robert feels the need to talk about dead kos. That is prima facie evidence that some other people will find what the page says useful. To me, that is reason enough to retain the page. Is SL not large enough to encompass different points of view?

X Re: ((no subject)) (2011-12-13 17:21) [#9031]

I don't think anyone suggested deleting it. It's more that it's nor really relevant for inclusion in these new ko paths.

RobertJasiek: Re: ((no subject)) (2011-12-13 17:31) [#9033]

If dead ko were not included in ko paths, then why include double ko, triple ko, molasses ko etc.? Re: ((no subject)) (2011-12-13 17:37) [#9035]

I imagine that double ko and triple ko would commonly interest people. Hence their inclusion in the main paths.

RobertJasiek: Re: ((no subject)) (2011-12-13 18:52) [#9038]

A bad culture. Rarities are celebrated, so new players get interest in them instead of spending more interest in the frequent fundamentals. I know, rarities are fun; this does not mean though that players should be separated from some kinds of frequent fundamentals. Let them read all - the frequent and the rare!

RobertJasiek: Re: ((no subject)) (2011-12-13 17:30) [#9032]

To repeat, the meaning of "dead ko" does not follow from the sum of "dead" and "ko". With the same logic, you might say the same about "double ko death" ("perpetual ko", the thing occurring in big nadare) - also wrong!

Dieter: Re: ((no subject)) (2011-12-13 17:37) [#9034]

It's tiresome for all of us, but I'll repeat that as soon as one active member sees a purpose in a page, that's enough to keep it. The problem is that Robert continues to claim that "dead ko" is as important a concept as "double ko" or any other ko type. The rest of us thinks it is nonsense at worst or a misguided way of presenting two concepts or an intricate concept only Robert understands at best. And the more resistance there is, the more important it seems to become in Robert's missionary mind, since the need for people to be educated in his way becomes ostensibly bigger.

reply Proposal (2011-12-13 17:57) [#9036]

I suggest that all of Robert's pages that describe a go term invented by him should have -(RJ) appended to the page title. This would be helpful to Robert as he would not have to worry about 'namespace collisions', and everyone else would get a 'heads up' about the content of the page.

RobertJasiek: Re: Proposal (2011-12-13 18:48) [#9037]

Things are not so easy. Some terms I have invented, other terms others invented and I have defined. Go terms are not my property but everybody can use them. Re: Proposal (2011-12-13 19:50) [#9040]

OK, lets also include all the pages that consist of your definition of previously existing go terms.

RobertJasiek: Re: Proposal (2011-12-13 20:07) [#9044]

Would you wish to mark all traditional terms as "traditional", all terms that acquired a new meaning after doubtful initial translation from Japanese to English as "Japanese origin, new English go term meaning", "snapback" as "snapback (James Davies)", terms invented by Ing Chang-ki as "invented by Ing Chang-ki" etc.etc.etc.?! It is a fact of go terms that at some time they were invented or converted from careless natural language into specialised go term meaning. It is a much better idea to create history pages for go terms, at least if people creating them would actually fill them instead of leaving many more gaps than contents, see e.g., Go Terms By Robert Jasiek, which contains only about 3% (?) of the non-private terms invented, defined or redefined by me. Re: Proposal (2011-12-13 20:21) [#9047]

No, 'traditional' can be assumed if absent. Invented by Ing Chang-ki? - Yes thats a good idea Robert. Invented by James Davis? - Yes why not. Proper attribution of these new ideas is important isn't it?

RobertJasiek: Re: Proposal (2011-12-13 20:51) [#9051]

Nothing is wrong with proper credits. But they should be at appropriate places - not where they distract from searching, reading, learning. Re: Proposal (2011-12-13 19:18) [#9039]

I too disagree. So far as those concepts are important, they should be included in the main flow of the discussion, though not to the exclusion of other terms.

The biggest problem is that there are many concepts that Robert uses but which the rest of us have yet to fully comprehend or evaluate. Incorporating them is currently very difficult. A namespace would be at best a band-aid. --Hyperpape Re: Proposal (2011-12-13 20:02) [#9043]

Nothing in my proposal concerns including or excluding Robert's definitions from any discussion. Whilst my proposal is almost certainly inadequate in solving the 'biggest-problem', it goes in the right direction - it allows people to easily distinguish well-known and well-(mis?)-understood terms, and those new and not well-understood terms of Roberts.

RobertJasiek: Re: Proposal (2011-12-13 20:17) [#9046]

The main problem is MUCH bigger: There are very many go terms of many origins or historical transfer paths. From whichever source, some terms are much more central and important than other terms. E.g., most of the Japanese terms still in Japanese I simply do not know because I expect about no benefit from it; by experience Japanese terms are overloaded with peculiar shape details. OTOH, there are also players liking to learn Japanese and becoming stronger by studying shape details. So maybe even all that, what I might call, rubbish still has some use for some other players. So why should I create an index of doubtful Japanese terms? Let everybody use his preferred vocabulary!

tapir: Re: Proposal (2011-12-13 19:52) [#9041]

I thought all had already agreed that better demarcation between basic/intermediary and advanced and specialist material would be helpful, no?

Dead ko, isn't really an advanced concept but it is a term that likely many won't use, but which may have its merits in the framework of Roberts terminology. Currently we are trying to use the difficulties to manage this demarcation, but it isn't necessarily my method of choice in the long run. I personally will keep on adding to the Go terms by Robert Jasiek page that I created and consider one or two topic paths. Maybe it might be useful to make one of the paths a taxonomy including all pages with proper attribution. Also, there is no necessity in getting hyped up about what is listed in which path, for both the current path structure is most likely not the final one and the paths are nothing but tools for navigation. Although what we try to do among other things is to add some user guidance for beginners that they don't stumble upon a (ko) monster unprepared.

RobertJasiek: Re: Proposal (2011-12-13 20:11) [#9045]

Actually komaster and komonster are useful ideas but overloaded with too specialised details. What one wishes to teach is something like "currently having enough ko threat material for winning this particular ko", but there does not seem to be a term for (only) that, or is there some? Re: Proposal (2011-12-13 20:45) [#9048]

No, there isn't (yet). That player shall be henceforth be called "the koko". Excuse me while I create a page about it (maybe).

RobertJasiek: Re: Proposal (2011-12-13 20:49) [#9049]


tapir: Re: Proposal (2011-12-13 20:51) [#9050]

Isn't the amount of compensation the opponent gains crucial for any evaluation? Usually you don't have a komaster, but a compensation of 2 tenuki plays is all you can calculate with without accounting for the ko threat situation in detail.

RobertJasiek: Re: Proposal (2011-12-13 20:52) [#9052]

Let's research in that, but... please not just here in this thread:(

reply A sufficient definition for teaching purposes (2011-12-14 09:09) [#9057]

For the purposes of teaching the concept to beginners, the page "User created go slang" already offers an adequate definition with enough detail and a more proportionate amount of effort expended.

  1. Belgian ko: A ko in a situation that will not have the desired result even if won.
Dieter: Re: A sufficient definition for teaching purposes (2011-12-14 09:44) [#9058]

All right, all right! Hold it right there!


emeraldemon: discussed in intro to life and death (2011-12-14 20:37) [#9078]

I tried to discuss this at a beginner's level in IntroExercise3 and and that page's answer. I thought at that level it was unusual enough to warrant an example and some explanation. If a beginner actually looks through the Intro Exercises (which I hope they do), perhaps that is enough?

RobertJasiek: Re: discussed in intro to life and death (2011-12-14 22:59) [#9082]

That example is not a dead-ko but a 1-eye-flaw (definition see my paper). Do not confuse these two!

RobertJasiek: Re: discussed in intro to life and death (2011-12-15 02:43) [#9084]

Correction: Under superko, it is a 1-eye-flaw only if currently Black may not make the ko capture and does not have any reasonable (even if trivially filling a third single intersection eye) tenuki; otherwise it is a dead ko. Under other ko rules it can be different.

RobertJasiek: dead ko or dead-ko (2011-12-17 11:37) [#9125]

A "dead-ko" spelling would make sense only if every ko type or at least every ko type of the life+death ko typology with naive ambiguity of the kind "modifier" + "ko" <> "modifier-ko" were also written with hyphen. In particular this concerns "fighting ko", "disturbing ko", "disturbing life" and "disturbing death". Supposing an agreement would be reached to write all of them and possibly some more as necessary with hyphens, how would that be enforced and which scope of application can be expected for the hyphen spelling? To start with, would all librarians wish to correct all occurrences on SL now and in future and whenever some innocent user confuses usage as a ko term with informal, non-ko-term usage? Could we hope to spead such a spelling consistently elsewhere in the English speaking internet? Would we be prepared to explain that prior spelling of those terms has been different whenever somebody asks?

X Re: dead ko or dead-ko (2011-12-18 11:25) [#9130]

The hyphen is not needed for most of the terms.

When writing, or communicating in general, you always have a target audience. Whichever conventions you choose to use will not matter in the least, as far as your target audience will understand what you wanted to say. So, if you want to approach the go player population of the world, choosing the Lingua Franca "English", vocabulary "commonly used go terminology" and the voice "speaking among peers" will be a fine choice. For teaching your neighbor's 6-year-old to play go, you would likely do well by choosing "German", "very little terminology" and "teacher". Again, for writing a research article on some go rule set, you need to reach scientifically interested go rules experts of the world, so the good choices would be "English", "special terminology rigidly defined in reference work X", and "scholar addressing other scholars".

The problem described on this page arises because two of the cases (the first and the third ones) happen to have conflicting meanings for a "dead ko".

Therefore, when writing the sciencey stuff on a wiki with a couple of rules experts and lots of other go folk, a warning light of some kind is needed to highlight the use of a common word in a very unusual sense. (In a scientific article, where you have defined "dead ko" as a term, you would need a similar warning light if you wanted to discuss "a ko that is dead".)

The most commonly used technique to implement such warning lights is typesetting. Small caps (hard to use on SL) or italics (dead ko) are the more common approaches. Starting the words with capital letters (Dead Ko) is also possible, but will look a bit weird. (In German orthography, the capital letter approach is, of course, unlikely to succeed at all.) On the web, the standard way of linking the terms to their definitions (dead ko) is a much more useful one.

So, for the purpose of chatting about the interesting aspects of go, no typographic trickery is needed, because scientifically rigid definitions are typically not necessary, and showing the situation on a board will nearly always adequately convey the meaning. For the purposes of scientific articles, it is an excellent idea to mark all newly defined terms in a way that instantly sends the reader looking for the glossary. For this purpose you have used the hyphen approach quite successfully so far, and it does not make the text look bad, so that is what I recommended for this case too. It would also remove the overlap of dead ko and dead ko, and that overlap is why there is such a strong opposition against the term; accepting it would create an unnecessary ambiguity in the meaning of "dead".

tapir: Re: dead ko or dead-ko (2011-12-18 12:15) [#9131]

Sensei's Library does not distinguish between "with-hyphen" and "without hyphen" spelling in page names, why should it elsewhere?

Also, while as a librarian I am willing to do library work, I am a bit puzzled by the "I don't care enough, but can the librarians please clean up" attitude portrayed here.

RobertJasiek: Re: dead ko or dead-ko (2011-12-18 12:32) [#9132]

I don't work the work, you don't, so what about the user

tapir: Re: dead ko or dead-ko (2011-12-19 11:59) [#9133]

Regardless of what you do on pages, SL does not support such a distinction in page names. E.g. Dead ko and dead-ko. So, I believe it is pretty pointless to try to make such a distinction systematically.

Back to forum     Back to page

New reply

[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library