Meta Discussion

The ko use case [#2681]

Back to forum

New reply

Dieter: The ko use case (2011-12-05 12:00) [#8908]

In the light of our recent discussions about the future use of SL and the forum at L19, I departed on a story: I am a relative beginner and want to learn more about ko.

The result is not very nice. Our pages on ko do not help, on the contrary. The basic pages of ko and ko threat are rather good, but then the ko fighting? path and the second course on ko? path fire off in many different directions, mentioning ko master where it is often not helpful.

I believe the subject of ko is one of those where critics rightfully claim that learning to play go from SL would be a disaster.

Let's see what we can do about that.

tapir: ((no subject)) (2011-12-05 14:30) [#8909]

I was such a case (started writing as a beginner) and while I found SL pretty confusing on ko. I don't know which other material is so much more structured. There is only one really good English book on the topic ("All about Ko"), too. Even Guo Juan's online lessons consciously stay away from systematically addressing ko, they discuss it in game reviews, but the only lecture dedicated to ko was made on my request.

What would help is clearly separating novice and advanced material, maybe even with different indices, and then to stick to the distinction - and make it transparent to the reader at every step. Ko master / ko monster concepts should be marked as advanced material. As obviously the term ko master and ko monster are more often misunderstood than understood when not used by Bill Spight. The whole handling of ko situations by referring to the ko master concept is especially misleading when it leads to the omission of the crucial concept of compensation (a "ko loser" (term) doesn't gain any compensation but an average value move, while any real player losing a ko gets compensation 99% of the time - so we have ko apprentices not masters most of the time). Another complication are the additions of Robert, that are not well integrated with the rest of the ko pages. The biggest disaster are the pages necessary because of special ko rules.

But in the end ko is a topic, where getting the basics right, isn't trivial at all, if we seriously start with examples where threats have to be evaluated in size.

RobertJasiek: ((no subject)) (2011-12-05 15:38) [#8910]

Literature on ko is still weak incl. the relatively best English book on the topic. I understand that you can't wait some more years until I will fill that gap.

SL has ko pages with topics for beginners, advanced players or experts. All serve their purpose. The overall SL ko coverage also has lots of gaps though. Nevertheless, one can try to make the best out of the current situation. I do not know if all ko-related pages carry proper tags "ko" and "beginner" / "advanced" / "expert". Regardless of the tags,...

It would be possible to create reference / topic path pages to all those ko pages that are for either beginners, kyus, dans, experts. So something like about four ko path reference pages for each of these levels.

To do that, somebody needs to list ALL ko-related pages. Only then we can classify or reference them properly.

Dieter: ((no subject)) (2011-12-05 16:41) [#8911]

Robert, one can only love that first sentence.

The material on ko is pretty exhaustive on SL but it is poorly organized. I believe that is one of the main complaints about SL in general, so if we want to improve SL as a whole, improving ko can be a showcase (if only for ourselves, to see if we can do it at all).

I can do the listing and start the exercise, but we must be sure about our intentions and from the past I seem to remember we have different ideas. To restate mine: an introductory page or series must make things clear for a beginner in an informal matter but nevertheless not compromise too much on factual correctness. It should not attempt to be 100% correct and complete at the expense of a steep learning curve. Anything resembling formal scientific language that is otherwise absent in the world of go (I think of n-territory) is to be avoided IMO, even if the introduction of such concepts would benefit overall comprehension in the long run.

The same goes for CGT-inheritance like ko master or monster. All these terms have their reason for existence and must be explored, but again be avoided at intro pages. That way we can sharpen a beginner's intuitive comprehension of ko and we can collectively grow an understanding of the more refined theory on the advanced pages until those crystallize in our brains and culture, so that they can make it to introductory pages by merit and not by force.

If we can agree to that, I'm all for a collective effort, captained by your experience and keen theoretical prowess.

X Re: ((no subject)) (2011-12-05 17:00) [#8912]

I actually think that the goals of factual correctness and beginner friendliness do not conflict so much. It merely requires care in choosing examples, so that complications will not arise, and informal explanations will map to higher level ones. But perhaps my faith will be proven wrong.

Robert's index may be a good idea here. Because of the quantity of material, and because it is not necessarily clear which pages are meant for beginners, some coordination is required, even if there are places we can just jump in and make improvements. --Hyperpape

tapir: Re: ((no subject)) (2011-12-05 17:10) [#8914]

I started a CGT about ko? as an index for CGT terms concerning ko. I started listing some pages there and tagged all of them as expert level (+ theory and ko keywords) for now. As far as I understand most terms Robert added are listed under Mathematical Ko Terms but they are not tagged with the ko keyword yet. Maybe we can join both indices :).

Another candidate index can be sth. like rule dependent ko terms? which can temporarily clear the field of all those disturbing kos :)

I don't know whether you should waste time with making a list of all pages now. Advanced Find Page is pretty comprehensive already (search for ko keyword) and without maintenance. If we find more ko related pages we can just add the keyword and everyone of us can find them in the next search. I don't object to someone doing it manually, but it might be too tedious and unnecessary.

My aim maps with Dieters for the introductory pages. For the advanced pages we probably do best by organizing access to the pages by means of well-ordered paths etc. without trying too much synthesis which inevitably would produce disagreement. Re: ((no subject)) (2011-12-05 17:19) [#8915]

That's reasonable. Perhaps a list is not necessary, more that we need to agree on what is for beginners. Fighting Ko path? The intro to the main page? What else? --Hyperpape

tapir: Re: ((no subject)) (2011-12-05 17:29) [#8916]

I believe beginner material should link to other beginner and intermediate material, but not all the way to the expert stuff. While this might look like a heavy-handed pedagogy, I believe much of the confusion results from jumping by one click from the introductory to komonster. That is I would be happy if we could have a full (though incomplete) cycle of ko pages (what is it, how to play it, examples) which would map to beginner / intermediate levels - pages on ko aren't introductory anyway (as a page like atari is). resolve the ko is a good example for what I imagine.

emeraldemon: ((no subject)) (2011-12-05 18:06) [#8917]

I like the idea of a full board ko example, maybe something like Basics on Kos/Example? . But that particular example seems a bit complex to me. Maybe start with some very simple full-board ko positions?

Also, the most common ko for beginners to see is the poorly named Half Point Ko. That could also use a bit more information, perhaps.

reply ((no subject)) (2011-12-05 18:20) [#8918]

When a CGT ko terms page is created, then another page non-CGT maths ko terms is needed. Therefore I dislike the idea of starting with a CGT page. Let me repeat my wish for beginner (or call it introduction) / kyu (or call it intermediate) / dan (or call it advanced) / expert reference pages. Surely an expert page would list all maths terms; it is too early though to split expert into CGT, non-CGT, rules-heavy etc.; better keep all on one expert reference page; if that grows too large, then we can still decide to split later.

All terms can be derived from rules. Therefore it is not a good idea to create a rules dependent ko terms page. Let us start to put on an expert reference page what is not easier.

Back to forum

New reply

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