[Welcome to Sensei's Library!]

StartingPoints
ReferenceSection
About


Referenced by
MetaDiscussion
WikiOrphans

 

Meta Discussion 2003
   

18 December 2003

Cheyenne: This is kind of request for an anchor/definition/study page. I know that there are basically two major styles of play; moyo based versus territory based (are there others?). What I would like to see is a page that would offer guidance on study based on these different styles. Which fuseki's to study, which pro players to study, pointers to example games, etc.

Cheyenne: Added the Territory Versus Influence Styles page and Territory Versus Influence Styles Discussion page.

mAsterdam: Hi Cheyenne. I would very much appreciate educational remarks on playing style by some more advanced deshi, too. When I put the page up initially that was what I hoped it would trigger. It did, but unfotunately it triggered just a few of them.

12 December 2003

BillStrider :

As long as I'm here, is there a convention on where new material should be added to a page? It seems as though text is added at the top of discussion pages, and at the bottom of topic pages. Is that an accurate observation?

Charles Not really. It is better in most cases to add to the bottom of discussion pages - interpolations can cause confusion. Some pages like this are dated, with new dates added at the top for convenience (also true of OngoingGame pages in general). Anyone can add anywhere, naturally. It may be better to add queries as a footnote.

Bill: Discussions tend to get messy, leading to the eventual necessity for an edit. Since it is now easy to return from footnotes (by clicking the bracketed number), I think that the liberal use of footnotes is a good idea for adding some structure.

28 November 2003

Dieter: On both joseki pages and fuseki pages, high quality pages are mingled with low quality pages and collaborative pages are mixed up with personal projects. In order to discern between the two, a Whisky Master editor could rely on page ratings. Arno, is there a possibility to have pages rated by users ? Or isn't that desirable at all ?

2003 November 26

DJ: I've done a small edit to BQM129, but it doesn't appear in RecentChanges, only in FullRecentChanges. I do not recall to have ticked the minor edit bow, though...

wms: I have a question. Why is it that an apostrophe in a page title turns into a capital letter in the URL? It's very disturbing to programmers who expect a capital letter to denote a new word! See KGS Beginner's Room (which becomes KGSBeginnerSRoom) for example, but there are several more like it. It seems just stripping the apostrophe would be a better choice.

Charles I've been offline for a few days. I had some concrete thoughts, to move on the recent discussion.

  • Can we look at the Page Type field, with a view to making more types? I think, at the least, that recognising a Discussion Page type, and formulating some conventions for it, should help.
  • I propose the creation of Pages in need of attention?, and Requested pages?. There might be a case for further pages, under a Maintenance Page heading: but these seem potentially useful to me.

Stefan: I'm too uninformed about the entire discussion thing to have an opinion on your first point, but I do like your second suggestion, Charles.

2003 November 21

BobMcGuigan: I sympathise with Bill and Charles, both. Often I have looked at Recent Changes and shrugged my shoulders on finding many (even most) entries of at most marginal relevance to Go. I am impressed with how wide-ranging discussions lead to interesting and useful summaries. Where some structure might be helpful is in some clear way to separate the peripheral discussion from the serious summary and teaching material.

2003 November 10

Anonymous: Is it possible/desirable for the text of recent edits to be in a colour other than the normal text colour? I like to read 'Recent Changes' but am lately at a loss as to whats changed in the pages it refers to.

Arno: us junkies only read the diffs ...


2003 October 26

(Sebastian:) Should we have an FAQ page for SL? Here are some (including a comment) that I found in the sandbox:

Charles Move to Quick Questions?

(Sebastian:) I wasn't aware of that page and yes, one of the questions was already there. Is that page also a good place for questions regarding SL itself?

Charles I guess it hasn't been much used recently - by all means bring it back into prominence, for example by linking to it.


Question: Is it possible for a twisted mind to erase everything ever written on these pages?

Answer: There are backups.

Answer 2: SL has been around for over three years. In that time a fair amount of minor vandalism has occurred, but the strength of the community keeps things running rather smoothly indeed.

Many wikis have been around for much longer than that, even; and the oldest ones are probably the best place to look for discussion on this sort of topic. Therefore I refer you to

[ext] http://c2.com/cgi/wiki?WhyWikiWorks .


I'm impressed with this living library!!! good idea fellas, I'd like to see more of this dynamic web stuff on the internet.... ahh what a Weird Wired World -

funkyBside


Q: How to write B.C. without the periods? BC BC !BC !!BC
A: Precede with an exclamation mark!


2003 October 24

Charles If I can make a brief general comment, on the launch of Talk pages: we can have 'encyclopedia'-style content here, we can have 'magazine'-style content too (quite suitable for the pages of the BGJ - RIP the AGJ), without detriment to either. Or to any of the other kinds of content that gets posted here: reference material, joseki, mathematics, links and so on. I'd just like to sharpen the demarcation a little at this point, because I see that as the way forward.

Bill: While I am in agreement with having Discussion or Talk pages, I am not so happy with the current implementation. For one thing, I think that Talk tends to trivialize the contents of the page. We already have Coffee Machine and Library Lobby for talk. For another thing, it seems kookie to me to have Agehama Talk and Omoshiroi Talk pages without Agehama? and Omoshiroi pages. In particular as regards omoshiroi, it is not clear whether that should be an alias for Interesting or not. We need some feedback from Sebastian or others to determine that. Some editing will need to be done, and having a separate Omoshiroi Talk page will complicate the editing. This also ties into the question of whether a discussion is about linguistics or go. The answer is not so pat. A good deal of learning go has to do with learning the vocabulary.

Charles Well, I don't see why Talk about something needn't have the same status as Discussion. I'm aware that the demarcation I'm suggesting isn't water-tight, but it is surely better to shuffle material about a trio of pages X, XDiscussion and XTalk than to scrabble around in different parts of the site.

Can we see how this beds down, before conclusions are drawn on the kookiness? The reaction that the vocab and learning go are close together is what I expect from the Japanese-speakers here. But in a sense they are excluded from the normal experience, which must go more along the lines of 'how much of this stuff do I need to know, right now?'.

Bill: The point about terms is not the words per se but the concepts. I think that it is important to learn about leaning attacks, but when I talk about motare on SL, I am almost certainly talking about go, not language. On other, not so straightforward terms, questions of translation (language) may blend with questions of concept (go). If I wrote about go and later found that what I had written has been moved to another page and labeled as being about language, I might feel that my writing has been misrepresented. It is one thing to express a differing viewpoint on certain material, another thing to impose that viewpoint.

I think that a wait and see attitude is fine. Why don't we wait and see if the people engaged in such discussions create Talk pages for them themselves? If it is a useful structure, people will use it of their own accord, won't they?

2003 October 22

Charles A rec.games.go posting has just appeared, suggesting that the content of the web site at

 [ext] http://myhome.naver.com/dajava/lifedeath/problem.html,

consisting of classical life-and-death problems, be somehow transferred to SL, before that page goes offline shortly.

I have no reason to doubt that the posting was from the site's owner (though there is every reason to take some care over this point).

I think the copyright situation, for what is a superficially attractive offer, is treacherous.

For example, the Igo Hatsuyoron problems there are (said to be) from the Yutopian edition: which because of the editing and removal of some deliberate 'false trails' in the original is quite probably not public domain.

I was never very happy with the Kanazawa problem series posted here, despite the strong defence made of the use made of the underlying classical collection. Anyone wishing to take up on this offer to copy material across to SL ought, in my view, to do some serious homework before posting it here.

2003 Oct 9

SnotNose: Am I the only one who finds it extremely hard to follow a game with diagrams with numbered stones without the use of some medium by which I can play the moves one at a time?

When I read a game commentary at home I always play it out on a board. I also review games (with and without commentary) by SGF on my computer, where I can see stones placed one at a time. But I almost never get very far through commented games on SL because I can't handle a board full of stones with more than 3 new moves (numbered). I just lose patience.

The reason is that I need to see the new move in the context of the whole board to begin to understand it. But, with all those other numbered subsequent moves sitting there, it is hard for me to see the board clearly. This is interesting because I have no trouble mentally adding stones to a board--like when reading out sequences--but I can't seem to mentally subtract them--as with those subsequent numbered moves.

Is this a problem for anyone else? Would it make more sense to think about an enhancement to SL to include a game viewer so commented games can be viewed one move at a time (with appropriate comments appearing at the right time in some other area of the viewer)?

Just a thought.

Charles Gobase does this very well. I feel there is an argument for keeping SL to a different approach: namely not to encourage the posting of diagrams with long sequences. (There a regular requests for longer runs of numbers; I have had regular comments from kyu players which are negative about diagrams with too many numbered stones.) I realise that there isn't currently a site to which one can contribute commented SGF which are then editable by anyone.


2003 Oct 8

wms: I'd like to see, in addition to the overall "most popular" list, per page hit counts in the "page info & history" link. Why? Because I'd be interested to see if anybody reads the pages that I update. In other words, is it worth my time to keep going or not. Having a "most peopular in the month so far" would be fun too, that way pages could compete for top notice without having to overcome years of lead time by the older pages!

Hu: If hit counts are introduced, then some accounting should be made for viewing of diffs, for example through Recent Changes. That is the way I most often read additions to SL (especially the KGS Wish List page), and I am sure that a lot of traffic goes through diffs reported in Recent Changes and its sister pages.

Arno: I have been thinking about having a monthly MostPopular list as well. Makes more sense. About the diff count: we had some 280000 page hits in September and only 20000 diffs. So even if the diffs are omitted it would not skew the counter too badly. Reading diffs is mostly done by junkies.


2003 Sept 22

AvatarDJFlux: Ever since the very first time I came to SL I've always thought that the name of our game is not prominent enough in the Front Page. Adding the picture was a step forward but not enough, IMVVHO.
I would like to propose all the fellow Deshi's the very simple modification below - what does everybody think?

Goban (image by Juha Nieminen) Sensei's Library is a collaborative web site about and around the game of

Go
(Weiqi/Weich'i
Baduk/Paduk)

It is a repository, a discussion medium, it is the pot of Go(ld) at the end of the rainbow! SL is whatever you make of it
Sebastian: Good idea. I'd leave out the two alternative spellings, though. They're listed on the respective pages already.


2003 Sept 21

Fhayashi: On FM1, FM2, I love the contents, but the page titles break the usefulness of the page title search, doesn't it? Wouldn't it be better as simply FavoriteMoments1? and FavoriteMoments2??

gimpf: Would indeed keep things more simple. When I first just saw FM in the Recent Changes, I thought of frequency-modulated broadcasting...

BlueWyvern: I did 'em like that to echo BQM. Maybe they should be changed. I dunno.

Charles How about FPM - Favourite Pro Moments?

Fhayashi: I think BQM is also breaking the title search... but it would be a pain to fix. BW, it's a chance to start your interesting series of wikipages in a more user-friendly fashion. Let's go with FavoriteMoments1?, please? (hmm, calling BlueWyvern 'BW', makes him into BC...)

2003 Sept 19

Sebastian: The pages regarding Go programs need a major revision. See Go programs for a proposal how to do it.

2003 Sept 12

I just added a Compost Heap. It's just an idea - let me know what you think of the concept. -- Sebastian

2003 August 26

HolIgor: I put the reference to a problem at goproblems.com as a problem in snapback workshop. I think that we shall encourage such references to other net resources as we are one great community, aren't we?

2003 August 16

MarkD: Moved from the frontpage to clean it up:

Temporary notice: 08/15/03 03:20 pm EST: Restored Frontpage after vandal attack from 80.14.207.127. MarkD

Temporary notice 2: Who is 12.228.141.176? He is deleting a number of pages and leaving a number of destructive comments, even while I am reviving them. --unkx80

Just a little comment: Why, oh why would these people want to destroy the pages? Could they not handle the game of Go and pointed their anger and wrath to this site? Just a thought Varinz

Plus i added a site where players can write about themselves, and their thoughts about Go. Player list?

2003 August 1

Velobici: A question regarding attribution here at SL. When adding to a discussion I attribute my comments. When adding a few words of explanation or a link to another page, I dont attribute the change. Attributing the change would ruin the flow of the document, making it hard to read. Is this honte? Or should all changes be attributed till a master edit of the page? Two recent examples of changes without attribution are the endgame page points only and the 34PointHighApproachInsideContact W2 move to reach nadare.

Bill: Moi, I occasionally make changes or contributions without attribution here on SL. They are almost all minor edits. :-)

As for the examples you give, Velobici, on the joseki page you clarified the text, while on the endgame page you offered your opinion. I think that's why Dieter suggested that you sign it.

Dieter: On the endgame page, the first edits were also intended as clarifying, so I don't object to them being unsigned. However, after Bill posted his objections (signed), Velobici went into the discussion unsigned and I added his name to his statements myself. So I am only talking about stuff which already is discussion, not about stuff which has the potential of turning into discussion. Once it does, someone else will have to WME it anyway.

Basically, sign as soon as there is apparent disagreement or confusion.

Hu: I often make contributions without attribution. Those who are really interested can always follow the page history. I even will start new pages without seeking credit. On the other hand, if I am making a statement that might be controversial or adding to an on-going discussion, or making a major contribution to an established page, or writing what is clearly opinion, then I will sign it.


unkx80: What do you deshis call the kind of box-like framework? (Of course, BC may be at a or b).

[Diagram]
Box formation



In BQM 102, I wrote that a similar kind of framework as a "box formation". But on closer inspection of the box formation page itself, the "box formation" is defined as occupying four star points all in one quadrant.

Bill: I call it a box, too. I think that any rectangular or almost rectangular enclosure can be called a box or box formation.

unkx80: I have moved the current contents of box formation to four star points box formation.

2003 July 30

mgoetze: I think the material currently at New Fuseki Era ought to be what you get when you look for Shin Fuseki, and that the first of John's posts ought to be on another page linked to from there. Is there any good reason to keep the present arrangement?

Also, is it just me or has only a third of the work been done at Taisha Five Way Junction?

2003 July 29

Charles It seems to me that BQM105 is getting into a tangled if interesting state. I wouldn't want our poster Sazn from Korea to feel any less than very welcome here. I would also like to see something more like a joseki page edited out of this.

Fhayashi Anyone know of any English commentary on the games of Rin Kaiho? There seems to be nothing in English written about him or by him.

Bill: He is the author of Golden Opportunities. :-)

2003 July 26

Charles Now that the technology to make pages like 3-4 point enclosures is available, I'm seeing more of them. Can I comment, though, that I don't like what was done just now to the 3-4 point page? There was a list of links to pages dealing with individual enclosures such as the 3453 enclosure. It has been replaced by just one link to 3-4 point enclosures. The 3-4 point page is a beginner page. I have never met any beginner who could handle the 'raw data' of all the kinds of enclosure listed, without comment. Enthusiasm for linking to all the information on a topic that we have on SL seems here to be running ahead of sensible pedagogy. These big link-from diagrams are good for reference material, but far too dense otherwise.

Blake: Actually, I'm not sure that the individual links to different enclosure pages were any more transparent to a beginner. I certainly find a diagram more intuitive than the 3-4 6-3 type of notation. Though if you think it's better, feel free to restore the links :)

Charles One point is that the single-diagram approach lumps together very unusual techniques and the most common: it's not selective.

Blake: I suppose that's a good point. A beginner (not that I'm a strong player, mind you) wouldn't be able to distinguish between the 'secure and standard' and the 'really bizarre.' I will restore the links.

Bill: Having seen this exchange, I took a look at the page and I am confused. There are links to specific enclosures there, but they are the weird ones. Maybe Blake hasn't restored the previous links yet?

Charles Would anyway be better in the small low enclosure style on that page. The general point would be to do with correlating difficulty levels, styles of indexing, and access to 'reference areas'. What happens for joseki here seems a reasonable model.

Bill: Well, if we are taking the idea seriously that the 3-4 point page is a beginner page, rather than a reference page (We still haven't revised our difficulty indicators), what the hell are links to the 3-4 7-3 enclosure and 3-4 3-7 enclosure pages doing there? (Among others.) They are certainly not common, as is claimed.

It's one thing to encourage beginners not to be too bookish, but to use and develop their own judgement, quite another to suggest plays to them that they will find difficult to handle.

Charles Yes, those links don't belong there.

mAsterdam: As Bill said: "We still haven't revised our difficulty indicators". The discussion seems to have stopped a few weeks ago. What is needed to get this ball rolling? (Dutch saying, don't know if it means the same in English).

Blake: Those are the links which were there before, and I actually made them more legible; originally they were in the notation like '[3437 Enclosure,' without separation of any sort.

Also of note: small stone (B1,W2,B3,...) notation doesn't work in diagram titles. See J groups. Can this be fixed simply, or is it a problem inherent in the diagram-rendering routine?

mgoetze: Well, while I can understand Charles' criticism of uncommented information on a page marked beginner, I found the way it was left to be unpleasing; enclosures were mentioned twice one the 3-4 Point page and you still had to jump around to get to the actualy information; too much redundancy for my taste. I have added the links to the main diagram about enclosures on the 3-4 Point page with additional comments about what is usual and what isn't, and hope this will be acceptable.

And while we're on the topic... I think many intermediate-level players use SL as a reference, and if they want information on the 3-4 Point, they're going to look on that page, regardless of whether or not it's marked beginner...

2003 July 13

Velobici: Lack of dates in reporting events. An example of this can be seen in the first comment on the KoreanDominanceDiscussion page. The first comment is that Japan has recently won a tournament. When? When the entry was first added, the question answered itself. Now that some time has passed, its not clear whether this happened this year, last year or earlier. Furthermore, we dont know which tournament. Please add basic information. Who? What? When? Where? and How? when at all possible.

MarkD: Very good point Velobici. Same thing is true for expressions like "last July", "last month" etc.


MarkD: I am not sure if this topic has been raised before, but I've seen many japanese words with an "s" at the end for plural form. (example "4-5 point Josekis"). Because there is no plural form for Japanese nouns (one Joseki, two Joseki,....) this looks a little strange.
Any suggestions for future edits?

mAsterdam: Fuseki appears to suffer from similar treatment. Any more victims?

Andrew Grant: I'd prefer to spell these words without the "s", but I've seen Japanese terms pluralised with an "s" many times in a wide variety of go books, not just on SL. There doesn't seem to be any accepted convention on this, unless we wish to establish one here.

2003 July 8-11

mgoetze: I'm rather new here (in fact I'm not even sure this is the right place to ask this...) so I thought I'd solicit some feedback before jumping in and changing anything. But doesn't it seem that the Jubango page is rather biased towards Go Seigen? I mean, I love Wu as much as the next guy, but I would be more interested in finding some information of a broader scope on the Jubango page than the match record of one player, which I can also get by following the link to his page. And aren't there any other players who had several Jubangos? If so, how about posting their records on the pages of those players?

Stefan: Hi Michael, and welcome to SL. Since I wrote the first versions of Jubango, I feel I can answer this question. Your criticism is justified: when I started the page it was not 'rather', but _entirely_ biased towards Wu. The reason is simple: that was the only Jubango-related information I knew of. Usually other people here quickly add whatever they know or research when you post incomplete stuff here. Sometimes it takes a bit longer though... Anyway, I encourage you to go ahead and edit the page, and make it a better one.

mgoetze: Well, a full-text search for jubango does reveal a few specific matches not involving Wu. But I'm afraid I would not have anything else to add in terms of content; I would only change the form. So I think I'll wait a bit and see if anyone else does a better job of it than I possibly could. (Hint, hint!)

mgoetze: Great job, AvatarDJFlux! I wonder if someone could provide similar short descriptions of some of Go Seigen's Jubangos? Then maybe someone else could contribute info on another modern Jubango, and afterwards the page would be ripe for a Wiki Master Edit.

2003 July 7

Dieter: I just got over with my own old habit of making pages called old... after a WikiMasterEdit. However, I'd suggest that in the WME statement we mention the number of the version preceding the WME.

2003 June 1

mAsterdam: The list on please review me only seems to grow. Maybe the pages on the list could be marked as "to be reviewed" in a more visible way (uneasy line color / warning sign - whatever). This way there would be (I hope) an incentive to get those pages off of the list. An easy way for the senior authors to let the less experienced deshi know that their work is 'good enough' to not urgently need a review anymore would help too, IMHO. (For instance: I would like to remove Priority from the list, but I simply don't know wether it is good enough to be removed from the list).

2003 June 27

Deebster: A petty, petty thing, but easy: "This page does not yet exist" pages shouldn't say "x last edited by on January 1, 1970 - 01:00"
This is probably why I never finish anything, too obsessed by trivialities like that one...

Andrew Walkingshaw: This is a Unixism - that's the first day of the Epoch (Unices count time as the number of microseconds from 00:00UTC, 1/1/1970. Thus, any uncreated file has that creation time, in some sense...

Deebster: Yes, I know it's the Epoch (i.e. date('r', 0); in php), I was just quibbling about the fact that the 'last edited' message shouldn't be on something that's never been created, therefore never edited. As I said, very unimportant.

mAsterdam: Indeed. So we'll get at least 20 edits about it.

2003 June 12

Frs: I suggest to create a page like How to play / record (experimental) games played at SL.
The seed for it is in the first section of 13 x 9 Game 01.

2003 June 8

Moved to Page difficulty discussion -- mAsterdam

2003 June 7

Charles A thought as I look back through edits of method of playing inside systematically. I adopt in writing here the convention standard in English prose, of writing words rather than numerals for numbers up to ten (inclusive). This would be typically be required by copy editors.

I believe this makes for readability. We have no need for move numbers in the text since we can do B1 and W2. We sometimes need numbers other than integers, and they of course should be written numerically. I feel happier writing 'five' rather than '5' when I can; and would ask others at least to respect this in what I write.

By the way, the whole 'liberty' issue was first aired there, as far as I can see; and it really was an imposition on my page marked Beginner. I'm not sure we are done with the issue, but can I flag here that this was never to my taste? A clean-up is urgently required. If we can ring-fence the ambiguity point, to the extent that (a) it is not mentioned on Beginner/Introductory pages and (b) [liberty] on such pages is really [liberty|liberty - introductory], then I think we might be able to make some progress.

Bill: I am happy not to talk about the number of liberties in introductory pages. I am not happy to give a false impression to beginners when you count above atari plus one. I think that the problem can be addressed without going into the difficulties by linking. That is, any page that talks to beginners about counting liberties need not present anything complicated, but should let them know that there is more to it and link to the Liberty page. When they are ready they can forge ahead. I think that this is a good approach in general when simplifying things for beginners. Don't confuse them, but let them know that there is more, and give them a link or links to the appropriate page or pages. This is a good way to use hypertext, I think. :-)

Charles Bill, we certainly know your views; and we know that they are strongly held. What I don't know is if they are shared by others here. Perhaps they are. But I really don't see, in the many postings here, anyone else who firmly believes that a true beginner - a genuine novice - really needs an ambiguous notion of liberty. We do know that Richard Hunter, writing a book-length treatment, makes an explicit switch of meaning; as is possible in a book.

I really feel we may have to get to votes on this. It would be a shame if Bill felt that this was in any way a reflection on the value others place on his prolific contributions here.

Bill: Charles, I do not believe that a true beginner - a genuine novice - really needs an ambiguous notion of liberty. I am sorry if I gave that impression. And I do not think that I am saying that by advising a link to the liberty page if you really get into the question of how many moves it takes to capture a stone or chain. That way you avoid confusion.

You want a vote for making things simple for complete novices? You have mine. You want a vote for not giving the impression when doing so that that is all there is to it? I vote yea to that as well. You want a vote for not linking to appropriate reference pages? I vote nay to that.

Charles Well, I'm going to take it upon myself to try to resolve this discussion. There are 13 Beginner pages now linking to liberty. I'm going to make a liberty - introductory page as a kind of preface to liberty, make it clear there that it's only the beginning of the story on filling liberties to capture, and divert all those links there. I shall make it 'work in progress' on that page for the moment. (Later - this has all been carried out, though the liberty - introductory page is still a draft. It occurs to me that the word 'tutorial' should be used more often on SL.)

Bill: Cher Charles, I took a quick look at Liberty - introductory, and it looks very nice so far. Thanks. :-)

2003 June 6

Moved to Page difficulty discussion -- mAsterdam

2003 June 4

Charles It occurs to me that the development of SL, and a much-needed 'page churn' bringing older pages back to current attention, is inevitably raising the question of authority. This is obvious in the cases where material is here because it has been written about in books by Ishida, Kageyama, even Hunter. It also comes in to all discussions based on Japanese usage. It is quite unavoidable when we talk about joseki, because the whole point there is that standard sequences have their status because they are used by strong players and then accepted on a communal basis.

My own 'solutions' here:

  • second-guess any assertion about joseki using a database;
  • try to state exceptions to proverbs and other conventional wisdom when I can find examples;
  • but not to assume that this refutes, just broadens discussion;
  • argue against what other authors say (whatever their level relative to mine).

From a wiki standpoint I don't think that authority as such exists. It's more like the old scholastic habit of glossing everything, to the point where the glosses overtake the initial material. Which is not so bad for SL: the older books in English do date back several decades.

2003 June 1

Bill: Great pic on Front Page! :-)

Deebster: Agreed. In WikiNews, Arno says we have Juha Nieminen to thank. There's some more of his stuff at [ext] http://www.students.tut.fi/~warp/pics/ (I feel no guilt at the extra bandwidth I'm causing as I don't think it's paid for by him :) )

2003 May 28

Charles Well, the miai values list is well launched, with 50 different positions there now. Obviously it will shortly need subdividing. Clearly, too, this isn't an ideal reference for learning the endgame basics. But it seems to work quite well, for the smaller endgame plays: as well as discursive treatments, I think.

HolIgor: I think that the most useful application of the list in the reduction of a little bit complex position to the position with known move value. But in order to do this efficiently one needs to index the positions somehow. I propose to reference the diagram by the move value and an index of a diagram for that value of the move.

Then the explanation would become like this.

0.88

[Diagram]
Take away point and reduces to 0.75(Diag 1) position

See corridor.

The rule would be to add new positions with the same value at the end.

Charles It was certainly my hope that such a list would clarify how to compute miai values - and so give an operational definition of what they are. The question HolIgor raises is the natural next step. It's really a choice of method: do we assume that in a well-developed list the smaller plays on which calculations depend have already been done? That's a kind of 'dynamic programming' approach and justification. Or on the other hand should there be some kind of pointers showing the way?

By the way, I think it is inevitable that some unreliable values will be posted; but that shouldn't be too much of a problem in the longer term. For someone reasonably expert, checking the value takes about the same length of time as creating the diagrams.

2003 May 25

Charles Matthews Thinking about the Endgame Reckoner discussion, I wondered where we were on reference material on SL.

There are these categories:

  1. People keyword, Names in go: now good coverage of pro names, but info not so deep.
  2. Joseki: more and more pages are being added, structure in place.
  3. Side patterns: my particular concern - a drop in the ocean.
  4. Openings: shallow, considering reference value.
  5. Middlegame joseki: just starting out, really.
  6. Tournament keyword: patchy coverage, indexing.
  7. Go Terms: expanding all the time, needs taking in hand.
  8. Life and death: reasonable coverage of shapes all should know.
  9. Endgame: no structure for reference material.

On the last of these, one can envisage, one day, a list of standard endgame positions, arranged by miai value. Points about that, in the light of what Bill Spight has been posting.

Therefore one could expect this project to end up spread over many pages of type MiaiValueStrictlyBetweenThreeAndFour. That is, it would be for the expert. It would not do the job the Endgame Reckoner seeks to do. It would probably not deal comprehensively with endgame mistakes, either. The tesuji would simply be shown as correct play, rather than flagged as such. The value to amateur dan players in general would be in fractions of a stone.

That is, it would have many of the drawbacks of a joseki dictionary. The fact is that (as far I know) a large-scale list of this kind isn't available, so this is one area where collaborative work by amateurs could move things on.

I feel much more committed to the joseki project right now. Considering how much detail the monkey jump reveals on close analysis, this is probably a three-year effort for the dedicated. Not to be defeatist, though: this area probably obeys some version of what I call one-ten-hundred in study, even if the 80 for 20 solution isn't a great one.

dnerra: I share your doubts about such a project. Even more than joseki, I think that you just have to learn to read endgame values out quickly. Relying on memorized patterns seems pretty dubious to me[1]. I think the most important part of precise endgame play is to take into account side effects that couldn't possibly be part of the patterns listed there. With side effects I mean sequences that will later make some defense moves necessary, cause another sequence to be sente etc.
I think a rather short list of the most common patterns (the 3/4/5/6/7 miai points 2nd line hanes etc.) would be useful for kyu players, however.

[1] As you mention the monkey jump, I always have to read out the correct response from scratch, and I don't feel bad about that.


Deebster: I was thinking about date formats. America and Europe use different date notation, i.e. America uses mm-dd-yy, whereas Europe uses dd-mm-yy. When I see 08.11.03, I have no idea if the date is the 8th of November or the 11th of August. I generally have to figure out who wrote it, and where they're from before I can decode it.

Could we decide on a standard? I don't think either side is willing to change, so maybe something like 21.Dec.03 could be our standard?

dnerra: I think if you use the standard notations, there is no ambiguity. American notation is mm-dd-yy, Europe notation dd.mm.yy., and British notation dd/mm/yy. Am I wrong?

(Benjamin Geiger: Actually, most Americans use mm/dd/yy. I try to use "dd Mth yyyy" (21 May 2003) unless otherwise directed. If a strictly-numeric format is needed (such as in a filename), yyyy-mm-dd (2003-05-21) is better than either of the aforementioned formats. Both are unambiguous. And two-digit years are evil.)

Velobici: I have often had the same problem. The European way makes sense in that the duration of the time period continuously increases (day to month to year). Unfortunately, when dates are sorted the results are nonsense (01012003, 01023002, 02012003). We can address both issues by using yyyymmdd or yyyy-mm-dd (e.g. 20030102 or 2003-01-02). The time period is continuously decreasing and the sorting works as expected.

dnerra: Hey, we are not talking about file names! IMHO, 010305 is unreadable no matter what convention it is intended to use.

Velobici: oh no, not two-digit years again...I just got through dealing with that a couple years ago. 2001-03-05 is so much better than 010305.

Deebster: Well for something that a computer might sort, the big-endian yyyy-mm-dd format is obviously best. However, I'm thinking of dates mentioned in the text, and it seems that we should be thinking of how humans can most easily and unambigously read it.

I think trusting that people will use whichever of the -./ separators is correct is probably a little hopeful.

unkx80: I usually prefer to write dates as any permutation of "2003 May 22", spelling out the month. But if written with all digits, I tend to use the yyyy-mm-dd notation, because of the sorting reasons already discussed above.

DJ: The international standard format for dates is yyyy(hyphen)mm(hyphen)dd, therefore today is 2003-05-22.

This was set by ISO (International Standardisation Organisation) and adopted by CEN (Comit? Europ?en de Normalisation) and all national standardisation bodies (UNI, DIN, BSI, AFNOR, ANSI, etc.)

Charles .. and by CRM.

Benjamin Geiger: But, do we really want 2003-05-22T12:58:10.3372-0400 ? (Yes, it's a valid ISO 8601 time stamp.)

Dieter: ... or need it, for that matter. Make yourself clear when a date could be ambiguous and the ambiguity would be bothersome.

Benjamin Geiger: Could we add something like this to Wiki Etiquette or SL Conventions?

  • Avoid ambiguous dates. Dates such as 4/2/2003 are ambiguous (is it 4 February or 2 April?)
    • Suggested date formats include yyyy-mm-dd (2003-05-22) and dd Month yyyy (22 May 2003), spelling out or abbreviating the name of the month.

Frs: I think, adding pages like Help?, Support?, FAQ? would make it easier to find help topics with SL. And I miss a help link in the navigation menue on the left.


Arno: Lately, I have been thinking about the wiki syntax. Two things that come to mind is, having "![" instead of "[[" to escape "[", as the "!" is used for all other escapes as well. And using ">" to indent blocks instead of (mis)using definition lists ";:".

Searching around on SL I see that many people would like to see the camelcase syntax go away - links could only be done with square brackets - I'm not sure this is a good idea - I like the name RecentChanges. And headings <h1-3> could use another syntax as the "!!!" is not really nice to look at. An idea is e.g. "== heading h3 ==", "=== heading h2 ===", "==== heading h1 ====" (on a single line).

Also there are alternatives to the current list syntax which would allow paragraphs in lists etc. However, that would induce a greater shift and would require relearning some wiki techniques - although nice I think it may be to disruptive. For an example see [ext] http://phpwiki.sourceforge.net/phpwiki/NewBlockMarkup

Another idea I had, was enabling short notes for path pages. This could be done with definition lists. I.e. on the path main page (e.g. HikaruHistory) you'd have "; [>Meijin] : Sai is said to be of Meijin strength." or some such. I.e. the notes would put the current page in context of the path. That way you could add a short note, how the current page "fits" into the path.

Also, the idea of allowing private sub-pages crossed my mind (e.g. for having your private "scrap book" where you can selectively make those pages public or not). However, I'm not sure if this would lead to too much a fragmentation of the "wiki space" (or whatever its name is :o)

Charles Arno, I'm a big fan of paths, and would be very interested in any enrichment possible of their syntax. For example standard footers, or thumbnails of a diagram, to be included in all pages of certain paths.

Arno: footer would be header as well (because the path navigation is shown at the top and the bottom of a page). Could you tell me an example for the thumbnail idea?

Charles For example, all the pages on the tsuke-nobi joseki path to carry the same starting position (joseki after first five stones). I don't know where: perhaps in the left side bar, perhaps where the Stones character is now.

Dieter: Another path fan here. If a page belongs to one path only I would like to have the previous/next feature included not only if you arrive on that page by the path itself.

David: I like the proposed "![" syntax. It's more consistent and intuitive. The ">" indenting would be useful, especially if it could work at different levels: "> > ". (I think whitespace should be allowed and ignored here; "> > Text" looks nicer than ">>Text". This would make it impossible to indent preformatted text, but you can do that just by adding spaces anyway.) Personally, I don't like "camel case" links, but there's no reason not to allow them. I prefer the current "! Heading" to "== Heading ==". I am slighly against the block markup because of the disruption you mentioned. My feelings are neutral about changes to

paths and the creation of private pages.

While we're talking about markup, how about having the wiki eat the first character of each line of preformatted text, so that the text itself is aligned with the rest of the page?


Tamsin: Is there any way to save material here while you're writing it, but in such a way that it is protected against editing by anybody else? I was half-way through writing a big chunk, saved it because I don't trust my machine not to crash, and found that it had been edited even while I was working on the next part, which resulted in my over-riding the other editor's work when I saved the new version of my work. Or perhaps is the only way to stop others editing works in progress is to head them up with the request "Please do not edit yet until this notice disappears"? I have nothing against people editing my work, but I want to save them the trouble of editing something that is about to change radically, anyway.

Dieter:

Under construction, please do not edit

should do fine, really.

Charles Well, that was probably me - if it's any consolation you did the same to me in reverse. Yes, post a 'work in progress note' by all means. If you get the SL notice informing you of someone else's edit, when you try to save, you can copy all that's in the window and re-use it. This happens to me all the time, by the way. In fact ten seconds ago!.

Dieter: I think I am a frequent frustrater of Charles' edits. Here's a vicious trick: when you feel someone else will soon mess with the page you're editing, save it quickly and then move on. He will lose valuable time doing what Charles writes hereabove.

Tamsin: Thanks for the above advice. I'm still quite innocent in the ways of wiki. I would like to point out that I've no interest in perpetrating "vicious tricks", however! (LOL) I only wanted to save people the effort of doing work that I would delete in any case. I'm nice, like that, you see.

Arno: if this is getting bothersome I could change SL's behaviour: e.g. locking pages for ~30minutes when one presses edit. I did not implement it, because I thought that simultanious edits don't occur too often. Another idea (instead of locking) is to show a red warning when someone has pressed the EditPage link during the past 15 minutes and has not yet saved the page.

Tamsin: The red warning idea sounds like a much better idea to me than locking the page. I often tidy up my pages very quickly after saving them the first time, so being locked out for 30 minutes would be very frustrating.

Arno: the page would only be locked until you press "save". So you can immediatly re-edit a page after saving it.

mAsterdam: Loss though concurrent update also happened at elementary moves, see JasonD. As the deshi population grows it will happen more often. To me locking for 30 minutes seems a little drastic for now. Furthermore (maybe I should do this another way) I press 'edit' in a second browser-window to get to the source of a page without intending to edit it, and usually without pressing cancel immediately. Maybe if everybody knows the locking time (say 5 minutes) it 'll work, even better if 'preview' re-initializes the timer. The red warning has my sympathy. I think it also would help to delay the appearance in recent changes for say 30 minutes if that is feasible. Just my 2 Eurocents.

NN?: How about a three-way merge like version control systems such as cvs use? This would prevent data loss at the price of requiring manual editing in case of conflicts.

Charles It is hard to see how to prevent 'competitive editing', without also removing the valuable possibility of real-time dialogue.

BobMcGuigan: Relative newcomer though I am I, too, have experienced a few editting clashes. It's really not much of a bother, but something like the red flag idea would be preferable to the lock-out, I think.

unkx80: I seldom encounter the competitive editing of a page (probably because I live in a different time zone from most of you), but I guess flagging a page as "under construction" will do. Locking don't seem a good idea for me, it is simply frustrating.

Arno: I have implemented the warning. Feedback? -> GuineaPigsFeedback.

unkx80: Very nice feature, thanks! =)

10 May 2003

Fhayashi: I love what AndreEngels is doing regarding the joseki listings. Whenever we all decide what the rational naming scheme for each variant is, we can go back and fix things, but looks like Andre is making the information accessible. I've already started to contribute my small bit...

Charles Yes, nice one, Andre.

6 May 2003

Fhayashi: Ericosman, what's up with the pages like "me" and "unfinished discussion"? Each page should at least attempt to make sense regardless of context. It's hard to imagine that anyone would look up useful go-related information about "me" or "unfinished discussion", and as it is now, someone has to take time to manually delete pages that don't make sense. You have to realize that SL is a community resource for information about the game go. There are plenty of free webspace available (geocities/yahoo hosting my own) if you feel the need to stake out your own share of cyberspace.

Eric Osman: I'll remove the "me" page. But please leave "unfinished discussion" as it's just that, one in which I'm composing the page. Thanks.

Morten - Eric, I look forward the form and content which "Unfinished Discussion" will take :-) However, as mentioned above, it might be useful if, next page you want to leave 'unfinished' you give a descriptive name which the page can keep afterwards as well. (e.g. MethodOfMultiples).

Morten- Joseki nomenclature discussion moved to OpeningSystematicClassification.

Eric Osman: 17:13 EDT According to Deleting Pages, pages can't really be deleted on this wiki. So in an attempt to delete the "me" page, please refrain from actually putting in any actual links to it. I'll also edit the orphaned "me" page to remove its content and replace it with one sentence saying "this page has been removed".

By the way, I'm surprised you even found the "unfinished discussion" page at all. I attempted to make that page be unreferenced until I completed it. Maybe the revision history info for my homepage is the only thing pointing at it ?

This brings up an interesting topic: Should orphaned pages be deleted, or should they be allowed as a way for users to temporarily keep pages private while they are still "works in progress" ?

Morten Some of the users can delete pages - you just need to leave a note here or elsewhere that you wish to see them deleted and it will be done (I removed Me) - You will need to do the same to have a page renamed. As for 'unfinished' being seen - most of the creatures who lurk here have a beady eye on the RecentChanges which shows pages being edited - and are usually quick to respond Big brother is watching you :-)

Charles WikiOrphans is cleaned up regularly. Arno seems to enjoy the effort to break the round number barriers like 6000 in the reverse direction from most of us: from above.

5 May 2003

Eric Osman at Sand Box, here edited slightly:
Hi, [email] ericosman@rcn.com here. I recently started composing a technical wiki page.

However, I didn't finish it, and when I came back the next day to work on it some more, to my dismay , others had already edited it.

Is there a way to create a wiki page such that it remains private, or at least un-editable, until we're finished creating it ?

Thanks. Please send email to [email] me . /Eric

David: Sure, do it the wiki way: Leave a polite note on the page explaining your intentions and asking that others not edit it until you're done. If you don't want people to even see the page until you're finished, you can edit it as a file on your computer and post it when it's done.

Eric Osman

       >  If you don't want people to even see the page until
       >  you're finished, you can edit it as a file on your
       >  computer and post it when it's done.

Well, I'd be happy to do that if there were a way to have the wiki formatter on my computer so that I could occasionally press "save" in order to see what the page looks like so far, just as I can here. Is that possible ?

David: If you just want to see what a page will look like, use the "Preview" button when editing. You can edit and preview a page without saving it. Copy the text in the edit box to a file and paste it back when you have your next editing session.

Are you sure that you want to do this though? Your Get Strong at Joseki problem discussion page has attracted thoughtful comments by strong players. In general, it's good to make your pages available so others can benefit from and contribute to them.

Eric Osman at Me, here edited slightly:
I want the sentence back there on the Sand Box, the one that says "send email to me" to have a link on the me so that when you click on it, it starts an email to me. But I don't know how to do that !

David: Do it like this: [me|mailto:ericosman@rcn.com]. There's more helpful formatting information at Text Formatting Rules.

11 April 2003

About how pages should look like. Take JosekiAsASourceOfBadHabits. I had a clear idea of what I wanted this page to be when I created it. It is rather of article bred, but of course I welcome other thoughts and examples. Now, with that clear structure present (1 up to 5) should it rather be a starting point with links to five (as of now) pages, each describing a specific situation, or does the grouping benefit the general idea. Myself I prefer many pages, many links and paths for a wiki, over few pages with big overlapping content. yes I have been an OO programmer. On this occasion, as the author of an article, I hesitate. --Dieter

I like pages being subdivided. I enjoy pages more when I can take in the content without scrolling. Of course more pages causes problems, but they are the same type as those caused by the rapid growth of SL. --Charles

On a different issue, a point was raised at Don't provoke damaging plays on the second line about labelling a given mistake as 'typical for 10 kyu' or anything similar. Is this an acceptable, positive or negative thing to do? Quite a number of years ago David Erbach expressed an interest in this kind of information: i.e. correlating levels of play with typical mistakes. When do players really start to see snapbacks, for example? I don't want to express my thoughts on this if it is considered derogatory. But teachers of the game might welcome information of this kind, if it was collected more systematically. -- Charles

dieter: Thanks for the explanation, I now understand better the reason for such a statement. I still think the times an 8 kyu receives it as a derogatory one outnumber the occasions that a teacher might take advantage of such information, all the more because we can only offer our best guesses.

2 february 2003

No pages StrategicGroup, Life, Death yet!!! --RobertJasiek




This is a copy of the living page "Meta Discussion 2003" at Sensei's Library.
(OC) 2004 the Authors, published under the OpenContent License V1.0.