Document Mode vs Thread Mode

Is it more important to strive for a beautiful wiki or to preserve the integrity of personalized statements? See e.g. Footnote 3 in version 47 of Sake bottle shape.

Charles Wiki theorists talk about ' document mode' (implication: good) and ' thread mode' (implication: so-so). What goes on at SL has more thread mode than the purists would like; and therefore a quite high proportion of signed stuff. There are two sides to this:

Pro: It matters greatly whether a poster is 20 kyu, 2 kyu or 2 dan, in knowing what importance to attach to a posting – the convention is to indicate strength on your home page (which is mostly better than writing it in the text). Some of the major contributors only write in threads, and don't edit.

axd 18k: as contributions can become outdated, it makes sense to "hard-code" the contributors strength at the time of writing. One's position can change over time, especially because of growing experience; the hard-coded rank would clearly indicate the staleness of one's remarks, and this would be confirmed when visiting the contributor's home page to see his rank had changed meanwhile.

Con: There is supposed to be a conversion process (WME) taking thread to document, so that we get some more definitive pages out of all the dialogue.

Since in fact few people engage in WME, it looks like thread mode is winning; with the result that there is always a proportion of unreliable material here. Now I'm not too unhappy with that, since closing down issues by premature WME seems also to be a loss – and I think wanting that sort of 'closure' on a topic goes against the grain of the nature of go.

Sebastian: Well, there are two other ways to mitigate the problems you mentioned:

1. During WME, use Compost Heap to give the author or experts a chance to extract useful information and incorporate it back into the main text.
2. Encourage Wiki apprentice edits (or "baby step WME"). Some people, such as myself, would like to help by editing those parts of the document they understand, while leaving others intact. This could be facilitated by
   • a convention how to handle named contributions
   • a convention how to mark these edited parts

There can be many ways to achieve this. One way, which would combine both objectives, would be to use footnotes with ridiculously high numbers that mark the beginning and end of the edited area and individual contributions.

Klaus: I think, names are useful during the discussion, but it should generally be allowed to ignore them in a master edit.

Charles Yes, in a WME one credits previous authors, at the end.

By the way, like Sebastian, I am not sure I am allowed to make a master edit of any page after the main discussion has taken place. For example, I would like to try one on invading deep enough within the next couple of days if that is OK.

I would suggest, that if someone stupid like me makes such an edit, it should first be presented on a new page for a week, and only later replace the original page.

Charles Enough to add some header say WME – please comment..., and see if anyone reverts to an earlier state.

Charles That rather assumes that the editing problem is the main one to solve. I'm actually more concerned about 'page churn': how long between mentions at RecentChanges for typical pages?

Bill: Before there was Wiki, people were talking about similar ideas. One that I found appealing we might call Network Mode. In the extreme, each person starts a new page for his or her comments and links to it. I doubt if that model would work well for SL, as there would be too many pages.

Another idea is one that was common on some conferencing software I used in the 80s, and we use to some extent here, and that is to have separate discussion pages. So you would have page XYZ cross-linked to page XYZ Discussion.

Both of these ways of organization have the virtue that there is little need to edit. IMO, that is a big virtue. :-)

Sebastian: I agree that it's always a virtue to have little need for anything. But editing can be a good thing if it saves readers from having to extract meaning from pages of discussions, and there are some people who would enjoy a little bit of editing. I wish we had an easy way in between the two "modes". Currently, there is a high threshold that deters potential editors, particularly if they're considerate.

Bill: I have not always been happy with how I have been edited. For instance, I am listed as one of the authors of the Basic Instinct page, but I do not recognize anything of mine there. ;-)

Sebastian: I'm sorry about this and I hope that none of my edits distorted your contributions. (If so, pls let me know by reverting it or by commenting on the page or by mail). Can you pls be a bit more specific about what bugs you and how it happens, so that we all can learn from the experience?

Bill: To me, a big problem I have seen with editing is a lack of care for context. After editing some things just don't make sense, because of the loss of context or of proper links. You cannot edit hypertext like regular text. When people are just interacting, they do not have much care for that, but someone producing a hypertext document needs to keep context in mind. Editing a discussion — moving from thread mode towards document mode, in a sense — is not easy.

Charles Bill doesn't appear in the page history, but the exchange on his home page implies he made comments in the past, perhaps on some other discussion page, and is being rather scrupulously credited for those. That's easy enough to fix.

Bill: It has been some time, but there were some things I questioned on the original page, and perhaps some links to other pages. As I recall, when I was first listed as an author, it was clear what my contribution was. That is not so for the current version, and the page appears to have grown. I'm not really sure of the material, and feel a little uneasy about having my name on it, but not enough so to take my name off.

Anyone writing here is contributing under the terms of the OCL (see SL Copyright); which is of course a different matter from the polite treatment of contributions. Actually this is a most polite place, by wiki standards. Compared to Wikipedia, there is much less noticeable 'bold' editing that has subsequently to be reverted (Wikipedia encourages writers to 'go for it' and sort out the mess later). There are few if any annoying deshis.

All this leads to an idea that writers here don't really need a policy of active management of their contributions. Here I differ from Bill. I like page churn, as I have said; and I'd see as consistent with that the thought that everything on SL should be reconsidered on a time scale of six months. Things in the SL archives are (in my opinion) best brought out into the daylight on a regular basis. Where Bill, or John F. or others write authoritatively on some topic, it is quite likely that as much harm as good is done by coarse editing: I see that. But as a general rule we should be working constantly to upgrade SL by (at the very least) creating paths and shaping what's already up here.

Sebastian: I agree with you, this is a very polite place. I found that the harshness on Wikipedia came from both sides, vandals and admins. One thing I like in Wikipedia, and which helps valuable, authoritative contributors keep track of what they've written is the option to monitor their pages by entering them in their watchlist?.

Charles Wikipedia also has talk pages for each page (big overhead): XYZ Discussion is automatically there to create for each XYZ. This is probably the way to go for an actual encyclopedia – but highly impersonal, really.

Bill: I get a sense of this discussion in favor of moving SL closer to document mode. I agree, because I sometimes have difficulty separating the wheat from the chaff. At the same time I feel that the discussions are quite valuable. I do not see SL so much as a reference as a community. :-) What we need is a good balance.

Charles, in my experience with the XYZ cum XYZ Discussion format in the 80s, we did not find it impersonal. For certain types of material, where XYZ could function as a reference, it was quite convenient. It was one way of achieving the balance I'm talking about.

Charles Oh yes: first priority having an open door, second priority SL the online community (on Wikipedia you cannot sign contributions, just talk, which is what makes it like a ghost town sometimes). After that, my personal hopes and aspirations are

1. continue as an ongoing 'seminar', without making a strict distinction between the deshis and senseis (as we know, there is only one real sensei here, and Sensei's influence is indirect);
2. assemble enough good material to document all you need to know to become 5 dan, practical play aside (about a third of the way there, I guess);
3. organise SL in a way that makes apparent what sort of content it has: i.e. give a good overall picture of go.

Now if we were closer to having covered #2 of these, there would be more reason to boil down what we have: perhaps when BQM200 comes round it may start to seem that the questions are of a familiar type. But not yet, really. Some of the basics on proverbs and joseki are here – hardly an exhaustive coverage, though.

dnerra: Being one of the moderately regular contributors, I almost only add in thread mode. That has a reason. I always feel a little uneasy trying to outline general theory. My feeling is that one has to be either extremely specific, or extremely vague, or partially wrong. Of course this feeling is contradicted by a couple of well-written books; maybe I cannot really judge since I have only read 5-6 go books. On the other hand, I myself found direct comments by stronger players on either my plays or on my thoughts and judgements most helpful for improving. So I try to be helpful in the same way to weaker players.

Dieter: Being one of the more zealous Wiki Master Editors, it won't surprise that I am more in favour of document mode. I wholeheartedly agree with dnerra that specific comments on specific positions can help improve a player much more than outline general theory. The problem with SL is that the effect of such comments has a few minutes' range but the page remains for ages. It is unlikely that new people will have the same question and if they do, the thread mode makes it unlikely that they will find the answer to their question. WME goes both ways: for a while, the "pupil" will have the opportunity to see all reactions to hs question. After the WME, other curious cats will find something more general, which may not fulfil their specific needs but is more likely to ring a bell.