What do you think of sensei's library

[RobertJasiek]

Subpage discussions don't help people find pages.

Good structure for ALL pages does.

[Thunkd]

Subpage discussions don't help people find pages.

Good structure for ALL pages does.

Agreed, but that requires people to edit existing pages and to maintain the structure on newly added pages. The lack of editors seems to be the problem. My suggestion was an attempt to avoid requiring more editors as they aren't available.

The biggest problem I have with SL is separating the poor pages from the good pages. The objections to my idea seem to be that ranking is subjective and that problem pages are already known. But we still have the problem that it's hard to find the good pages amid all the not-so-good pages. So if you are going to critique my suggestion then I challenge you to provide a better suggestion to resolve the issues with helping users find good pages and avoid bad ones.

[RobertJasiek]

Good structure is my suggestion, regardless of whether it requires work. If people cooperate and share work, it can be done. I would do part of the work, say, all pages related to rules. Before a start, we need a list of all existing pages.

[Kaya.gs]

It is clearly indecent to have the bold text into the definition of the word.

Many English language go terms are invented and spread by amateur players without being considered indecent, as they are used in the books we all used to learn from. Should new English go books don't invent anything new? I don't think so. Should innovation be allowed only in books but not elsewhere? I don't think so, closure of terminology comes with maturity. In the special case of the proposed translation, it is at least a genuine attempt to offer a translation as opposed to the vague descriptions for an untranslated term we have. It has been adopted by at least one professional teaching in the west. It is hilarious that writing "has been proposed by XY" makes it sound indecent, while a blunt statement without attribution likely would have been perceived differently, right? Aji as a term is in my experience as over-used as it is under-understood, which is caused by having many vague descriptions on display, but no English term.

Disclosure: I am somewhat responsible for the current state of the aji page. I took lessons from the early adopter professional and I wouldn't feel comfortable to discard a term my teacher uses.

'

Regardless of what any opinion i or anyone could have about how good the term "funny business" is to describe Aji, personal opinions are way out of place for a definiton. I challenge you to find a term in a dictionary that includes an individual's interpretation included.

The inclusion of a name is as akward in a defintion as mentioning some people's opinion. "Some people think..." is as bad as "x guy thinks.

But its not the point to discuss or mention if that line is appropiate or not. The point is that the text is not organized in a paragraph or template model like it could be ideally. If there was a template for go terms, it could have a "definition" paragraph, "content" "opinions" etc, and it would be entirely possible to mine it out automatically. Now my only option is to edit the solutions, which will take a lot of time.

[emeraldemon]

So are you looking for something like formatting the first sentence of every page to be extractable as a definition?

[Kaya.gs]

So are you looking for something like formatting the first sentence of every page to be extractable as a definition?

Maybe. There are many formats possible, i could take whole paragraphs. Just any organised way to order that clears out the defintion would give me excellent results. Potentially i can just take as much text as i want without being intrusive. But the problem there is things that i think make them poor definitions, like the case above , or the inclusion of peoples comments.

[tapir]

But its not the point to discuss or mention if that line is appropiate or not. The point is that the text is not organized in a paragraph or template model like it could be ideally. If there was a template for go terms, it could have a "definition" paragraph, "content" "opinions" etc, and it would be entirely possible to mine it out automatically. Now my only option is to edit the solutions, which will take a lot of time.

It is very unlikely that the whole structure of SL is changed to provide easy extractable definitions. In fact, there is debate whether we have too much definitions on SL. See: [sl=Definitionism]Definitionism[/sl] - and there is a practical problem too: a definition isn't worth anything if it is just an opinion in disguise. What makes definitions work, is a stable demographic that uses a special language and can agree on the definitions. English go terminology simply has not reached this maturity yet, that it can provide stable meanings for words. Especially for terms such as Aji, where we most of the time help us with vague descriptions. Looking at CJK practical use of words instead of providing definitions, which may be right or may be wrong or may leave everyone confused, seems like the better choice to me.

[hyperpape]

I think the point about rigorousness and importance of definitions is a bit orthogonal to the point about organization. Even a loose definition can be set apart in the text so that it can be extracted by a script. A Robert Jasiek style definition can be shoved into the middle of a discussion.

Also, this discussion is a bit confusing. Sometimes I think Kaya is objecting to the presence of "funny business" (misguided, imo), and sometimes I think he's objecting to the "has been proposed by Ed Lee" (which isn't good, but is also minor).

[hyperpape]

Here is another example where it is easy to help contribute to sensei's. There have been several rules disputes mentioned in our discussions of L19. Last year, there was a case where a team that was playing to win on time was somehow forced to resign or was disqualified. There is a sensei's page on rules disputes.

In general, pages like this are not likely to affect the overall quality of sensei's, but they could be an important adjunct to individual memory. And of course this will not just be true of rules disputes, but any notable events in the Go world.

[cata]

I think that it's totally appropriate for SL to be as unstructured as it is. The nature of a wiki is to provide a very low-friction way for someone to put their thoughts on a page, and it's the content that counts, not the organization.

Someone like Dex can come along afterward and clean things up, if he wants to use the wiki for some useful semantic purpose from his software, or people can reorganize in the spirit of making information easier to find. If you don't like the bold text in the middle of the paragraph, then make some footnotes, or put it at the end in parentheses, or something. That's fine. But the most important thing is keeping people writing in the first place, and if you make contributors adhere to a bunch of writing and organizational and citation standards, then that won't happen as often.

[Mivo]

I actually prefer the discussions and comments on the content pages, rather than on separate discussion pages. Though I'd agree that a consistent format would be preferable to the current "mixed" approach, even if this means consequently separated discussion pages. I agree with Cata that the nature of a "true" wiki is to be a little "wild" and perhaps even "chaotic". For popular projects like Wikipedia this may be impractical, but the Go community here is so small still that a less strict way works well, too.

[Kaya.gs]

Just to make it clear, my observation had no intentions to put work on senseis to get organized, i just stated the shortcoming i suffer.

Be that as it may, as senseis is already half-way included into Kaya, it wouldnt be bad to brainstorm or explore the possibilities on how can both parties get more out a synergy.

As i skim over this post, one of the key issues in Senseis is getting people to work and edit it. I said before that i would expect Kaya's pages on senseis to have activity because users will be able to edit them and see the glossary changes right into the server, but its a very small window of benefit for senseis, other than branding (as any go term looked-upon will have sensei's name).

Given a potentially big flow of people going from Kaya.gs to Senseis, what could senseis get from it? or how could senseis transform them into regular users?

From our standpoint and as it is, what we get from sensei's is content(although we will have to re-process all the content we use), but maybe it has a lot more to offer us.

Time to get creative

[gogameguru]

My biggest question is, why are so few SL contributors 4d+?

Why are the top players completely disinterested in submitting authoritative content?

Most edits are anonymous so you don't really know anything about the people contributing the content. I've contributed a reasonable amount over the years and I know some stronger players do too. It's just that it's done in an occasional, adhoc way, rather than being a formalised and regular routine. Isn't that how wiki's work? And it's hard to track what's happened unless you look at Sensei's every day.

Also, I think Sensei's took a hit when the community sort of forked and people went over to Wikipedia. Sometimes now I find myself wondering whether it's better to add new content to Sensei's or Wikipedia and I think a few other people do too. Sensei's has the advantage of supporting go diagrams.

I think more integration between go sites would be better for the community too. As a simple example, can the people working on josekipedia set it up to autolink go terms back to the Sensei's pages? Can Life in 19x19 do this? We've already set this up on Go Game Guru and we already link to a small number of terms. I'm in the same position as Gabriel, where it takes time to review each page and decide whether it should be linked to. Some definitions are really confusing and I'd rather write my own, or rewrite the Sensei's page eventually. That's partly why I haven't done it yet.

If someone wants to help out and prepare an organised list of terms with matching URLs, I can add them to our database and it will autolink those words whenever they're used in our existing articles and in all reader comments. Here's an example of what the formatting for the list looks like:

eye|http://senseis.xmp.net/?eyes
two eyes|http://senseis.xmp.net/?TwoEyes
ko|http://senseis.xmp.net/?ko

One new term per line.

It's most useful for language that is very specific to go and foreign language terms. For example 'furikawari', 'two eyes' or 'eye' (singular) - words that go players use when they talk or write comments, but are rarely used in normal speech. Words like 'jump' or 'opening' are too ambiguous to autolink (but fuseki is clear). Maybe someone could start a master list of the good quality definitions on a Sensei's page (my preference is the format above and if you enclose each line in square brackets it will work on Sensei's too) and then other go sites could use it as a resource for creating links to Sensei's? The page itself could be used as a checklist of sorts too.

I think if we start sending more visitors to Sensei's it might encourage more people to contribute to it. Wikipedia has the advantage that nearly everyone knows about it.

I think I mentioned this somewhere previously, but since there seems to be more enthusiasm for working on Sensei's at the moment I'll mention it again. Go Game Guru is licensed under Creative Commons. Isn't Sensei's too? If so, Sensei's can just take excerpts from our content (including bit of commentaries) on the condition that Go Game Guru gets attribution (in the form of a link back to our original page for each excerpt) and Sensei's doesn't use it for commercial purposes (not a problem I think). Don't just copy our pages wholesale because then Google will think Sensei's is a content scraper and Senseis' pages may start disappearing form the index. Relevant excerpts (or what Google call 'snippets') are not an issue though when they acknowledge the source.

We decided to license the content this way largely because we thought people might use some of it on wikis or translate it. There are quite a few people translating under the terms of the license already. If we let people take excerpts and translate instead of just linking to us, then some of the more important content will survive even if the three of us are killed in a plane crash.

Just because we don't personally want to write all our content on Sensei's library, doesn't mean we can't work together.

[RobertJasiek]

My biggest question is, why are so few SL contributors 4d+?

Because there is an only small percentage of such players?

Why are the top players completely disinterested in submitting authoritative content?

Whom do you call top players? With Bill Spight and myself there are, e.g., two top researchers. If you follow RecentChanges, you might not always notice because we do not always login. If you asked why top players would not contribute 24h a day, well, their (like everybody's) time is limited. E.g, it can happen that I don't have any time for SL for a couple of months; it is simply impossible to write a go book while editing SL.

[hyperpape]

Sadly, Sensei's is not creative commons. Creative commons did not exist (or wasn't ubiquitous) when Sensei's was started, so it uses an older license that is incompatible. Given the amount of content on there, it is likely that it will never be possible to mass import Sensei's content to any other site.

It is possible to dual-license your own content so that it can be added to Sensei's.

License incompatibility is a real shame, in my opinion.