To Sign or Not to Sign
Bill: The Wiki Etiquette page states:
- You are free to contribute anonymously, but it is preferred that you sign your comments with your name (or handle). It is common to prepend[1] the signature with '--' like this: "-- [Arno Hollosi], 1dan" (While you're at it, you are free to create your own wikipage and tell us about yourself. If you prefer you may put your rank on your homepage)
Recently Dieter and I have had a brief exchange about signing contributions on SL, and the question has arisen in regard to the SL II proposal, where signing is discouraged or forbidden in certain domains. So I thought I would create this page for discussion of the question: to sign or not to sign.
Bill: I'll kick off the discussion. Mostly I sign my remarks on SL, but sometimes I don't. But since we are talking about shoulds here, I have one. I think that opinions should be signed.
There are a fair number of unsigned opinions on SL. Mostly I ignore them. A lot of them seem to have an authoritative tone. "This is not just my opinion, this is how it is." When I see that I say to myself, there's a kyu player, probably a double digit kyu. Dan players usually know how much they do not know.
damien "Those who speak, do not know. Those who know, do not speak." Zen Proverb
To sign your opinion says to me, "I don't hide behind anonymity, I don't pretend to be the voice of God, I stand behind my opinion, I take responsibility for it." If someone isn't willing to stand behind their opinion, how much can it be worth?
(Strange. To me, it says "I put my name on this so that I can participate in status games and puff myself up a bit." We're all different in this matter.) -- Letmethinkofanamenow
I am also concerned when opinions and other ideas are not signed that casual readers or beginners may be misled into thinking that what is said is authoritative. For instance, in the 80s I coined the terms, primary, secondary, tertiary, and virtual ko threats. (See Ko threat functions.) Now, I hope that people will like them and that they will catch on. But I present them as mine. I do not present them anonymously, as though they were standard terminology. That would be misleading. People should adopt them, if they do, because they like them and find them useful, not because they think they are already accepted terms.
So I think that opinions should be signed. And a great deal of material on SL is opinion, even on reference pages. (There are few authorities on anything related to go here. We are amateurs, after all.) Do I think we should have a rule about it? No. If I see something here that appears authoritative, but I think is not, I may question it in a comment. Signed, of course. ;-) I think that that is good enough.
What about the proposed rule for SL II not to sign, say, reference material? Well, you don't have to be a deconstructionist to know that facts, well, may not be. Other reference material is even iffier. But at the same time, well researched material is not just opinion. And it may, like the definitions of some terms here, be the result of research by several people. So maybe it's appropriate not to sign reference material. But encyclopedia articles used to be signed. To sign something is to take responsibility for it. That's not such a bad idea, is it?
Bob McGuigan: I, too, am in favor of signing contributions. GIven the varying levels of experience and go ability among the contributors on SL, signed contributions allow for communication regarding possible errors or disagreements. Not having any name associated with something makes it harder to discuss, too. We do have the Anonymous user so that unsigned contributions can be identified with a name in discussions.
DrStraw: For a reference page I would prefer not to have the clutter of signatures on every paragraph. It disrupts the flow. However, I agree that the source of information should be indicated. How about this proposal.
In the footnotes for a reference page we should have a block for contributing authors. The information about each author should include name (or at least handle), rank and source of rank, and date of contribution. For me, this would look something like:
- Steve Fawthrop, AGA 5d, Oct. 2005
In the main text there should be a link at the end of each paragraph or item indicating the contributor. The list of contributors should be separated from any other footnotes, perhaps by creating a new link style (something like [*1]?). In this way the main flow is not disrupted, but all contributions can be identified by those who are interested.
This is system is still open to abuse. Anyone could claim to be 5d. Perhaps reference pages should only be editable by registered contributors, who must complete personal information in their preferences in order to be eligible to edit a reference page. Using this, the contributions could be automatically footnoted. This would be less open to abuse because the contributor infomation could be made available for all to see and comment on.
dnerra: I think this is very easy. If something is an opinion, or not authoritative, or original research, then of course it should be signed. But at the same time, it also doesn't belong in the reference section. (Think Wikipedia. No surprise one of the strict rules there is "no original research".) So I think it is right that contributions to the reference section should not be signed. I hope this might also increase the level of self-discipline in that section, by reminding everyone that he shouldn't be speaking for himself on these pages. In the teaching section, certainly both makes sense. An explanation of "This move is worth one point in miai counting." can be fact-checked by everyone here and needn't be signed. Other stuff might better be signed.
I expect the reference section to be rather small. Factual information on pro players, tournaments, ranks, maybe definition of common go terms (but no teaching on how to apply them).
Bill: From what I see on the SL II proposal page, the proposed reference section appears to be fairly large. Just looking at the proposal, I would think it would include go terminology -- which, as we know, has led to some of the hottest differences of opinion on SL. I would also think that it would include go proverbs, which, again, contain a great deal of opinion. Perhaps the overall plan needs more community discussion.
dnerra: Bill, I agree with you in the sense that I think that terminology for which there is no consensus shouldn't be in the reference section (unless of course the relevant article objectively describes the different uses of a term; for example an article on "Hane" shouldn't be included in the reference section until it describes the difference between the original Japanese meaning and the more general English usage (if I understand this correctly)). As for go proverbs, they should, in my opinion, belong to the Playing or Teaching section, but definitely not in the reference section.
Charles Matthews People can use page histories to track who adds what. The more time goes on, the less I like much of the signed stuff: signed questions on major pages, for example, should become query subpages if they have merit. Self-promotion and neologisms (sorry Bill, but I now find this unacceptable) in coining theory terms is something that seems out of place; use your own web space for this, and then importation here can be done as 'indirect speech'. The whole Japanese terms thing seems to have proved a dead end; basically few Western go players know enough about the game and the language to employ them in a way that doesn't cause winces from the cognoscenti. So, while recognising that the terms discussions have pushed the envelope on our understanding, I think they are something of a special case.
Bill: I don't want to get off topic, Charles, but, even though we disagree about the signed coinage of terms, can we agree that the unsigned coinage of terms on SL is unacceptable?
Charles I'd credit anyone able to coin an annoying new go term with the capacity to come up with a matching pseudonym. Why make empty rules?
Dieter: Repeating my earlier argumens here (BTW: I should move this page to the forum):
My main point is that, for a reference page, while it is in its embryonic state, signed contributions are best. After some time the discussion dies down, either or not with agreement. Next, a WMEditor cleans it all up and makes it a reference page without signature, possibly moving the major discussion parts to the discussion page (widen this idea to multiple pages and spin-offs with their discussion pages). Next, if any, discussion starts on the forum/discussion pages of the reference page. Even if that discussion involves questioning the work of the WEditor.
The reason is that you don't always know in advance if a certain discussion will contain material suitable for reference. Often it will, but as often it will start with what looks like mere discussion. So you cannot in advance create the reference page for which the discussion is at hand.
I don't like mentioning contributors. It's enough for me to know that I'm part of the project. My "accomplishments" don't have to be widespread (that would be my reason to sign, I'm not judging others'). I like even less signed contributions on a reference page, long after it has become a reference page.
DrStraw: Any reputable encyclopedia has the name of the author at the end of an entry of any significance. This gives an indication of the authority of the entry. Why should SL be any different.
dnerra: Dieter, do I understand you right that a reference page in its "embryonic state", when there are still signed contributions and possible discussion, should not be made part of the reference section yet? If so, completely agree. Once a page is more or less finished and WME'd, I think it should be a community consensus decision to move it over to the reference section.
ilan: I am against signing one's contributions.
dnerra: Let me phrase my argument against signed contributions on reference (and partly even on teaching) pages more to the point: A habit of signed contributions encourages thread mode and subjectivity.
And this has its place in SLII, namely in the Social/Playing section. But if we want to the reference and teaching section of SLII to be something different than what we have in the existing SL, we have to change some of our attitudes (I am not excluding myself here), and discouraging thread mode/subjectivity is one part of that.
DrStraw: dnerra, read my proposal on footnote signatures above. This will reduce the thread mode.
Bill: dnerra, except for cut and dried material on the one hand, and well it seems to me material on the other, I have not noticed any difference in subjectivity between signed and unsigned material. Anyway, that's how it seems to me. ;-)
nachtrabe: For what it is worth I agree with dnerra. Particularly in reference, there should be absolutely no need to sign contributions. Footnote citations just shift the problem, and don't directly deal with it (I anticipate it will look something like <statement X[1], but some think Y[2], and others think Z[3]> which basically leads us back to thread-mode.
DrStraw: No nachtrabe, you miss my point. I am not proposing footnote signatures for anything which is subjective. Merely acknowledging you wrote the reference material. Subjective material should not be on a reference page and so the example you give cannot exist there. Open up a page of the Encyclopedia Brittanica. There is no discussion there, but alternative viewpoints may be expressed. However, a single person writes the article and signs it.
Charles But this is a collaborative web site - says so on the front page. The analogy is flawed. The page history gives the audit trail, but the pages are collective (and in no sense 'owned', outside user pages).
Bill: Nachtrabe, if footnote citations lead to thread mode, then, as others have pointed out, maybe the page is not authoritative enough to be in the reference domain. Consider that there are several example pages on SL where the examples are not cited. Compare those where the author cites the example, giving book or article, page number, and professional author. Which examples do you trust more?
nachtrabe: Reference is very controversial, and teaching more so evinced by the hamete debate for the former and aLegendWai's commentary on Basic Instincts for the latter. As to what I trust more: There's a huge difference between citing a book or professional author and citing one's own opinion at any strength short of professional-level. If I write something and say "this is jeongseok[RUI98]" ([RUI98] as a footnote referencing Rui Naiwei's Essential Joseki or a [GUO05] for a 9-October-2005 Lecture by Guo Juan) is very very very different than me saying "this looks like it gives an even result" and citing myself as a 5k. If it is "not authoritative enough" to be in the reference domain or teaching domain, then it is not where my issue with signed contributions lies. If it comes from an authoritative source, it should (already) be cited, but that's very different from me signing my own contributions.
For reference, signed contributions just seem silly to me--what advantage is there in having them? Okay, so a 5k said that this person was born in 1982, why does it matter? A 20k wrote the definition for a peep, not a big deal. Signed contributions in this case just make it harder to edit and respect what everyone wrote and disrupts the flow of the page needlessly.
With most teaching (e.g., "Basic Instinct"-style pages) a statement that is "true in my experience" is probably worthy of discussion before being added to the page if the page is already in the teaching section. If it is controversial it can be flagged and discussed, but again, the author doesn't especially matter. Where possible, it is probably better to cite where it comes from (e.g., Lessons in the Fundamentals of Go), rather than to sign who contributed it to the Wiki.
Agree, but we have very few pros contributing here. When literally citing, we must take care with intellectual property (ooh, opening that box again). I wrote much of the basic instinct page: I think it is ok for 15k to 2d to write reference material, if the subject is basic enough for them to fully grasp it. Even then, I made some heavy statements while building basic instinct. It has been contested later and then adapted. I wouldn't mind if that page's content changes considerably, as long as that's what's the community regards as reference material - I'm sure I have overstretched my abilities when creating that article.
DrStraw: Having read all these contributions I am beginning ot think that we need a new type of page: InDevelopment?. It could be development for anything, but most likely reference. Such a page would be a working version of a reference page, with footnote contributors as suggested (or other method of identification if that is not to the majority liking). Once there has been enough agreement (through a forum page) the InDevelopment? page can be cleaned up and inserted as a reference page without signed contributors.
Bill: One thing I hear from people who object to signing reference pages is that having names scattered through the text breaks the flow. When I first came to SL, it was usual after a WikiMasterEdit to list the authors at the bottom, without attributing any particular text to anybody. Now, that would not break the flow, but would still give information to readers who want to know the source of the material. Would that be acceptable?
DrStraw: That is exactly the idea I have been pushing for the last week.
Dieter: It is acceptable to me, but I don't see the point besides pride.
DrStraw: I see a few advantages. 1) The author is known in case of issues. If someone has made an effort to write a reference page I believe that it is common courtesy to consult them if a change is to be made. I agree this information is available in the change log (but not easily) so it is not a very compelling reason. 2) People are less likely to write reference material without thoroughly thinking it through if a contribution is signed (and hence improve the quality) . 3) Pride apart, I think the efforts of frequent contributors should be acknowledged. (No I am not looking for personal pride here - I doubt I will have time to write a lot of reference material, if any.)
dnerra: I strongly disagree with your reason 1), and I feel this is the essence of our disagreement. A wiki where you first have to ask the original author of a page before you improve it, isn't a wiki anymore. And besides, it just doesn't work. Probably the author is following SL's changes. If so, he will see your change anyway and can make comments if he disagrees (and is much easier to discuss an actual change than to discuss a suggestion to change something). Or just leave it at that if he agrees, whereas he would first have to signal agreement if you just suggest the change. However, if the author isn't following SL's recent changes, you will have made a comment, forget about it, and the change will often just not happen. I don't believe your reason 2) btw. Whatever, I feel like Dieter about signatures at the end -- nothing much wrong, but I don't see the point. But if it gives the impression that the original author first has to be consulted before a change is made, I am against it. Wikipedia has an informal rule be bold. If you see something needs improvement, improve it (after checking of course, that this is not something that has been discussed before).
Bill: I don't want to get off topic, but why do you say that it is easier to discuss an actual change than a suggestion? (As you point out, there may be no discussion at all. But there are often edits that you know are likely to lead to discussion.) We had extensive discussion before changing the hamete page, and I think that made it much easier than if we had gone back and forth on the main page itself. The final edit was quick, easy, and clean.
Likewise, in the current WME of the ten thousand year ko page, suppose that I had deleted the reference to thousand year ko, then velobici had not only reinstated it, but made thousand year ko the main page, and then Bob Myers had changed the text to standoff ko. What confusion! Discussing a substantial edit ahead of time seems much better to me.
And I am not arguing that the original author needs to be consulted before making edits. I am just talking about the ease of discussing edits at what time.
Bill: When I kicked off this discussion, I was of two minds about signing reference material. For myself, usually I do not, but sometimes I do. However, I was thinking as a writer. This discussion got me thinking as a reader. And as a reader, except for such cut and dried stuff as the list of modern Honinbos, I want to know the source. For instance, I have more faith in something authored by you, Dieter, than by a lot of other people, questions of playing strength aside. Now if you or any other author wish to remain anonymous, I respect that, but I still find the author information valuable.
Charles I don't like the signed WME concept. It seems to be a covert way of implying closure of a discussion.
Bill: Well, not signing it suggests that this is how it is. No discussion at all.
Dieter: Take the Jowa page. It starts: "considered by many as the strongest tactician --Bill."
When you enter this page, there are many thoughts possible:
- Bill thinks that many think Jowa is a strong tactician.
- Bill thinks Jowa is a strong tactician
- Bill knows more about Jowa than other people
- Bill is a strong tactician
- Whoever Jowa may be, his existence is opinionated. Should I read on?
It's all very Bill and not so much Jowa. I don't think this is your intention: you just want people to know that Jowa is believed a strong tactician. I also don't believe your authority is necessary for most people to believe that many people think Jowa to be a strong tactician. I think people want to find out about Jowa, without being too distracted by the personalities that populate Sensei's Library.
DrStraw: I have to disagree. It is not all about Bill. It is Bill who is lending his authority to it. Assuming I know who Bill is, and that he has some credibiility I am must more likely to accept this statement than I would if Joe Beginner stated it after only two weeks of playing because he wanted to get his name in lights. That is why a signed footnote, stating Bill credentials, is useful.
Bill: I think this is a good example, Dieter. I added that comment to the very first version of the page, which had very little about Jowa on it. The fact that it is still there four and a half years and fifty six versions later is a surprise to me. Why it heads the page, I do not know. If I am the best source on the page for how Jowa's tactical skill is viewed by history, that's a shame. Anyone who knows me knows that I am well read, but not a historian. So my impression is worth something, but not much. It would be good for someone to do the research and replace my comment with better material. What would not be good would be to leave the comment without the attribution. Then the reader would have no way of assessing it. Maybe it would be better to delete it entirely, but so far the people who have edited the page have seen fit to retain it.
Dieter: Maybe the example was way stronger than the point I'm making. More funny than fair. It still doesn't deal with signed advice on how to play. In that grey area we disagree, it seems. I'll WME Jowa.
Bill: If you retain my comment, please attribute it, so the reader can assess it.
Charles Really, that doesn't make much sense to me. Data on Jowa's reputation must be in many places.
Bill: Fine! Somebody go get it. Replace my comment with good research.
Gronk: Right! Why isn't the standard more like a scholarly article, in which nearly everything documentable references other sources writen by experts on the subject. Anything that is opinion is clearly just that (either explictly or implicitly). By this standard, Bill's original contribution on Jowa ("considered by many as the strongest tactician --Bill.") would be judged as poor scholarship and probably stricken. But something like "I recall having read that he is considered the strogest tactician--Bill" might be okay. Now, by my standard, what Bill actually wrote amounts to the latter since it is pretty clear from context that this is more-or-less what he meant. But that's only because I've read enough of Bill's work to more-or-less trust that he doesn't make stuff up. This places a significant burden on the reader though. So, to sum up, I would recommend documenting the documentable or not writing it at all. If it is undocumentable then it is opinion and that should be made clear some how.
Charles Frankly, it takes nerve for an amateur to have an opinion on Jowa. The guy might have made it as a go saint, if he'd been less of a Machiavellian. But Bill has a good knowledge of the Japanese literature.
Gronk: I don't mind who has an opinion of whom. Point is, by academic standards, a reference to something that is complete enough to be able to be judged as a source of authoritative information is perferable to a signature, no matture who is signing. Of course reputation does exist in academic circles too and goes a long way. But it assumes quite a lot of the reader (that they know this guy is an expert on this topic) and it does frequently get abused. Anyway, from the point of view of serving the SL community, a reference is a very nice thing to provide. Presumably some readers of the Jowa page may want to know more than a contributor wishes to write there. If that contributor knows a source, it would be quite nice to give it. If not, on what is the contributor's contribution based? Is it just opinion? Fine, make it plain...which, in fact, is what the signature may do, if that is the convention we adopt.
[1] Hu: That is not prepending, it is appending. I have prepended my name to my comment.
Bill: I think what was being said was that you could prepend your signature with --. like so: --Bill, when you append your signature to your comment.