Scholarpedia talk:Further enhancements

Discussion!

Shouldn't the task of improving the article be worth more points than clicking a reject/approve button?

- Nick Orbeck 11:58, 30 October 2011 (UTC)

Perhaps ? This would be rectified somewhat by the first proposal, no? Leo Trottier 17:56, 3 November 2011 (UTC)

I attempted to edit the unpublished article bar code scanning, and received the following error:

Error: You are not a (co-)Author of this article

Often new articles contain markup errors (usually leftover LaTeX), or do not conform to the article template. I try to copyedit articles before they are submitted for review. Clearly changing/correcting someone's unfinished work has the potential to cause aggravation. However, I have received only positive feedback.

"Unpublished" articles are publically accessible, indexed in Scholarpedia, AND indexed in Google. The warning message: "This article has not yet been published; it may contain inaccuracies, unapproved changes, or be unfinished." appears in a very very light grey. This message should be more overt.

I suggest that assistant editors be re-granted the authority to edit unpublished articles.

Also, the error message above should read: "Error: You are not an author of this article."

- Nick Orbeck 19:47, 30 October 2011 (UTC)

This is happening very soon. Every user with an summed index > 0 will have edit privileges on in-progress articles. Leo Trottier 17:56, 3 November 2011 (UTC)

The math font is too small:

1. This is normal text.
2. \(\textstyle \mathrm{This \ Mathjax \ text \ should \ be \ the \ same \ size \ as \ normal \ text.} \)

\[\displaystyle \mathrm{This \ Mathjax \ text \ should \ be \ larger \ than \ normal \ text.} \]

fixed it for you :) . Why should it be bigger? We are currently using the MathJax defaults: http://www.mathjax.org/ Leo Trottier 17:56, 3 November 2011 (UTC)

The 2nd line is rendered in "\textstyle", which is used to specify inline math markup.
The 3rd line is rendered in "\displaystyle", which is the default.

- Nick Orbeck 14:35, 1 November 2011 (UTC)

Is it really good to treat "Encyclopedia of X" pages as articles?

I'm puzzled: I'm not convinced that treating the "Encyclopedia of X" pages as articles authored by a few editors is a good thing for the following reasons.

• This tends to create inequalities among editors responsible of a category page and editors responsible of a category page AND of an encyclopedia of page.
• Within the framework of a given encyclopedia of X page, if there are editors of subcategories of X that are excluded from authorship of the encyclopedia of X article this is even less fair. In particular the editors authors of the "encyclopedia of X" article could refrain from inviting other editors of subcategories, not to have too many authors around.
• After say 20 years from now the editors of X will have to write in a page authored by the editors of the page in 2011 (but not say those of 2015...).
• Editors tend not to maintain the list of articles in the "encyclopedia of X" page, while this is automatic in the category page.

The only positive point in having authors/curators of"Encyclopedia of X" pages is that they will responsible of the articles listed in the page, vouching for them and eventually ignoring potential bad articles entered by mistake in the circuit. (But in alternative the [[Category:X]] could be removed from articles the editors do not want listed in their category page.)

So, why not simplifying and be fair, having only category pages maintained by editors? Or if those pages are needed, why not having "encyclopedia of X" pages without authors (i.e. in Scholarpedia name space)?

NB: I'm author of an Encyclopedia of X page, but I would be ready to renounce to that. --ric

Leo: I'm also a bit confused about whether we should have Encyclopedia pages in the main space. They might work, insofar as they coud (e.g.) be regarded as "books", or some kind of more prestigious kind of category. But to be honest I'm not entirely sure. --Leo

Eugene: My main motivation for "Encyclopedia of X" page to have an author - its editor - is to give the editors more power in deciding what articles go to the encyclopedia and what do not. For example, there is nothing prohibiting an article on "Consciousness" to be included into the category "Physics" (e.g., by Penrose or others who think that consciousness has a connection to quantum physics) and the editor of Physics would be powerless in blocking such an article from being listed in his category. However, if he is an author/curator of the page "Encyclopedia of Physics", then he has all the control not to list this article there. This is the only motivation. If the confusion resulting from "Encyclopedia of X" having authors is greater than the benefit, then let us remove such pages completely and use the category pages "Category:Physics" for the portals to the encyclopedias.

Categories of users

Ric:My proposal is to have the following categories of users:

• category "expert (or sponsors)": members are endowed with sponsorship power equivalent to curators (and any other power not related to curatorship of an article). This category is composed by: authors of published articles, curators, former advisory editors and other invited people.

Comment The reason for this category is, effectively, for reviewers, right? I don't think we should have this user type because it adds complication with little benefit; there are enough scholarly topics that nearly every genuine expert could become a curator of a topic if s/he was so inclined. Furthermore, this would prevent non-experts from being authors, which could stifle participation from students. Finally, any non-curator expert that is so enthusiastic to review (etc.) can petition to join as an editor. I propose another way of accommodating external reviewers: we allow any user given the reviewer URLtoken to approve of an article, but if they are not a curator this approval is not sufficient. When the Editor sees that the external reviewer approves, the editor will approve, noting that this approval is based on the approval of the external reviewer. As you mentioned, this will mean that one more Curator/Editor approval will be required. I think that it is reasonable, here, to expect the Editor (or the article's authors) to find a reviewer who is already a Scholarpedia Curator, or alternatively to persuade another Editor to sign-off on a second external reviewer. --Leo

Actually, my main reason for the category experts/sponsor is 1) to empower in some way (and thanks) co-authors that have no (senior) curatorship (I assume that right now they have not curator-powers and that they cannot act as sponsors-- if this is not true stop me) 2) to create at the same time a place for advisory editors (role which is lacking in SP2 but I think is important, and all in all should be distinct from editorship, to avoid multiplication of stakeholders in the same category). Never thought to reviewers. A young student will become an expert if s/he will co-author an article, that would be an incentive, non? But maybe you're right and simpler solutions can be found....--ric

The fundamental power of Scholarpedia, going forward, is the determination as to who will be each article's single Curator, and the subsequent pairing of power and responsibility. It is on top of this that the long-term viability of the project rests. The fact that it will often be a difficult, and perhaps painful, decision is a good thing, for it means that there is a revealing of information as to who genuinely is the best authority on a topic. The process begins with article Sponsorship, in which a Curator stakes his/her reputation on the fact that a team of authors has within it at least one individual with Curator qualifications. The authors themselves must then agree on who will become Curator: even the most generous senior expert should not be willing to award this privilege to a less-qualified individual, as this would mean that the senior expert would lose the ability to determine future changes to the article. It is precisely this person who will later on, as a Curator, have the most to lose if he or she vouches for someone untrustworthy. --Leo

Agree --ric

By giving Curator-like privileges to non-Curators, Scholarpedia would end up relying too much on Editors for quality assurance, and would weaken the privilege of Curatorship. As it stands, for legacy reasons there are already individuals in Scholarpedia that have Curator status but are not Curators of anything -- perhaps these might be called "Curators of Scholarpedia (or "Curators at large", but this to me sounds too political). Might these become the "Advisory Editors" you speak of? --Leo

I didn't propose to give "Curator-like privileges to non-Curators". I proposed to give the part of those powers that concern sponsorship of new authors (and other curator with no article privileges) to the members of a new class, call it sponsors if you do not like experts. In agreement with the power=>responsibility I proposed that authors --that had the responsibility of writing an article-- should belong to this category. (Note that it is not in the interest of authors to vouch for new non-expert authors enter in Scholarpedia). I instead agree that admitting advisory editors as being members of the sponsors class somewhat violates the "power=>responsibility", because they would have nothing to loose and we would only trust their allegiance. --ric

As for recognition for non-Curator authors: they already are recognized on the article page, and in every citation of it. They might also be called "Authors" of Scholarpedia, though creation of a formal category here is likely unnecessary. Article authorship has been incentive and recognition enough throughout history -- I strongly believe it will continue to be sufficient. --Leo

Also, I don't believe we should be in the business of anointing or deeming individuals to be "experts" -- authorship of one review article does not an expert make, and I see no reason why we should only allow experts to participate in authoring articles if the article's authorship is supervised by someone who (a) is a recognized expert, and (b) would look bad if they allowed anything incorrect in an article that they have signed-off on. Indeed, all the procedures that Scholarpedia has in place are to ensure that everyone who is a Curator is also an expert. This is the only way of permitting individuals who most certainly are not experts to act as authors. --Leo

As I (poorly) stated "expert" is just a tentative wording as member of the new category. As for the bottleneck to non-qualified authors you misunderstood my proposal: anybody can achieve the status of author if s/he manage to be sponsored and to convince two reviewers. Then in quality of author of a published article if s/he is not a curator s/he's promoted to the new class. --ric

The advantage of restricting Curatorship to one-per-article is that it forces people to reveal the information that is most vital to the creation of a trustworthy network while allowing amateurs to initiate new articles. Sure, it may make people uncomfortable, but this is a small price to pay for the information provided. It is far worse that many experts not be Curators than for many Curators to not be experts (and, furthermore, any expert who wants to be a Curator knows well what s/he needs to do!). --Leo

Never proposed to have more than one curator per article. --ric

We need Editors to get this project off the ground, but in general we must avoid having individuals with power that is not equally balanced by responsibility. In the long-run the authority of Scholarpedia cannot depend on Editors, as this would lead inevitably to arguments over academic "territory", to the brokerage of power, to stratification of topics, the development of petty fiefdoms, etc. This would be the result of Scholarpedia's having a class of individuals with privileges obtained only because of who-they-know. We must avoid at all costs the possibility that privileges will be awarded as a favor (or that they are not awarded because of a grudge). --Leo

I strongly disagree on the fact that in the long term Editors will not be needed. Scholarpedia will always need the role of editors as supervisors. Quality requires supervision. Having curators without supervision is much more prone to misuses, the first temptation is in inviting his curator pets as reviewers. On the other hand I think that editorial powers must be temporary, and that editors should be ideally elected by and from curators, to avoid the "who-they-know" problem.

I personally believe Editors can and should play a valuable role for as long as Scholarpedia exists. They can be vital to Scholarpedia's promotion; to its development of policy; to tracking and maintaining its overall quality at a high level; etc. That said, the point above was that, when it comes to Scholarpedia's being scholarly and authoritative, we should, if at all possible, minimize the need to depend on the contributions, authority, and good will of Editors.

As for non-registered reviewers (which is another problem) your proposal is more or less equivalent to allow non registered reviewers to approve as "User:Anonymous" and require a mandatory editorial endorsement for such articles (we should maybe avoid asking editors to approve twice, once as sponsor-reviewer and once because some reviewers are anonymous). Your proposal is ok but we first have to foreseen all situations an editor could face and make a more organic proposal to fix the issues "anonymous reviewers invited by editors" and "editor=sponsor=anonymous reviewer" (i.e. the paradox that an editor acts as a reviewer and wants to be anonymous but is obliged anyway to publicly sponsor because this is mandatory). --ric

When a Curator or Editor approves an article for publication, it does not necessarily mean that he or she alone is able to personally vouch for the article's quality. Rather, it just means that he or she has enough reason to believe that the article is ready for publication. In many cases this will be because he or she can personally vouch for the article's quality, but it might also be because someone he or she trusts has done so. Approval is deliberately meant to be slightly abstract. --Leo

My proposal was not for non-registered reviewers who want to remain anonymous, but rather for non-Curator reviewers who wish to not be anonymous. It might make sense for us to distinguish here between "review", which I'm going to define here to mean "engaging in a back-and-forth conversation with authors about their article", and publication "approval". Anonymity might be requested for one or both aspects, and I think that this is accommodate under the current system, but with the following proviso: more anonymity simply requires the participation of more non-anonymous parties to vouch for the anonymous ones. I think this is, actually, a good thing, since this will prevent unconstructive anonymous commentary. Since anyone can comment on an article non-anonymously, if an Editor wishes to do so s/he need only ask another Editor to review on his/her behalf. --Leo

• category "curator": experts that have also the curatorship of an article (in the current sense)
• category "editors": experts that have editorial powers in the current sense
• category "user" in the current sense
• category "contributor" in the current sense

(Actually I do not really like the name contributor, because it is misleading: in the usual language an author is a contributor. Alternatives: "maintainer" or "developer" or "collaborator" or "supporter")

Comment It's not my impression that the English word has any particular connotation in academia. I suppose, however, that it is often used in a reporting context to mean an individual who works on his/her own to produce news articles.

One reason why I liked the term is that it encompassed everyone who had a trust relationship with the article, and thus includes sponsors, reviewers, editors, etc. I believe the term is generic enough that it won't be mistaken, especially if the table is presented in the explicit way proposed. It is also somewhat flattering while having a somewhat fuzzy meaning, and is associated with the idea of a "donation" and other warm notions. --Leo

In all our rules on sponsorship "curator" must be replaced by "expert".

Ric: All in all, after reading Leo's arguments and thinking a little more, I will drop my proposal for the following reasons:

1. it is better to be minimalist in the categories of users
2. the issue of giving sponsorship power to non-curator authors, while leading to academic experts (excluded by the unicity of the article's curator) the power of sponsorship, it may also lead to (non academic expert) authors obtaining the sponsorship power. I agree that this latter event would be worse than loosing the action of academic experts due to lack of power.
3. I admit that giving sponsorship power to "advisory editors" based only our trust can lead to misuses."advisory editorship" must be a class with no power that must be cleanly distinguished from that of editors that work and take responsibility.

Editors

When we will implement the "2 curator sponsorship rule" for article sponsorship we will have to introduce also the exception "1 editor sponsorship=2 curator sponsorship", because editors are trusted people that are there to invite good experts, and we do not want to slow down their action

We might never implement this rule :) . Editors have plenty of opportunities to nix in-development articles that don't look good, and article review still will require two Curators. Everyone here is using their real name, and their reputations will always be on the line: unless we see a flood of bad single-Curator-sponsored articles from a large number of Curators, we should probably just keep it to a single Sponsor rule. If a couple of Curators misbehave, then they can be talked to or have their permissions changed.

If no "2 curator sponsorship rule" then no 1 editor sponsorship=2 curator sponsorship": ok.

Because of the fact that an editor will be a sponsor (possibly the only one as proposed above), he will be obliged to approve the article anyway. The ideal behavior would be that any editor sponsors an article and invites and registers 2 alternative reviewers (so that there will be 2 reviewer approvals followed by one sponsor approval). In reality there are 2 problematic situations. The first situation arrives when the reviewers want to keep their anonymity, the editor will approve on behalf of one reviewer, but who will approve on behalf of the second reviewer? The second situation is when the editor=sponsor is not a real expert on the subject and does not want to approve an article as a reviewer: who will approve the article in behalf of the sponsor? even worse the two situations might be combined. Way-outs:

• editorial approval (different from reviewer approval) of an editor bypasses the sponsor+curator rule and leaves anonymity to reviewers

Comment Perhaps we could allow Editors to formally "confirm" at most one of the reviews by a non-Curator? Another option might be to allow every reviewer to provide a 70 character comment, and the Editor might write "On behalf of Reviewer XX"

• refuse non anonymous reviewers

There are a few parts to this. (1) anonymous commentary and review, and (2) anonymous approval. (see discussion above)

Because of the fact that an editor will be a sponsor (possibly the only one as proposed above), he will be obliged to approve the article anyway. The ideal behavior would be that any editor sponsors an article and invites and registers 2 alternative reviewers (so that there will be 2 reviewer approvals followed by one sponsor approval). In reality there are 2 problematic situations. The first situation arrives when the reviewers want to keep their anonymity, the editor will approve on behalf of one reviewer, but who will approve on behalf of the second reviewer? The second situation is when the editor=sponsor is not a real expert on the subject and does not want to approve an article as a reviewer: who will approve the article in behalf of the sponsor? even worse the two situations might be combined. Way-outs:

• editorial approval (different from reviewer approval) of an editor bypasses the sponsor+curator rule and leaves anonymity to reviewers
• refuse non anonymous reviewers

Now that editors are only morally linked to a category, would it be possible/advisable to enlarge the editorial board for each category to 2-3 editors (with the rule that each editor should be approved by the previous ones)

L:Yes, I think this would be good. Perhaps category editors should have to add the text: [[Category:whatever]] to their page?

R:Well it would be simpler to manually add their name to a category.

Fair enough ;)

Change contribution measurements

This will result in problems of comparing two assistant curators contributors, and to have a Curator Scholarpedia Index that reflects overall contribution to Scholarpedia. The system with a single number is better. -- Eugene

This suggestion just concerns the way the contributions were presented. It does not prevent the consolidated from being used outside of the article. Furthermore, it's so straightforward that it's unlikely anyone would protest it, and people are already protesting the current mechanism. To permit better within article comparisons, we could do a secondary sort based on A/(A+D) or (A-D)/T. --Leo

Only count post-publication edits in contribution measurements

IF the scholarpedia/activity index is there to rank new contributors ok, see elsewere my proposal. ric