User talk:Anne Delong/AfcBox

Voting for approval of this proposal

Per a request at WP:AN/RFC, this RfC is closed with near unanimous support for the proposal. -Nathan Johnson (talk) 15:10, 27 May 2013 (UTC)[reply]

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

The proposal is posted at User:Anne Delong/AfcBox. Please indicate your supporting/opposing views and/or concerns below-

This is a proposal to insert a new section (either a web page or a pop-up box) into the Afc submission procedure to encourage new Afc acticle submitters to add reliable sources. It would reduce the number of articles that would be rejected immediately for being unsourced. I apologize for the poor graphics. As you can see, the proposed procedure would NOT check to see if there were references (pretty tricky to implement without error), but the editor would have to be totally confused or actively lie or to get the article submitted without any. Users who did this repeatedly could easily be picked out for special attention. TheOriginalSoni (talk) 21:47, 29 April 2013 (UTC) I have added another choice (D) because some were worried that new editors would give up too soon. —Anne Delong (talk) 00:05, 3 May 2013 (UTC)[reply]

Adding an RfC tag to generate wider community views. TheOriginalSoni (talk) 01:57, 28 April 2013 (UTC)[reply]

Support: FoCuSandLeArN (talk) 13:03, 15 April 2013 (UTC)[reply]
Support, provided a better way of achieving the same goal isn't done first. davidwr/(talk)/(contribs)/(e-mail) 00:31, 28 April 2013 (UTC)[reply]
Certainly - This will bring about a streamlining of the AfC process, and remove a lot of bad AfCs we see today. Will also relieve the help desk from quite a few questions. TheOriginalSoni (talk) 01:15, 28 April 2013 (UTC)[reply]
Support per Davidwr. • • • Peter (Southwood) ^(talk): 07:30, 28 April 2013 (UTC)[reply]
Support Looks good.. TheStrike Σagle 12:27, 28 April 2013 (UTC)[reply]
Support Implementation if it is technically non-taxing to the system. Good job! GenQuest ^{"Talk to Me"} 19:09, 28 April 2013 (UTC)[reply]
Support but include an exception under option c for articles that don't necessarily require independent sources, i.e. schools, long-standing newspapers (seems to be the consensus anyway), and probably a few others I haven't thought of --BigPimpinBrah (talk) 02:43, 29 April 2013 (UTC)[reply]
Support with inclusion of good suggestions below. Roger (Dodger67) (talk) 07:55, 29 April 2013 (UTC)[reply]
Support Looks good to me! -Fumitol|talk|cont 13:26, 29 April 2013 (UTC)[reply]
Jumping up and down support Fundamentally our biggest problem with article creation, here and elsewhere, is that we don't actually give editors the information they need to write an article that will survive before they write it. This is the very first attempt I've seen to correct that. Well done! --j⚛e decker ^talk 15:07, 29 April 2013 (UTC)[reply]
Support I've said it elsewhere, but what the heck... Pol430 talk to me 15:11, 29 April 2013 (UTC)[reply]

Comment I like the idea but I don't think it is realistic. Newbies who are not adding sources have one of several issues and this check is not addressing these. I think that this point has been missed by a number of voices here. I believe that ony 2% have simply forgotten about putting in citations but in most cases the newbies simply lack the skills required to provide (inline) citations. Sending them to a policy pages is not going to fix this either - I am actually worried that this will just backfire and reduce new article creation success rates.

So if this extra step is added I suggest to set it up as A/B engagement test first and get a measurement on the results. If testing proves this as an efficient measure I will support it. BO | Talk 12:48, 30 April 2013 (UTC)[reply]

(I added two new questions below to expand on this.) —Anne Delong (talk) 14:48, 30 April 2013 (UTC)[reply]

Support with modifications Strongly discouraged, yes! There are enough exceptions that the author should be able to submit the article unreferenced, though strongly discouraged. A person could say, I'm putting in what I have in the hope others will add to it--which is why "unreferenced' is not cause for deletion. That's not a good way to work, but we've gotten many good articles that way 'DGG (at NYPL) (talk) 20:07, 2 May 2013 (UTC)[reply]

How about adding unreferences articles is allowed only when an article goes into AfC, and not the main space? TheOriginalSoni (talk) 21:16, 2 May 2013 (UTC)[reply]

Since this is an addition to the Afc submission process, it doesn't affect mainspace at all. The point of the proposal is to prevent all of the unreferenced articles in the Afc. If you think that it's okay for Afc articles to be submitted without references, you should vote against this proposal, since that's what is happening now. —Anne Delong (talk) 23:36, 2 May 2013 (UTC)[reply]

Ah. I confused AfC and article creation wizard. Carry on. TheOriginalSoni (talk) 23:40, 2 May 2013 (UTC)[reply]

I do not think it is a good idea, but I think it remains permissible. We should add with a warning. ~

Oppose with appreciation for the author and the concept, as I think that this will discourage people from not only making requests for non-notable articles, but from making requests for articles in general. I understand the need for reliable sources - I really do - but if a user is using AfC in the first place, I'm going to go ahead and make the assumption that they aren't going to be really knowledgeable on reliable sources. Wikipedia editors are more than competent at using search engines, in general, and those who participate in the AfC projects certainly know policy. I understand that it can be time-consuming and frustrating to have AfCs listed that are non-notable, but at what expense do we limit public requests for articles? Let my opinion be taken with a grain of salt, though, and feel free to push me off my soapbox, as I do not help to create articles, I merely nominate them for deletion and revert vandalism. --Jackson Peebles (talk) 13:42, 16 May 2013 (UTC)[reply]
Support Anything which helps new editors submit articles in a state in which they are more likely to be kept is a huge bonus for new editor engagement and retention. Anything which improves the AFC process and helps our AFC volunteers is to be embraced. I'm sure it won't be perfect on day 1, and I'm sure we will improve the process quickly in use. None of that will happen if we don't take the first step. Thanks for the well targetted idea, and the thought that has obviously gone into it. Begoon ^talk 12:44, 18 May 2013 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

General discussion and questions

I added the fact that there are also templates. ;-) mabdul 07:17, 15 April 2013 (UTC)[reply]

Thanks, mabdul, good catch. —Anne Delong (talk) 14:46, 15 April 2013 (UTC)[reply]

Q: Won't users just ignore it as they have done to suggestions in the template?

(as suggested by Someguy1221)

A:

Yes, some will; it won't always work, but rather than just ignoring something, the user would have to actively LIE to submit a sourceless article. It would help separate those who are just inexperienced (we can teach them) from those who are intentionally noncompliant (administrators could take action).—Anne Delong (talk) 14:06, 15 April 2013 (UTC)[reply]

Comment actively LIE - please try to assume good faith. AfC is here to serve newbies not to make liars out of them or to ban them for noncompliance. :( BO | Talk 12:55, 30 April 2013 (UTC)[reply]

I agree that the wording was not the best chosen, but what she meant was that We would know when a user clicks on yes without actually reading all the statements. It might not help all the users, but it will give the users an idea of what's absolutely essential for our articles - references [As a helper at the help desk IRC, I can verify that there are plenty of such users.] TheOriginalSoni (talk) 13:10, 30 April 2013 (UTC)[reply]

Q: Will Javascript be needed?

(asked by Technical 13)

A:

I don't know enough about how the original submission template is added. Does it require Javascript? A web page needs to be generated with six links, four of which include info as to what page the user was trying to submit, so that if they choose options A or F the user's article page is displayed again with no submission template added, and for options D or E the link would be to the submission process that now exists. Couldn't that be done server side? —Anne Delong (talk) 14:06, 15 April 2013 (UTC)[reply]

It could be done server-side, however, that would require writing a PHP extension and getting Wikipedia to adopt it, which due to the specific nature of it I doubt would happen.

JavaScript is the better alternative (and that is what I think the request for a new extension would turn into, "do it with JavaScript or a bot"). In order for anything to "popup" and require input from the user, some kind of scripting is required. This would mean that the user would have to import the script to their Special:MyPage/skin.js at very least UNLESS we wrote it as a gadget and got Wikipedia to implement it as a default. Then it would be "turned" on by default for everyone.

The alternative to this (which I think we should probably consider if it is feasible first), is to add a new section to the submission template of something like "Request for submission of article" of which a bot could go through and check for citations, possible links to the article, links from the article, basic formatting errors, and so-on. Then, the bot could either decline the article and offer helpful links for the editor to make adjustments to the article or make necessary changes to the article and change the status to "Articles waiting for review". Any articles that are not clean cut denials, the bot should do what it can and then change the status to "Articles waiting for review". Technical 13 (talk) 15:13, 15 April 2013 (UTC)[reply]

Would there be any need for a "pop up"? Couldn't the submit button lead to a link to a regular web page with links, and the user clicks on one of the links? —Anne Delong (talk) 19:50, 15 April 2013 (UTC)[reply]

@Anne Delong – My thoughts exactly. – Ypnypn (talk) 23:35, 16 April 2013 (UTC)[reply]

While it technically "could" be done with templates and multiple page reloads, I feel that method would be really non-user-friendly and would deter people from using AfC. Also, using templates and page reloads would leave you unable to know the answer to "Did the user perform some other action such as leaving the page?" Technical 13 (talk) sometime, 27 April 2013 (UTC)

Good point. —Anne Delong (talk) 01:41, 28 April 2013 (UTC)[reply]

Q: Will a bot check the article for references to see if there really are any?

(asked by mabdul and Technical 13)

A:

My proposal does not suggest that. It would be very complex - and error prone, too. Maybe someday... —Anne Delong (talk)

I'm suggesting it probably should do that. It would likely be less complex than you think and any errors would default to the article being submitted for human review. Technical 13 (talk) 15:13, 15 April 2013 (UTC)[reply]

It would be nice if this could be done, but I have been reviewing a lot of submissions which appear NOT to have to references but really do, such as those were the ref tags are left out or messed up, those where the user forgot the http://, and those with only offline resources who listed them but didn't create citations. I believe that your idea, while a good one, would be trickier to implement. —Anne Delong (talk) 19:50, 15 April 2013 (UTC)[reply]

Perhaps it could be checked for ref tags and if not present, send a warning to the user that refs are not apparent or are incorrectly configured, which would also be useful feedback. • • • Peter (Southwood) ^(talk): 07:33, 28 April 2013 (UTC)[reply]

There is no requirement that ref tags be used for referencing--it is just one of the many possible methods. It's the one most commonly used here, and the one we recommend, but if someone uses parenthetical references properly, there is no reason for them to be converted, and no reason for there to be any warning. There are also informal ways of expressing references, as in . John smith, according to his obituary om p.100 of the February 30, 2000 New York Times, was .... -- which not only is an adequate reference but proves notability.

Q: Won't users just add unreliable sources such as Facebook and Twitter?

(asked by Ritchie333)

A:

For sure, some will (okay, many...). But those who do are one step ahead of the ones who don't know what a reference is (or submit their mothers as references). One step at a time!

Per the above comments of mine, these could be detected, the article declined if it relies solely on these references, and a helpful link to what reliable sources are. Technical 13 (talk) 15:13, 15 April 2013 (UTC)[reply]

I do think this would happen. I think you could probably add a sentence or so to the script that would help. " Have you included citations to independent, reliable sources in your article? (Generally books, newspaper or magazine articles from reputable publishers, rather than social media pages like Twitter or other self-published sources, see identifying reliable sources for more detail.)" This might be too much, but I think we can hint to well-meaning editors what we actually mean by reliable sources to good effect. --j⚛e decker ^talk 15:12, 29 April 2013 (UTC)[reply]

As pointed out below, independent sources are not strictly necessary for some categories of articles, such as those covered by WP:PROF or elsewhere where the alternatives to the GNG apply. 'DGG (at NYPL) (talk) 21:47, 6 May 2013 (UTC)[reply]

Indeed. But I'm not sure I'd take a twitter account as a source for the claim of a Richard Feynman Professor of Physics, or what have you. A university bio? No question, I pointed out PROF C5 somewhere below myself. --j⚛e decker ^talk 02:43, 7 May 2013 (UTC)[reply]

Q: Can we do it with categories?

Q: So, what if we put all of the ones that claim to have references into a category, and/or have a kind of "intermediate" template, that reviewers can go through real quick and say "yes has references" or "no does not, warn user about lying", without checking credibility, or reviewing the actual article? That would make it easier to find these liars before having to do a full review of an article. gwickwire^talk_editing 01:01, 18 April 2013 (UTC)[reply]

A: This might be a good idea, although I think right now there is a bot that removes categories in Afc articles, so that might be a complication. I am beginning to wish I hadn't used the word "lie"; perhaps we could suggest that they misunderstood and direct them to Referencing for Beginners. Hmmm... if all of the submissions went through the filter page being suggested, then every one of them would be claiming to have references, so there may be not need for a category. In any case, this is something that would only work as an addition to the original proposal, so if this means you like the proposal please indicate so at the top of the page. Thanks! —Anne Delong (talk) 00:26, 28 April 2013 (UTC)[reply]

A: One more thing - although in my proposal I said "citations to reliable sources", if the submitters had just added some references at the bottom of the page and hadn't yet learned to insert them as citations, I think it would still be an improvement and we wouldn't chide the submitters for not knowing the difference. —Anne Delong (talk) 03:43, 28 April 2013 (UTC)[reply]

Agreed, except it is not a matter of teaching them to do better, it is not necessary to teach them at all if they express the references in any format whatever as long as they can be identified.

There's even a MOS policy, that we do not change references from one form to another as long as they are consistent. My personal opinion, as someone who for long had the job of teaching students to use references in whatever arbitrary style a professor or a journal chose to require, is that we put much too great an emphasis of reference format, as distinct from reference quality. The advice I and others give to beginners is to put in references however you please, as long as you put them in. reference format is one of the unnecessary hurdles to editing. 'DGG (at NYPL) (talk) 21:47, 6 May 2013 (UTC)[reply]

Q. Are there some articles that don't need sources?

A: It's my understanding that although schools, newspapers, geographical features, etc., do not necessarily need references to pass the notability test (as pointed out above by BigPimpinBrah), they still need references to back up facts that are in the article and to show that they actually exist. Perhaps someone who knows could comment on this. The proposal already exempts disambiguation pages, categories, and lists. —Anne Delong (talk) 03:20, 29 April 2013 (UTC)[reply]

As far as my understanding of notability goes for these topics, geographical features (like towns and cities) do not need any source to be added, as long as they are not hoaxes. Many articles already exist without any sources, as finding a reliable one can be difficult for most of those which will be added through AfC. Schools have a similar status I think, except it needs to be proven that they exist. [No idea about newspapers though]. I think that we should be exempting any of these three categories from having sources, by adding another question. If the user answers no to the reliable sources question, we ask if its about a school, town/village or newspaper, and direct them to the next step if they answer yes. TheOriginalSoni (talk) 04:23, 29 April 2013 (UTC)[reply]

Personally, I don't support the promotion of unsourced articles. I've seen hoax village articles. You're right, however, that high schools are generally considered to be notable by default, as well as populated places. I'm not sure I recall that newspapers do or don't, but I'll take your word on it. I also don't think that "go find a source" is an unreasonable request, if the author can't find a source for their material, I have to ask, what did they write the article from? --j⚛e decker ^talk 15:21, 29 April 2013 (UTC)[reply]

Please note I did specifically say independent sources, since the question in the box is "Wait! Have you included citations to independent, reliable sources in your article?". --BigPimpinBrah (talk) 15:37, 29 April 2013 (UTC)[reply]

Ahh, indeed. Another example of a case where we generally don't force an independent source--generally we'll take a university biography of a prof holding a named chair as meeting WP:ACADEMIC #5 (assuming the institution looks correct for the critieria, etc.) --j⚛e decker ^talk 01:09, 5 May 2013 (UTC)[reply]

Even if notability is assumed, as in the case of professors, I believe that a reliable source is still needed to prove that the subject is legitimate, and not a hoax. —Anne Delong (talk) 02:12, 5 May 2013 (UTC)[reply]

they do need some sort of source to show existence, but it just has to be reliable enough for the purpose. It does not have to be independent if it is in some manner official or otherwise assumed to be genuine. . But in fact for such basic things as geography we normally simply add a source; which can be as little as a link to the community's web page. That would not be independent and reliable enough to meet the GNG, but it is when only existence must be proven. Don't confuse the technical requirements of the GNG with Notability.'DGG (at NYPL) (talk) 21:47, 6 May 2013 (UTC)[reply]

Ayup. --j⚛e decker ^talk 02:57, 7 May 2013 (UTC)[reply]

Q. What if new users are confused and discouraged by reading policies?

(as suggested above by BO)

A: BO has made a couple of good points in his comment above. "Referencing for beginners" is not a policy page and should be useful to new users, so I'm fairly confident in this option, but perhaps in the case of Option C - "I've looked and I can't find any", sending users off to WP:Notability is not the proper response. There are two very different reasons why users can't find sources when they have looked for them (1) there aren't any and (b) the user doesn't know where to look. What if this were split into two choices :

"There aren't independent sources. I'm promoting a subject that is new and/or as yet unknown." -> WP:Notability or something less daunting if someone knows of such a page, and
"I've looked for sources, but I am a new editor and would like suggestions as to where to find them." -> Tea House or New Contributor's Help Page (I hear they may be combined). —Anne Delong (talk) 14:30, 30 April 2013 (UTC)[reply]

The first should discourage a lot of the spammy things and fringe guys, and maybe that's okay.

What if they select *"I've looked for sources, but I am a new editor and would like suggestions as to where to find them." we offer them {{Find sources}} or something similar before sending them to the Teahouse (with an optional link to go there anyways if they want). FYI, yes, there is a merge discussion going on and there is overwhelming support for this merge at this time. Technical 13 (talk) 15:41, 30 April 2013 (UTC)[reply]

Q. How can we tell if this is discouraging editors from creating articles at all?

(as suggested above by BO)

A: BO is right that there should be a process in place to tell if the process is working as intended. Would it be difficult to add a line of code that would add a Category:Potential Afc submissions not submitted, and add that to the user's article if they got that far and chose not to submit? Interested reviewers could browse the category and see if there are potentially acceptable articles being missed; I would be willing to do this for a few weeks and make a report, unless you think it would be better for someone else to do it. —Anne Delong (talk) 14:30, 30 April 2013 (UTC)[reply]

No, it would not be difficult to transclude such a category in the last stage of template before submission, although I still think trying to do it with templates and forced page reloads is ugly, messy, and not user friendly which will deter people from using the AfC. I still think a JavaScript gadget that will do all of these things would be the better option and could even offer the editor an additional sidebar "AfC toolbar" offering them links to sources and ways to test their own project (such as doing a copyvio scan that tells them "The highlighted words/phrases are too close to the words on {these sources}, please try and reword them in your own words. Thank you"). I think something like that would actually encourage people to use the service just for the extra tools at their disposal. Technical 13 (talk) 15:48, 30 April 2013 (UTC)[reply]

Whether the tech people end up implementing this with page loads or with one kind of software or another, I still think that the category would be useful as a means of checking to see what effect it is having on the submitters. —Anne Delong (talk) 21:24, 30 April 2013 (UTC)[reply]

As do I. This can be done as easily either way in my opinion. Technical 13 (talk) 21:46, 30 April 2013 (UTC)[reply]

I think it would also be useful to track total accepted submissions per time period. If this discourages (I doubt it will) "Fred is my best friend in second grade" biographies, that's not a problem. If discourages articles with serious potential that's a real problem. If it saves enough effort that those working AfC submissions to help more articles up to par and through the gate, that's a big win. I can imagine that all three of those will be factors. To my mind, a primary metric for "winning" is "more articles mainspaced, assuming equivalent quality." Figuring out that we want to track some information on this is an excellent suggestion, Anne. --j⚛e decker ^talk 02:55, 7 May 2013 (UTC)[reply]

Implementation

Dear Pterb, mabdul and Writ Keeper:

I think that the talk has died down on this proposal, and the votes are mostly positive. I have made some adjustments to satisfy some concerns, hopefully, and offered to keep an eye on the category added to the articles of those who don't choose to submit so that we can tell if it's working or just scaring off potential editors.

Technical 13 offered to do some scripting on this, but on reflection remembered that it is his exam week!

I wonder if one of you might take it on. Peterb wrote a couple of weeks ago that he might help once we sorted it out. mabdul has shown some interest and made suggestions. Technical 13 suggested Writ Keeper.

I myself have never done anything like this, and although I can program in C++ and Java, I don't think these skills will help me here. —Anne Delong (talk) 00:50, 5 May 2013 (UTC)[reply]

Hi, thank you for your work on this. I will now read the whole discussion and once it is clear to me what exactly the bot is supposed to do, we can go on with brfa and implement this. Petrb (talk) 07:06, 5 May 2013 (UTC)[reply]

So what you need from me is: The bot should check if a newly submitted article contains a reference, and if not, it should do something? This is very simple task. Can you explain to me 1) when the bot should check it (I suppose for every new article?) 2) what action should it take in case there are no references. Petrb (talk) 07:09, 5 May 2013 (UTC)[reply]

Thanks for responding, Peterb, but no, that is not what is wanted here. Please look at the proposal itself, at User:Anne Delong/AfcBox. What is needed is pop-up box (or web page, whichever does the job must efficiently) which asks the submitter to make one of several choices, and then does the appropriate action as listed on the flow chart. We are trying to get the submitters to pay more attention to adding references, and to put them in before they submit. —Anne Delong (talk) 09:09, 5 May 2013 (UTC)[reply]

there is approval for this in general. There is not agreement on the specifics. I see some degree of consensus for something like "this is theb est I can do, or all that is appropriate." . Better have in in afc than added in desperation to mainspace becauser they can;t figureo ut the right answer. 'DGG (at NYPL) (talk) 21:47, 6 May 2013 (UTC)[reply]

Resuscitate this for the new Draft-space processes?

Hi Anne, I think the ideas here would be a useful contribution to the discussion about Draft-space procedures. Roger (Dodger67) (talk) 09:37, 12 February 2014 (UTC)[reply]