Wikipedia talk:Invitation to edit/Archive 2
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 | Archive 2 |
Stats
Anthony has manually evaluated the contributions to Pain (to date) while the template has been on the article, compared to the matching calendar dates from a year before. The list is at User:Anthonyhcole/ITE_Stats. We show a drop in new editors, but a tiny (statistically insignificant?) increase in good-faith edits.
It might be interesting to compare some of the article statistics during that time, and page views. Page views for this article seem to compare like this:
Month | This year | Last year |
---|---|---|
October | 89408 | 67483 |
November | 85984 | 66087 |
December | 53386 | 57314 |
January | 84577 | 67892 |
The last data point (January of last year) may be artificially low by about 7%. Two days reported zero page views, which seems unlikely.
So overall, it looks like we have a trend of more page views and fewer edits. I wonder whether this is a general trend. It might be, since better-written pages tend to attract fewer edits, and the overall trend in Wikipedia articles is improvement. WhatamIdoing (talk) 17:31, 2 February 2011 (UTC)
- With only 20 editions in 2 months by ip editors I would say that it is going to be hard to demonstrate anything, probably not even a trend. In addition the measure "good editions" is quite subjetive. Can we know the number of visits by ip editors? We could measure the ratio of ip editions of number of ip visits. In addition to being more robust it could serve as to compare different months with different number of visits since it is easy to assume that percentage of new editors out of total ip viewers would not depend of different (but consecutive) months; while the number of editions or behavior of editors can change from one year to the next. --Garrondo (talk) 20:11, 2 February 2011 (UTC)
- Pain is a mature, stable B-class article with 2,500 readers per day. The trial articles are all of poorer quality and higher popularity so, hopefully, there'll be more to measure there.
- I proposed comparing the new editors in the trial period with new editors in the same period a year earlier to avoid confounds like school holidays and weather. When analysing the trial articles, we'll also be comparing the templated articles with a number of similar control articles in those same periods that didn't have the template on. So group differences between the trial articles and control articles should be informative.
- Wikipedia_talk:Invitation_to_edit/Archive_1#Study_Design summarises the initial vision of the study design, but improvement by experts is appreciated. --Anthonyhcole (talk) 03:01, 3 February 2011 (UTC)
- It might be useful to do multiple kinds of comparisons. Weather and holidays and celebrity news can all affect readership and editing, as well as Wikipedia growing in popularity over time. WhatamIdoing (talk) 21:37, 3 February 2011 (UTC)
- A control group will deal with these sorts of fluctuations, assuming that they are pretty much the same across the board. Wrt to increasing popularity, percentages are better metrics than raw numbers because they can be compared over time. (ie, it tells you something if 60% of readers in 2002 edited versus 6% now, whereas saying that 100 people edited in 2002 and 100000 edited now may just mean that the website is more popular) --Danger (talk) 11:32, 7 February 2011 (UTC)
- If you don't already have something like this, I'd like to develop a coding scheme to quantify the quality of an edit. Then we can figure out whether the average quality of the edits is different between a control group and a test group. (I hypothesize that this won't have an impact on bad faith edits but will improve the quality of good faith ones.) --Danger (talk) 11:43, 7 February 2011 (UTC)
- That sounds like a good plan, Danger. I agree with your point on percentages. That was where we were heading. And the control group will be a number of articles with similar low vandalism, high readership figures. They're not chosen yet, but can be chosen at any time, when I can make the time to record the stats. Real life's being a bit demanding just now. --Anthonyhcole (talk) 14:31, 7 February 2011 (UTC)
- If you don't already have something like this, I'd like to develop a coding scheme to quantify the quality of an edit. Then we can figure out whether the average quality of the edits is different between a control group and a test group. (I hypothesize that this won't have an impact on bad faith edits but will improve the quality of good faith ones.) --Danger (talk) 11:43, 7 February 2011 (UTC)
Wiki Guides study/project
Hi Everyone!
User:WhatamIdoing left a message on the Wiki Guides talk page for here and I agreed, this looks like a great trial to think about. I wanted to let everyone here know about the study that he posted on. Obviously given the project you're aware that we've had a decrease in new editors over the past couple years.
![](http://upload.wikimedia.org/wikipedia/commons/thumb/1/13/ENnewEditors.jpg/230px-ENnewEditors.jpg)
As a community we have a lot of ideas but we’ve been stymied by a lot of options and little data.
We want to conduct a study over the next couple months (with some resources from the Wikimedia Foundation) to help craft strategies to develop new users, to get data on exactly how our new users are finding their first, and later, experiences on Wikipedia and of course to help share the experiences of the experienced users who are here to find out what works, what doesn't and what resources they need to make their work easier.
The plan at the moment is to have several groups of users, 1 group that is just followed (the control) and several other groups with guides who actively reach out and try to help them edit and join the community. I hope that you can help us as we get ready for the study start next month and help the new users once we start! You can find out more information and sign up on the project page and if you can think of anyone who might be interested please please PLEASE point them this way or let me know so I can reach out to them personally!Jalexander--WMF 00:32, 4 February 2011 (UTC)
Concision
James posted this at Template_talk:Invitation_to_edit/tutorial
How to edit a medical article:
- Find a source or reference. A list of suitable sources can be found under medical resources for editors
- Properly format said source. A number of tool can help including diberri's.
- Place the formatted reference between <ref>..</ref>
- Summarize the content, find the appropriate heading per the medical manual of style or create it if not present.
- Click the edit button and add your content followed by the reference.
- If you have questions about editing medical content drop us a note at WT:MEDICINE
Congratulations on joining Wikipedia!
I'd like this tutorial to be that simple and clear but feel the information being imparted in the current version is essential. Garrondo is making it more concise while retaining what I believe to be the essentials. Can that be improved upon? And should we cut some of the message back? Should we say less? If so, what? --Anthonyhcole (talk) 13:48, 6 January 2011 (UTC)
- I also think that most of the info is important. However I would rather eliminate the line on medical resources for editors. This subpage is from my point of view very general and may confuse or overwhelm more than help a new editor: Pointing an editor towards google books and pubmed in general is as much as nothing.--Garrondo (talk) 09:30, 8 January 2011 (UTC)
- OK with me. Done --Anthonyhcole (talk) 10:50, 8 January 2011 (UTC)
- Instead of having the title and at the end of the paragraph a link to the full policy what we could do is to directly link the policy in the title. I have been bold and carried it out. If somebody disagrees feel free to revert.--Garrondo (talk) 15:48, 9 January 2011 (UTC)
- I agree with your comment about linking to Google, etc. I'm waiting for this proposal currently being tossed around at here. Does anybody know what's happening with the proposal to put RefTools above the edit box for all users by default? --Anthonyhcole (talk) 14:40, 14 February 2011 (UTC)
- OK with me. Done --Anthonyhcole (talk) 10:50, 8 January 2011 (UTC)
What is a talk page anyway?
I like this idea of encouraging the passerby-reader with some nugget of info to offer. Clicking to unhide the template, the first line is "The easiest way to edit is to go to the article's discussion ("talk") page". If an individual has never come across the concept of a talk page before, this is a mental leap in and of itself. ("You mean, Wikipedia has pages that aren't encyclopedia articles? Pages that talk to you? What's that about?") To make things easier, I suggest that the text should point out the location of the button ("at the top of the page on the left"). BrainyBabe (talk) 20:27, 10 February 2011 (UTC)
- We're trying to keep this tutorial as concise as possible. I've stretched the blue link from just discussion to discussion ("talk") page so that, hopefully, it is clearer that it links to the talk page. --Anthonyhcole (talk) 19:13, 12 February 2011 (UTC)
Just increase the size of the edit button
I have always felt that the edit button should be the most prominent button in the upper bar. I think making the button slightly bigger and with a darker background would make it more obvious for people that you can actually edit this page. GoEThe (talk) 14:48, 3 February 2011 (UTC)
- I like that. I'm pretty keen to have a mini-intro-tutorial obviously and easily accessible. The ideas may not be mutually exclusive. --Anthonyhcole (talk) 14:57, 3 February 2011 (UTC)
- Yes agree that this is a good idea. Doc James (talk · contribs · email) 01:32, 4 February 2011 (UTC)
- Would it be possible enable the mini tutorial template to appear only after users have pressed the edit button? Like sometimes if a page is protected a message appears above the edit field, perhaps something similar could be done with the tutorial so then it doesn't interefere with the look of article mainspace? Hope that makes sense. Polyamorph (talk) 08:50, 4 February 2011 (UTC)
- Yes, that's called an editnotice and can be implemented just be editing Template:Editnotices/Page/Pain. — Martin (MSGJ · talk) 10:46, 4 February 2011 (UTC)
- Yes another great idea. Doc James (talk · contribs · email) 05:23, 5 February 2011 (UTC)
- This is interesting. --Anthonyhcole (talk) 02:05, 10 February 2011 (UTC)
- If the trial in anyway returns positive effect with respect to increasing the quality of good faith edits then I think this might be a way forward to eliminate the concerns of users who think a template in article mainspace is too intrusive? Polyamorph (talk) 21:06, 10 February 2011 (UTC)
- The focus needs to be on improvements, not just edits. If we can hook readers into wanting to fix things, where the edit button is will be the least of our problems. I listed some ideas in the bottom section. —UncleDouggie (talk) 08:43, 16 February 2011 (UTC)
- If the trial in anyway returns positive effect with respect to increasing the quality of good faith edits then I think this might be a way forward to eliminate the concerns of users who think a template in article mainspace is too intrusive? Polyamorph (talk) 21:06, 10 February 2011 (UTC)
- This is interesting. --Anthonyhcole (talk) 02:05, 10 February 2011 (UTC)
- Yes another great idea. Doc James (talk · contribs · email) 05:23, 5 February 2011 (UTC)
- Yes, that's called an editnotice and can be implemented just be editing Template:Editnotices/Page/Pain. — Martin (MSGJ · talk) 10:46, 4 February 2011 (UTC)
- Would it be possible enable the mini tutorial template to appear only after users have pressed the edit button? Like sometimes if a page is protected a message appears above the edit field, perhaps something similar could be done with the tutorial so then it doesn't interefere with the look of article mainspace? Hope that makes sense. Polyamorph (talk) 08:50, 4 February 2011 (UTC)
- Yes agree that this is a good idea. Doc James (talk · contribs · email) 01:32, 4 February 2011 (UTC)
Could someone who has the tools please paste the tutorial into an editnotice for this page and/or the proposal page? It requires admin privileges per this --Anthonyhcole (talk) 10:29, 21 February 2011 (UTC)
- Sure I could, but I can't see the point of that. The tutorial is not aimed at people editing this page ... What might make sense (if the trial was continuing) would be to add it to the editnotice of the articles involved. — Martin (MSGJ · talk) 12:24, 21 February 2011 (UTC)
- I was thinking it would show collaborators here how the tutorial would look as an editnotice. I tried it on my User page and it doesn't look too bad. --Anthonyhcole (talk) 12:45, 21 February 2011 (UTC)
- Understood and done. Too long in my opinion. Perhaps a shorter tutorial along with a collapsed box for the rest would be less intrusive. — Martin (MSGJ · talk) 12:49, 21 February 2011 (UTC)
- I agree. Once RefTool is on everybody's edit box (User:Kaldari was hoping to deploy it today) that will enable us to slash quite a bit of the "how to cite" stuff. Actually, regarding that, I'd like RefTool to be even simpler. I'd like to only have to type a number into the doi, ISBN or PMID field, and the page number(s) into another. But I'll propose that once the site-wide implementation has stabilised. --Anthonyhcole (talk) 13:18, 21 February 2011 (UTC)
- Understood and done. Too long in my opinion. Perhaps a shorter tutorial along with a collapsed box for the rest would be less intrusive. — Martin (MSGJ · talk) 12:49, 21 February 2011 (UTC)
- I was thinking it would show collaborators here how the tutorial would look as an editnotice. I tried it on my User page and it doesn't look too bad. --Anthonyhcole (talk) 12:45, 21 February 2011 (UTC)
Usefulness of the trial and possible extension
It was mentioned in the Tfd that some people would like to extend the current trial to 60 days. I strongly recommend not extending it in the current form. There should be an RfC to determine the best UI before a trial because the UI dramatically affects the results. When I went to pain, I saw a glaring message that "Wow, this article is open for editing". Of course, it would have been a surprise to me if it wasn't. However, to a regular reader, I think the message could easily be interpreted as "this is the one and only page on Wikipedia that you can edit", and that mentality will dramatically affect the trial. After all, the other pages don't have that neon green thing on them. Why should I read that whole blob of green and deal with a weird editing interface just so I can add a comma to this page? The goal of attracting new editors is admirable, but we need to take a bigger view of the problem than just a single article. Thinking through these issues before a trial is critical to it's meaningfulness. —UncleDouggie (talk) 09:33, 14 February 2011 (UTC)
- I don't think I've understood your point. Are you concerned that the invitation on the trial articles will mislead readers into thinking they can't edit other articles?
- Consensus for this trial was first obtained at Village pump (proposals). I'll be going back there tomorrow to ask permission to extend the trial. --Anthonyhcole (talk) 10:01, 14 February 2011 (UTC)
- One of the earlier versions of the tutorial began with something like "Unlike corrections of spelling, expression, etc, changes of meaning to a Wikipedia medical article must be supported by an appropriate scholarly article or textbook." That's a paraphrase. Should we reinsert something like that so it doesn't scare off simple fixers?
- Regarding the complexity of the tutorial, shortly, per this discussion, RefTools, a simplified automated citation button, will appear by default above all users' edit boxes. Until now, you had to opt in for it. Though RefTools could/will be made even simpler, this implementation will vastly simplify the citation instructions in the tutorial. --Anthonyhcole (talk) 09:58, 16 February 2011 (UTC)
- A little tweak might be "You can edit this article" -> "You can edit Wikipedia, including this article". Rd232 talk 13:21, 14 February 2011 (UTC)
- I like that very very much. --Anthonyhcole (talk) 14:14, 14 February 2011 (UTC)
- This solves the concern I expressed above, but I still don't think that it's the best we can do. It's a lot of words and can be seen as inviting vandalism. We need a short message that somehow conveys these thoughts in an inviting way without asking people to just do an edit test:
- Wikipedia needs your help.
- Please help improve Wikipedia.
- Please improve this article right now because it really needs it badly.
- See a problem? Fix it now!
- Help make our articles easier to understand.
- When you improve Wikipedia, you help 200 million people.
- Wikipedia is viewed 250 million times every day! Make it the best it can be!
- I'm not sure how to accomplish this. Perhaps rotating banner ads for each type of phrase with a small static graphic to catch a reader's attention without making it look like part of the article. —UncleDouggie (talk) 08:38, 16 February 2011 (UTC)
- I love options 2 to 6, especially 4 and 5. Do you not think a banner ad would be opposed by many on aesthetic grounds? --Anthonyhcole (talk) 09:17, 16 February 2011 (UTC)
- It's really depends how you read it, but #4 could be taken as ... more demanding then intended. ("Fix it! Now! This second! Or else!") --Yair rand (talk) 09:30, 16 February 2011 (UTC)
- Here's an updated set of messages:
- Please help improve Wikipedia.
- Please improve this article if you can.
- See a problem? Just fix it!
- Help make our articles easier to understand.
- When you improve Wikipedia, you help 200 million people.
- —UncleDouggie (talk)
- I really like 2, 3 and 4; though for 4, I'd prefer "Help make this article easier to understand." --Anthonyhcole (talk) 10:13, 21 February 2011 (UTC)
- This solves the concern I expressed above, but I still don't think that it's the best we can do. It's a lot of words and can be seen as inviting vandalism. We need a short message that somehow conveys these thoughts in an inviting way without asking people to just do an edit test:
- I like that very very much. --Anthonyhcole (talk) 14:14, 14 February 2011 (UTC)
I'm asking for the extension because I want to improve the sensitivity of the trial. The more edits we have to evaluate, the higher our likelihood of drawing meaningful conclusions. I know I should have announced this at the outset but then I thought 20 articles over 30 days would be plenty, but the longer I've had to think about this, the more data I think we need. Another 30 days would be good. I think there is consensus to run this trial. What I'm saying is I'd like this trial we have consensus for, to be better, more useful. --Anthonyhcole (talk) 14:14, 14 February 2011 (UTC)- On second thoughts, given the brouhaha at the template deletion discussion, I'd prefer to see what effect, if any, can be detected from 30 days before asking for additional time. It shouldn't make any difference whether this trial is extended, or another 30 day sample is taken later on. I was hoping to have collated some interim results by now, but haven't gotten round to it, and don't want to go back to Wikipedia:Village pump (proposals) for an extension without bringing figures to support the request. --Anthonyhcole (talk) 06:09, 15 February 2011 (UTC)
- That sounds sensible. — Martin (MSGJ · talk) 09:21, 15 February 2011 (UTC)
Could we try to keep this to non-medical, non-technical articles?
I don't really what help the banner brings to articles like scabies (say). Are you hoping to draw in human parasite experts who were just passively reading along, unaware that they could change anything? How likely is that? SBHarris 06:00, 17 February 2011 (UTC)
- I don't know what the odds are, but that would be a good outcome. --Anthonyhcole (talk) 11:16, 17 February 2011 (UTC)
- With 70% of medical students using Wikipedia drawing some of them into editing would be great.[1] I met a peads resident in El Salvador last week. She used Wikipedia extensively ( as did the rest of her class ). Yet she did not realize that she could edit. We need to get the word out. Doc James (talk · contribs · email) 14:54, 20 February 2011 (UTC)
The look
If something like this is implemented across the project, it's interface will not necessarily be a green strip just above the lede. It is presently in that configuration because the template technology is easily adaptable to small trials like these, and it is so blatant that we can be fairly confident most readers will read it. My preferred option is
"You can edit this page. Click here to find out how."
immediately above the first line of the lede, where here drops the tutorial onto the page below the invitation. That is, a simple line of black text, in a different font, maybe, with no box or background colour. In that position it will be read by most readers without having to be distracting. My second preference is having it as a line of black text under the article title, displacing "A featured article from Wikipedia, the free encyclopedia" (which could move to the top right where the FA star is at present). --Anthonyhcole (talk) 09:52, 2 February 2011 (UTC)
- I would favour some format which clearly distinguishes it from the article itself. The green background does this quite well, although it could perhaps be paler and/or smaller. What about linking to the tutorial, instead of transcluding it on the article itself? This would avoid all the various issues with browsers that don't display or collapse the box correctly. It might also be clearer to read without having the article on the page as well. — Martin (MSGJ · talk) 11:03, 2 February 2011 (UTC)
- It is easier to refer to the tutorial when it's on the article page, you just scan or scroll up and down. Opening and closing pages takes time (quite a lot on some articles). As I said at the top of this page, I hate boxes and templates, but I'm sure I'm in the minority. --Anthonyhcole (talk) 11:40, 2 February 2011 (UTC)
Would the bar eventually only appear to anonymous editors, or will it appear for all editors? Gary King (talk · scripts) 14:59, 2 February 2011 (UTC)
- Ideally it would only appear to unregistered users. I did look into this once but never managed to persuade anyone to write the code for it (I did try pestering User:Alex Smotrov but he doesn't seem active anymore). Maybe you can help us with this Gary? — Martin (MSGJ · talk) 15:10, 2 February 2011 (UTC)
- It could be useful to add to MediaWiki:Common.js code that would then allow anyone editing templates to make them visible only to a select few people. For instance, all you'd have to do is add class="for-administrators-only" to a template to make it only visible to administrators (that should be obvious enough). Thoughts on whether or not this could be useful in more situations than just here? Gary King (talk · scripts) 18:47, 2 February 2011 (UTC)
- Okay, I wrote the script. Do the following:
- It could be useful to add to MediaWiki:Common.js code that would then allow anyone editing templates to make them visible only to a select few people. For instance, all you'd have to do is add class="for-administrators-only" to a template to make it only visible to administrators (that should be obvious enough). Thoughts on whether or not this could be useful in more situations than just here? Gary King (talk · scripts) 18:47, 2 February 2011 (UTC)
- Go to Special:MyPage/skin.js.
- Paste the following into that page: importScript('User:Gary King/user groups.js');
- Perform a hard refresh (by following the instructions on that page that says "After saving, you have to...").
- To test the script, go to any page (preferably a short one so that it doesn't take as long to load), and then preview that page. Wrap some random text in <div class="for-anonymous-only"></div>; so for instance, you could have <div class="for-anonymous-only">This is a test.</div>. This will show "This is a test." to anonymous users only, so you won't see it.
- This works for all 20 user groups (sysop, bureaucrat, oversight, etc.)
- I brought this suggestion up here and it was mentioned that this could be used for editors to create vandalism that would only appear to anonymous users, so this idea needs some tweaking first. My suggestions were: "Perhaps hidden content could be marked with an icon? Or just tag anonymous edits including these classes. It could also be a gadget for those who want to hide templates such as {{Invitation to edit}}, so then by default nothing is changed." Your thoughts? Gary King (talk · scripts) 21:51, 4 February 2011 (UTC)
- I haven't fully understood. Are you saying your suggested code would do the job but monitoring for vandalism would be impaired because of its invisibility to logged in users? As for vandalism; couldn't we fully protect any templates carrying that code? --Anthonyhcole (talk) 00:14, 5 February 2011 (UTC)
- Nope, because this code is used by the browser and not by templates, so anyone could create it. Even if it was code that could be inserted into a fully protected template, it could still then be inserted anywhere, so it would have had the same problems, anyway. Gary King (talk · scripts) 03:08, 5 February 2011 (UTC)
- OK. Am I right in thinking there is probably a way to make the invitation visible only to anonymous editors, but it requires more investigation? Or would there be any benefit in taking this to the Wikimedia Foundation? They seem to have a budget allocation for this very kind of thing. --Anthonyhcole (talk) 03:50, 5 February 2011 (UTC)
- There's actually already an extension to MediaWiki, which Wikipedia runs on, that has code to only show content to a specific group, called Conditional Show. Getting Wikimedia to install it to Wikipedia will probably be difficult, though. The solution that I gave above is a JavaScript solution, which is easier to implement but is applied by the client (the user's browser) rather than the server. Gary King (talk · scripts) 05:21, 5 February 2011 (UTC)
- Gary, perhaps you are not aware but we already have such code in MediaWiki:Common.js to control visibility of content for administrators and account-creators. This was added when the editnotice links were put in. So I don't see that extending it to registered/unregistered users should be a big deal. — Martin (MSGJ · talk) 12:45, 7 February 2011 (UTC)
- I think that the primary concerns seem to be that unregistered users are less familiar with how the wiki works, so if they come across vandalism they don't know how to fix it. So the fact that vandalism could possibly appear to unregistered users while remaining invisible to registered users is a problem. One idea, as mentioned earlier, is to have these types of banners just show to all users as they do now, but then create a gadget that only shows banners specific to their group. This approach would be an "opt-in" program instead of an "opt-out" one. This ensures that at least someone will always see vandalism, even when it is only intended for an unregistered user. Gary King (talk · scripts) 18:31, 7 February 2011 (UTC)
- Gary, perhaps you are not aware but we already have such code in MediaWiki:Common.js to control visibility of content for administrators and account-creators. This was added when the editnotice links were put in. So I don't see that extending it to registered/unregistered users should be a big deal. — Martin (MSGJ · talk) 12:45, 7 February 2011 (UTC)
- There's actually already an extension to MediaWiki, which Wikipedia runs on, that has code to only show content to a specific group, called Conditional Show. Getting Wikimedia to install it to Wikipedia will probably be difficult, though. The solution that I gave above is a JavaScript solution, which is easier to implement but is applied by the client (the user's browser) rather than the server. Gary King (talk · scripts) 05:21, 5 February 2011 (UTC)
- OK. Am I right in thinking there is probably a way to make the invitation visible only to anonymous editors, but it requires more investigation? Or would there be any benefit in taking this to the Wikimedia Foundation? They seem to have a budget allocation for this very kind of thing. --Anthonyhcole (talk) 03:50, 5 February 2011 (UTC)
- Nope, because this code is used by the browser and not by templates, so anyone could create it. Even if it was code that could be inserted into a fully protected template, it could still then be inserted anywhere, so it would have had the same problems, anyway. Gary King (talk · scripts) 03:08, 5 February 2011 (UTC)
- I haven't fully understood. Are you saying your suggested code would do the job but monitoring for vandalism would be impaired because of its invisibility to logged in users? As for vandalism; couldn't we fully protect any templates carrying that code? --Anthonyhcole (talk) 00:14, 5 February 2011 (UTC)
- I brought this suggestion up here and it was mentioned that this could be used for editors to create vandalism that would only appear to anonymous users, so this idea needs some tweaking first. My suggestions were: "Perhaps hidden content could be marked with an icon? Or just tag anonymous edits including these classes. It could also be a gadget for those who want to hide templates such as {{Invitation to edit}}, so then by default nothing is changed." Your thoughts? Gary King (talk · scripts) 21:51, 4 February 2011 (UTC)
Implementation
How about we implement it and then see if it causes any issues. If it does, it can be removed again easily. We don't know that this will cause any problems: until we try it we are just guessing. And it's only one editor who has expressed this concern. It would be far neater if we could combine the code for WP:Upload with this, and there are definately other applications for this functionality (the article wizard for one). — Martin (MSGJ · talk) 18:48, 7 February 2011 (UTC)
- It's up to you, ultimately, if you want to implement this since I'm not an admin. I can help with the integration if you want, though. Also, I've moved the discussion from WP:VPT to MediaWiki_talk:Common.js#Add_new_features_to_show.2Fhide_content_based_on_user.27s_groups.3F as that's where the code would ultimately be implemented, so the guys watching that page are more technically savvy than most. Also, I do agree that if this were to be implemented, then the existing code for sysops and accountcreators would be merged in. The biggest decision is to decide whether or not it's worth waiting days, or possibly weeks or more to receive sufficient discussion on this to determine if it's worth going forward on or not. It may just be the best option to implement it to see if anything breaks, especially since awareness of the code would be slow at first (meaning vandalism would be minimal, at least to begin with). With regards to my suggestion to create a gadget for "opt-in" purposes, another idea is to make this "opt-in", meaning by default the script acts as normal, but admins and other users have the choice of showing all hidden content, which would help in patrolling vandalism. Gary King (talk · scripts) 19:08, 7 February 2011 (UTC)
- You'd have to tell me exactly what to do because I don't want to break anything. (BTW why are you not an admin?) I'm not quite understanding your opt-in suggestion yet, but I think it might be better to keep things as simple as possible to begin with. By the way we have sysop-show and accountcreator-show. What about sysop-hide or autoconfirmed-hide? Are these possible using the same system? — Martin (MSGJ · talk) 19:18, 7 February 2011 (UTC)
- Speaking of the existing classes (sysop-hide, etc.), which do you prefer? "for-sysop-only" and "not-for-sysop", or "sysop-show" and "sysop-hide"? I chose the former for the script but the latter is what currently exists. The advantage of the former is that it's easy for someone unfamiliar with the code to know what the code does. The advantage of the latter would be that nothing would have to be changed when the new code is implemented. Or we could use something like "show-for-sysop" and "hide-from-sysop", which is similar to the latter but is made a bit more clear. Do we know which pages use sysop-show/hide, then we can change those when the new code is implemented? Gary King (talk · scripts) 19:51, 7 February 2011 (UTC)
- Another thing we could do is if the content is supposed to be hidden for registered users, we could make it so that it still appears for admins. Therefore, regular, registered users would not see the content as intended, but admins would see it, so if there was anything wrong then they could do something about it (since admins are more familiar with the inner-workings of the wiki than the average user). Also, as I suggested before, we could also show an icon whenever content is hidden, so that it's always clear that something is intentionally hidden from the user. A simple, small square icon, about the size of the {{Featured article}} icon, could be used. And perhaps we could make it so that if the user wants to, they could click on the icon to expand the hidden content. This would definitely help in decreasing any malicious use of this hidden content feature. Gary King (talk · scripts) 01:21, 8 February 2011 (UTC)
- The show-to-autoconfirmed and hide-from-autoconfirmed schema sounds good to me. And your other suggestions sound reasonable as well, i.e. showing everything to admins and trying out the icon thing. The sysop-show and accountcreator-show are currently used at Template:Editnotice load/core; I'm not aware of any other uses. — Martin (MSGJ · talk) 09:33, 8 February 2011 (UTC)
- Another thing we could do is if the content is supposed to be hidden for registered users, we could make it so that it still appears for admins. Therefore, regular, registered users would not see the content as intended, but admins would see it, so if there was anything wrong then they could do something about it (since admins are more familiar with the inner-workings of the wiki than the average user). Also, as I suggested before, we could also show an icon whenever content is hidden, so that it's always clear that something is intentionally hidden from the user. A simple, small square icon, about the size of the {{Featured article}} icon, could be used. And perhaps we could make it so that if the user wants to, they could click on the icon to expand the hidden content. This would definitely help in decreasing any malicious use of this hidden content feature. Gary King (talk · scripts) 01:21, 8 February 2011 (UTC)
- Speaking of the existing classes (sysop-hide, etc.), which do you prefer? "for-sysop-only" and "not-for-sysop", or "sysop-show" and "sysop-hide"? I chose the former for the script but the latter is what currently exists. The advantage of the former is that it's easy for someone unfamiliar with the code to know what the code does. The advantage of the latter would be that nothing would have to be changed when the new code is implemented. Or we could use something like "show-for-sysop" and "hide-from-sysop", which is similar to the latter but is made a bit more clear. Do we know which pages use sysop-show/hide, then we can change those when the new code is implemented? Gary King (talk · scripts) 19:51, 7 February 2011 (UTC)
- You'd have to tell me exactly what to do because I don't want to break anything. (BTW why are you not an admin?) I'm not quite understanding your opt-in suggestion yet, but I think it might be better to keep things as simple as possible to begin with. By the way we have sysop-show and accountcreator-show. What about sysop-hide or autoconfirmed-hide? Are these possible using the same system? — Martin (MSGJ · talk) 19:18, 7 February 2011 (UTC)
No other comments at MediaWiki talk:Common.js. I'm ready to implement if you tell me exactly what to do. Just go with whatever you think best, keep it simple, and we can tweak it later. Thanks for your help with this. — Martin (MSGJ · talk) 08:36, 9 February 2011 (UTC)
- Touching wood, holding breath. --Anthonyhcole (talk) 15:00, 9 February 2011 (UTC)
- I think progress is happening. We will just have to be patient! — Martin (MSGJ · talk) 15:48, 9 February 2011 (UTC)
- exhale. --Anthonyhcole (talk) 16:18, 9 February 2011 (UTC)
- Sorry for taking so long. Is this required immediately? What I have so far at User:Gary King/user groups.js is the basics, meaning it doesn't include the icon or the ability to still show content to admins. If this is sufficient to implement at the moment, then we can do that. Otherwise, I'm quite busy this week, but the following week of Feb. 21–25 is a holiday for me. Gary King (talk · scripts) 22:02, 12 February 2011 (UTC)
- Sorry for the late reply. I think I will speak for everyone here when I say that we would still like this to be done. However as the trial is now over it may not be needed imminently. So take your time and do it when convenient for you. By the way, it seems that this code seems relevant and may duplicate part of what you are doing. — Martin (MSGJ · talk) 18:09, 22 February 2011 (UTC)
- Sorry for taking so long. Is this required immediately? What I have so far at User:Gary King/user groups.js is the basics, meaning it doesn't include the icon or the ability to still show content to admins. If this is sufficient to implement at the moment, then we can do that. Otherwise, I'm quite busy this week, but the following week of Feb. 21–25 is a holiday for me. Gary King (talk · scripts) 22:02, 12 February 2011 (UTC)
- exhale. --Anthonyhcole (talk) 16:18, 9 February 2011 (UTC)
- I think progress is happening. We will just have to be patient! — Martin (MSGJ · talk) 15:48, 9 February 2011 (UTC)
- Sorry for the wait. I've updated the script just now and posted instructions on how to use it, here: User_talk:Gary_King#Usergoups. I'll also copy the instructions below:
- Go to Special:MyPage/skin.js, and add the following line to it:
importScript('User:Gary King/user groups.js'); // [[User:Gary King/user groups.js]]
- Then, bypass your cache (usually, clicking on your browser's "Reload" button while holding the Shift key should be sufficient). And finally, go here, and there should be some hidden content there. (If you go to that page while logged out, you'll see the hidden content in full. Otherwise, you can just click on the "(+)" links to show the hidden content.) Gary King (talk · scripts) 21:54, 25 March 2011 (UTC)
- Also, regarding that link to that code change you linked to, that does look like something similar. However, it's more for controlling how the wiki looks like to members of each group rather than controlling who sees what content. My script lets the user control who can see content based on their user group, while the code change shows the entire wiki differently based on the user's usergroup. For instance, the code change could be used to hide the "delete page" button to pages from non-sysops. Gary King (talk · scripts) 21:57, 25 March 2011 (UTC)
Can we farm this out?
I'm wondering if, once we boil the interface down to something that won't start a revolt, we might hand the trialling of the concept to the same crowd that trialled the fund raiser this year. They obviously have the expertise. I guess that would mean getting the foundation to take it on, but I think trialling this strategy to encourage more editors is something the foundation could fund. Thoughts? --Anthonyhcole (talk) 13:33, 16 February 2011 (UTC)
- I like the idea, but only if they are as concerned about participation as they are about getting money. I know that some board members will support it, but I'm not sure that's enough. —UncleDouggie (talk) 17:50, 16 February 2011 (UTC)
- I think that if it shows promise, that it will be supported. There's a big discussion right now about getting more women to edit, for example, and this could be one part of a strategy to encourage editing. WhatamIdoing (talk) 18:58, 16 February 2011 (UTC)
- There was a discussion of running an invitation to edit banner after fundraiser finished this year. Not sure where this ended up. Doc James (talk · contribs · email) 14:50, 20 February 2011 (UTC)
- Are you suggesting that the foundation might take this on but do a poor job due to lack of interest, UncleDouggie? I have to say the management of the pending changes trial as described at the current request for comment seems a bit ham-fisted, and the foundation had some involvement in that. But much of that problem seems to be down to poor/unclear communication with the community.
- I think that if it shows promise, that it will be supported. There's a big discussion right now about getting more women to edit, for example, and this could be one part of a strategy to encourage editing. WhatamIdoing (talk) 18:58, 16 February 2011 (UTC)
- When we have a form of words and interface most collaborators approve of, let's put it with the results of this trial and a detailed study plan to the community via a centralized discussion. If we get approval, that may be the right time to decide whether to involve the foundation. --Anthonyhcole (talk) 10:07, 21 February 2011 (UTC)
- The approach you are using here is the right way to go, as opposed to the Pending changes debacle. Perhaps this trial was a good idea just because it attracted more community interest in doing it better with a relatively small impact to the project. I'm also impressed that you actually ended it when you said, which is rather essential for goodwill in a future trial; something that the PC supporters can't comprehend. I don't know to what degree the Foundation will support this effort. I was expressing concern because I see them directing efforts in other areas. However, they have supported several big outreach efforts such as GLAM. Calling this effort outreach to new editors will probably help get support. —UncleDouggie (talk) 12:27, 21 February 2011 (UTC)
- We'll see. The trial certainly attracted community interest; and the TfD discussion was highly productive, I thought. --Anthonyhcole (talk) 10:26, 24 February 2011 (UTC)
- The approach you are using here is the right way to go, as opposed to the Pending changes debacle. Perhaps this trial was a good idea just because it attracted more community interest in doing it better with a relatively small impact to the project. I'm also impressed that you actually ended it when you said, which is rather essential for goodwill in a future trial; something that the PC supporters can't comprehend. I don't know to what degree the Foundation will support this effort. I was expressing concern because I see them directing efforts in other areas. However, they have supported several big outreach efforts such as GLAM. Calling this effort outreach to new editors will probably help get support. —UncleDouggie (talk) 12:27, 21 February 2011 (UTC)
- When we have a form of words and interface most collaborators approve of, let's put it with the results of this trial and a detailed study plan to the community via a centralized discussion. If we get approval, that may be the right time to decide whether to involve the foundation. --Anthonyhcole (talk) 10:07, 21 February 2011 (UTC)