Wikiversity:Colloquium/archives/September 2014

UploadWizard
UploadWizard is a MediaWiki extension that greatly simplifies the process for uploading files to a MediaWiki wiki. To see the UploadWizard in operation, visit Commons:Special:UploadWizard. In order to activate the extension here at Wikiversity, we need community consensus to do so. Please discuss as needed and then reply to this thread and indicate or  the addition of this extension. -- Dave Braunschweig (discuss • contribs) 01:28, 7 July 2014 (UTC)

- I've used upload wizard on commons and found it easy and efficient. As I recall it also allows us to specify fair use images and provide appropriate licensing information before the upload can occur. This may help us to obtain proper licensing per image at the time of upload or prevent upload without licensing info. --Marshallsumter (discuss • contribs) 02:13, 7 July 2014 (UTC)

Some administrators should be responsible for the local importations in fair use as it needs some licensing skills (did you know that we can publish the Eiffel Tower photos unless it has been taken during the night because of the light system copyright?) and Commons:Special:UploadWizard is forbidding the fair use.

Apart from that every other media should be hosted on Commons to allow our courses to be translated and benefit to the Commons licenses templates, huge categories and licensing specialists. Moreover when we copy some images from one wiki to another the image can change behind our back, as its local name can be identical as the Commons one.

That's why I propose my bot to migrate at least our 3,773 Category:Public domain images as I did for the French Wikiversity (please see Commons:Commons:Bots/Requests/JackBot).

Sorry for my late answer and. JackPotte (discuss • contribs) 08:42, 16 August 2014 (UTC)


 * If JackPotte is correct and the Upload Wizard cannot be locally modified to allow Fair Use images then I agree; however, migrating our Public Domain images to Commons may not be a good idea. Deletionists on commons may cause the loss of some of our images which creates more work here to re-upload them as Fair Use. --Marshallsumter (discuss • contribs) 12:04, 16 August 2014 (UTC)


 * To me they are separate issues. The current process and the UploadWizard are two different approaches to *how* content is uploaded.  Neither one controls *what* is uploaded.  Either one can ultimately be controlled by filters and/or a bot to manage the content.
 * Regarding the proposal that we should be Fair Use only, in theory I agree. In practice, however, it presents problems.  Content is often deleted from Commons without adequate notice here.  It's happened to me, and it frequently happens to Marshall.
 * Dave Braunschweig (discuss • contribs) 13:08, 16 August 2014 (UTC)

As long as non-free use files are not locked out. As to migrating all our free use files to Commons, this is an idea with no benefit and much harm, as we have seen. Commons files have been used here, they stand for years, and then they are deleted on some obscure technicality, which may have been necessary or not (Commons uses the Precautionary principle, but may take years to apply it, and then it's applied quirkily, and who has time to engage in these discussions?), and our pages are then damaged, even if we could claim non-free use with a rationale; to obtain the file is extra work, i.e. we'd need to get a Commons admin to undelete it and supply it. There is no benefit to Commons; if any user thinks a file should be on Commons, they may easily copy it there. However, our files should not be deleted just because they have been copied to Commons. A bot, however, could clean up names. I'd think that ideally, our public domain files should have the same name as an identical Commons file; and be linked that way. In that way, if the Commons file is deleted, there is no work to do here. --Abd (discuss • contribs) 13:46, 16 August 2014 (UTC)
 * As far as I know Commons, if an image is considered of poor quality it's not deleted if it's used in a course. And letting a duplicate here will make the image on Commons unused in the paragraph at the bottom. JackPotte (discuss • contribs) 15:19, 16 August 2014 (UTC)
 * That is correct, usually. However, if an image is of general utility, it should remain on Commons even if it is not used here, but that's not really our problem. (If a file is hosted in both places, that it does not display as used on Commons is a bug, my opinion. However, I can see that there could be a version problem.)
 * Commons is a repository of free content. We are for the creation of educational resources, and we can use images that are not free, with a non-free rationale. An image may be used here with the belief that it is properly licensed. If that changes, it is then possible to still use it with a fair use claim.
 * Our policy on non-free usage may change over time, it's within our prerogative per WMF policy. The problems that have arisen have been over a decision on Commons that a license, for content standing for years, was somehow defective. The *main* thing that we need to do is to make sure that any non-free content is machine-readably tagged. That creates a warning for any content re-user, the concern of the WMF. Jack, the content is protected if it is here, and we are responsible for determining what we protect. Commons is fantastic. And not our mission. Basically, leaving images here is the most efficient practice for our mission.
 * Jack, your opposition is over an Upload Wizard that allows users to upload content here. Do you really intend to make it more difficult than it need be for users to upload images for usage here?
 * By the way, we specifically do not want to confine fair use claims to custodians. That's backwards. Fair use, under WMF policy, requires a content judgment. It does not require usage of custodial tools (unless a deletion is required, but, even then, deletion decisions are still a matter of community consensus, outside of uncontroversial deletions.) --Abd (discuss • contribs) 18:06, 16 August 2014 (UTC)
 * In spite of your pleading I choose to maintain my vote because we can already provide the UploadWizard for all by redirecting toward Commons. When you say that fulling it is not our mission I understand, but it seems to don't take into account that every Wikiversity course can easily be translated in several languages sooner or later. JackPotte (discuss • contribs) 18:25, 16 August 2014 (UTC)

Strong oppose - Simply because I cannot use commons. --Goldenburg111 21:15, 14 September 2014 (UTC)
 * What does not using Commons have to do with whether or not we use an updated user interface here? -- Dave Braunschweig (discuss • contribs) 00:26, 15 September 2014 (UTC)
 * Goldenburg, I think you have misunderstood. JackPotte seems to be arguing that we only allow uploads to Commons (which conflicts with our allowance of Fair Use), that is not the proposal here. The proposal here is to enable the Upload Wizard for Wikiversity as part of our MediaWiki setup. --Abd (discuss • contribs) 01:21, 15 September 2014 (UTC)

My apologies for being an idiot. I this discussion. Thanks. --Goldenburg111 01:48, 15 September 2014 (UTC)
 * Goldenburg, you are far from an idiot, and the reason you are far from it is that, early on, here and in your life, you developed the capacity to recognize and acknowledge your mistakes without being attached to being right. That, combined with your developed ability to express yourself -- so that you reveal those mistakes! --, makes you a fast learner. You will go far, it's predictable from your trajectory now, because you have already gone far. You are an inspiration to us. --Abd (discuss • contribs) 12:14, 15 September 2014 (UTC)

i enjoy too much wikiversity it helps us the student to improve our knowledge in our study
--41.73.114.229 (discuss) 00:07, 27 August 2014 (UTC)

Very true. I always use Wikiversity for my studies, and I am successful in school. I hope you have a great time here! --Goldenburg111 21:31, 14 September 2014 (UTC)

Rotlink Bot and Archive.is
Recently I have become concerned by the edits of User:Rotlink, which appears to be primarily bot-driven. Short explanation is that Rotlink searches pages, seemingly at random, and replaces dead links with links to various Internet archives. However, the user behind the Rotlink account runs archive.is, which now redirects to archive.today. Wikipedia has discussed this issue extensively at Wikipedia:Wikipedia:Archive.is_RFC. User:Rotlink is now blocked at Wikipedia for bot use, archive.is links are blocked from addition to Wikipedia, and all links to archive.is are being rolled back or replaced with links to other archives.

How does Wikiversity want to address this situation? -- Dave Braunschweig (discuss • contribs) 13:06, 28 August 2014 (UTC)
 * The Wikipedia response was particular to en.Wikipedia. The Wikipedia RfC was run and closed in a way that many considered a problem. There was no actual consensus. What is clear, in fact, is that Rotlink is doing good work, and if you review the Wikipedia RfC, it was not shown that Rotlink was damaging content. I have suggested to Rotlink, on the Talk page here, that COI be formally disclosed, but it is well-known that Rotlink is affiliated with archive.is.
 * Is Rotlink a bot? Meta policy allows unapproved bots if they do not edit at a rate higher than 1 edit per minute. However, humans can easily edit at 6 edits per minute, I've demonstrated 10, even for extended periods. Rotlink has 171,352 edits globally, which is huge, but there are ordinary editors, not bots, with more than 500,000 edits.
 * Looking at recent global contributions, I see 50 edits from 12:32 to 13:14. That's 42 minutes. Very slightly over 1 per minute. There is no way for us to know if Rotlink is a bot or a human, perhaps using an automated editor. (The difference is that an automated editor presents a human with an Accept button, one approval per button push, a bot just goes ahead.) Bots, though, usually hit much higher edit rates unless throttled back.
 * How many of the Rotlink edits created an actual content problem? Rotlink is now almost entirely adding links to archive.org. Archive.is or archive.today links are relatively rare, hard to find.
 * In none of the extensive discussion of Rotlink has a bad edit been shown. Negative opinion about archive.is was often of the nature of "they might do something bad in the future.
 * There is also RotlinkBot which has bot status on 6 wikis, is blocked on more than that.
 * Rotlink demonstrated that you can follow w:WP:IAR Wikipedia Rule Number One, causing no harm at all, and be blocked or banned. That RfC failed to implement what it demanded: a blacklisting of archive.is, and a global blacklisting request also failed. Essentially, the active core wants one thing ("obey our authority"), the real community actually wants better content.
 * AFAIK, Rotlink has violated no Wikversity policy. If we think that Rotlink is harming Wikiversity, policy or not, we may request Rotlink stop, and warn if it continues, and block if the warning is ignored. But it would be entirely contrary to our traditions to request that beneficial behavior stop.
 * We don't want unapproved bots running, for good reason. However, this user, bot or not, has, for a long time, only done good work. If the account is a bot or we think it is, we might consider approving it. Is.wikipedia apparently did. Pl.wiki actually gave it the flood right.
 * The issue with unapproved bots is flooding, overwhelming the capacity of the community to review edits. Rotlink is not doing that here. However, after checking a lot of Rotlink edits, I'm not doing it any more. Rotlink has been assigned Trusted status on some wikis, and it's no wonder.
 * I welcomed Rotlink's work on the user talk page. I suggested specifying possible conflict of interest, per WMF policy. However, there is no COI with the vast majority of edits, in recent edits, some time ago, I think I found only one archive.is link. This is trivial and actually harmless. Anyone may communicate with Rotlink, and the user has email enabled. Rotlink was not notified of this question, and I'm not doing it, because I consider it a waste of time. A formal warning would be necessary before taking action. --Abd (discuss • contribs) 14:49, 28 August 2014 (UTC)


 * Archive.is is still being used. One or more links were added today, and many were added this month.  You can do a full text search for "archive.is" (quotes required) to verify.  Separately, while meta policy may allow bots at a low rate, Wikiversity does not.  According to Bots, 'The operation of a bot requires approval.'  This is official Wikiversity policy.  If we don't like the policy, we can change the policy, but we shouldn't allow users to ignore the policy.  As to whether or not this is a bot, the editing occurs often enough that it should be approved and designated as a bot to put the proper edit tag on it if we want this activity, and it should be blocked if we don't.  -- Dave Braunschweig (discuss • contribs) 16:43, 28 August 2014 (UTC)


 * Archive.is is still being used. One or more links were added today, and many were added this month.  You can do a full text search for "archive.is" (quotes required) to verify.  Separately, while meta policy may allow bots at a low rate, Wikiversity does not.  According to Bots, 'The operation of a bot requires approval.'  This is official Wikiversity policy.  If we don't like the policy, we can change the policy, but we shouldn't allow users to ignore the policy.  As to whether or not this is a bot, the editing occurs often enough that it should be approved and designated as a bot to put the proper edit tag on it if we want this activity, and it should be blocked if we don't.  -- Dave Braunschweig (discuss • contribs) 16:43, 28 August 2014 (UTC)
 * Dave, first of all, policy exists to serve the project and users, not the other way around.
 * We have the same policy here as meta, look at the fine print, my emphasis:
 * Bots running without a bot flag should edit at intervals of over 1 minute. Once they have been authorised and appropriately flagged, they should operate at an absolute minimum interval of 5 seconds.
 * Our policy governs local editing. The 1 edit per minute approximate rate of Rotlink, when active, is global, not local. Rotlink is not close to the margin here, but far from it.
 * This discussion is not a warning of the user. This is a request for Wikiversity comment. My opinion is that no user should ever be blocked if their editing is not harming the project, and particularly if it is benefiting the project. This is an application of a rule that we have not formally implemented here, but it was made Rule Number One on Wikipedia, and that, in fact, was merely common law.
 * I did do more research on this, including looking at links to archive.is, but I did not report it, since my response was already long. Since Dave brought it up, here is what I found:
 * Rotlink, from CA, has 293 edits to this wiki.
 * There are 37 links to archive.is on en.wikiversity. One was added by Marshallsumter.
 * 22:23, 26 August 2014 added three links to archive.is
 * 14:54, 4 January 2014 added one link to archive.is.
 * 17:18, 22 August 2014 added one link to archive.is
 * that leaves 31 links, and unless I missed something, all were added in October, 2013.
 * Archive.today has 4 links on en.wikiversity, none were added by Rotlink.
 * There is no problem. It appears that almost all of the activity of Rotlink is adding archive.org links. Dave, I would guess that you saw two edits, listed above.
 * What I have not researched: were there archive.org links available for those four pages linked to archive.is? My guess is not, but that's just a guess. And is it worth the effort to find out? How will this benefit Wikiversity? I've put two hours into this because Rotlink is benefiting this project and I don't want it to stop. I don't really care if links are to archive.is. There are still 16,292 (just now) links on en.wikipedia to archive.is, almost a year after the RfC was closed. The RfC had it that there were 10,000 links when it was filed. Now, I know that the number of links went far higher than that, it went to over 30,000. Links stopped being added because an Edit Filter was written to prevent ordinary editors from adding links. That was not consensus, it appears it was unilateral.
 * So the question for the community. Rotlink may not speak English, and the user may not have time to bother with separate wikis, he's working on hundreds of them. Do we want to put the user through some bureaucratic process to get an approval that isn't needed by our policy, for the edit rate here? I've looked at a lot of Rotlink edits, maybe hundreds of them, and every one has been good. That was the strong report on the en.wiki RfC. Where Rotlink is blocked, it is not for bad edits!
 * There is no harm in allowing Rotlink do what Rotlink does, as long as the edit rate here is below 1 per minute, which, the way that Rotlink operates, it will always be.
 * If we don't want Rotlink to edit, we can block the user. Here is what happens when Rotlink is questioned. Nothing. The question raised was an esoteric one. Rotlink fixes dead links. Wiktionary quotes sources, that may include dead links. The user complaining thinks they should be left as-is, but when those materials were created, the links were live. Insisting on maintaining the literal link when it is dead is ... strange, very literalist, it will waste user time. However, a compromise would be to add a note, using the Rotlink link in addition to the original. We are very, very unlikely to see a problem like this. I think I know what happened with https://en.wiktionary.org/wiki/beigist. That was an archive.org page, and it was taken down after the link was added. That's all. The restored link is behind a pay wall (if it still exists). Any link can break.
 * We have lots of users who do not respond to warnings and requests. We do not therefore block them. We make an assessment, overall, if the harm outweighs the benefit of allowing the user to edit, and we don't punish for "failure to respond," as some wikis do. --Abd (discuss • contribs) 19:16, 28 August 2014 (UTC)


 * We're obviously looking at different data, because I find 19 archive.is links from August 2014 alone. And unfortunately, I am unable to vouch for the safety of this resource.  I can only tell you that if I wanted to create a botnet, I would do it by having unsuspecting users click on links that would bring them to my server before directing them to their requested content.  If the information at Wikipedia:Wikipedia:Archive.is_RFC is correct, such a botnet already exists, and the links here would serve to expand that network.  Because there is this potential for abuse, already identified on a sister project, I bring it to the community's attention.  If the community finds archive.is links to be a valuable service, so be it.  -- Dave Braunschweig (discuss • contribs) 01:47, 29 August 2014 (UTC)
 * At what data are you looking, Dave? I linked what I used. I did not, however, compile an analysis, imagining that this would be unnecessary and wasted work. Since this has been questioned, I now have, User:Abd/Archive.is. That page shows the four additions in August that I mentioned above, no more. Is there a bug in the special page for external links? Or have I made some other error? Diff or diffs, please! --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC) Daven's comment below split my comment, so I'm adding this sig to it, copied from below. --Abd (discuss • contribs) 20:08, 29 August 2014 (UTC)


 * As indicated initially, I am doing a full text search using "archive.is". I can't give a link, because the link doesn't interpret correctly with quotes in the URL.  It also doesn't interpret correctly using ASCII values for the quotes.  That's why I instead noted 'You can do a full text search for "archive.is" (quotes required) to verify.' -- Dave Braunschweig (discuss • contribs) 19:35, 29 August 2014 (UTC)
 * Awesome. Thanks, Dave. Now, why are some of these links not showing up on the normal special page used to study external links? For example, Molecular Biology contains a link to archive.is. The search shows a date of 03:20, 22 July 2014, which was the date of last edit (by Dave). That is not the date the text was added, . I am, by the way, fixing the reference to show what was archived. This was not properly created in the first place.
 * When I edited the link, which hid the "archive.is" text, it disappeared from the search. The search does not show hidden text (this is a bug in the search engine, in my opinion.) But we also are seeing a problem with Special:External links.
 * If we want to know Rotlink activity in August, neither of these approaches is clearly accurate at this point. Both are tedious and may be incomplete. Instead, this is Rotlink contributions for August (to 21:09, 29 August 2014), examining which, while tedious, will also give us other useful data. I'm compiling this on User:Abd/Archive.is and will come back with a report. --Abd (discuss • contribs) 21:52, 29 August 2014 (UTC)
 * See [] for an analysis of Rotlink August contributions. Summary: Rotlink made 162 edits in the study period. About 184 links were added to archive.org and about 43 links to archive.is. Peak day had 25 edits, peak hour had four edits. In only one case did two edits show up with the same minute. --Abd (discuss • contribs) 01:04, 30 August 2014 (UTC)


 * There are two issues here, Rotlink, a user, probably an unauthorized bot, but operating below non-authorized limits, and links to archive.is. The large majority of Rotlink edits add links to archive.org (the ancient Wayback Machine, or Internet Archive). It appears that where an archive.org page is not available, Rotlink uses archive.is, and Rotlink is apparently being run by the same people who run archive.is.
 * When the archive.is flap arose on enwiki, Rotlink did not respond as "the community" expected. Rotlink's attitude seems to be, this is purely helpful, so what's the problem, I don't want to hassle this, so if you try to stop me, I will follow w:WP:IAR and do it anyway. And so Rotlink edited anonymously. We do not know to what extent the ensuing IP editing was affiliated with Rotlink or was some general support, but that activity is not known to have continued. Rotlink is not editing anonymously here, but openly. So, rather than run a formal Community Review, which would be needed to ban Rotlink, some questions for the community follow below. --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)

The historic alternative before any automatic URL modification, was Dead link. [//fr.wikipedia.org/wiki/Mod%C3%A8le:Lien_bris%C3%A9 On the French Wikipedia this template provides in addition four archive site URLs], including archive.is. So this solution is quite better than Rotlink, and that's why it engendered a development request [//fr.wikipedia.org/w/index.php?title=Utilisateur:JackBot/hyperlynx.py&action=history of a similar bot in 2010, which I took], and modified [//fr.wikiversity.org/w/index.php?title=Triangles_et_parall%C3%A8les/Th%C3%A9or%C3%A8me_de_Thal%C3%A8s&diff=386660&oldid=386650 thousands of French pages] since then. Fortunately now my bot is open source and anyone can submit a modification I could run after. So I propose to: JackPotte (discuss • contribs) 20:52, 29 August 2014 (UTC)
 * 1) Modify Dead link in order to offer a few URLs to get the initial sources.
 * 2) Create Category:All articles with dead external links as a HIDDENCAT.
 * 3) See if we replace all the Rotlink URLs by the new template version.
 * 4) Vote if we periodically launch a bot like that, eventually in parallel of Rotlink.
 * Great. First of all, if Rotlink can operate an unapproved bot making good edits automatically at low rate, so can anyone, without approval, and we don't need to vote on whether or not dead links should be fixed. I.e. we want them fixed, we don't just want to notify users that they are dead. If anyone thinks that having a dead link template (even with suggestions) on a link is superior to having an linked archive of the page, the user is welcome to revert the archive.org or archive.is link and place the template, Rotlink having served to automatically identify the dead link. And then someone else could review the situation, etc.
 * Or one can get bot approval. It's obvious why Rotlink hasn't done this. The process is a pain in the rump; he's running Rotlink globally, and there are something on the order of 600 wikis. So he has apparently elected to go for low rate, tolerating a few wikis blocking him. If they don't want the fixes, that's up to them!
 * So there is already another bot, great! Demonstrate it, please! We will not block you for operating an unapproved bot without warning you first! Just keep the rate as low as Rotlink's and you will be fine. If you want approval, again, go for it! Then you can easily make Rotlink editing here unnecessary, because you can run the bot once a day and stay way ahead of the game, with up to about 10 edits per minute, I think. --Abd (discuss • contribs) 01:31, 30 August 2014 (UTC)


 * JackBot is already approved. -- Dave Braunschweig (discuss • contribs) 02:31, 30 August 2014 (UTC)


 * Great. However, I just reviewed JackBot operation with broken links, below. I'd not be happy to see that here. However, if the rate were low, we could supervise it. This is, in no way, a substitute for what Rotlink does. Take a look at my experience below. --Abd (discuss • contribs) 02:43, 30 August 2014 (UTC)

Allow Rotlink to operate?
Rotlink is running an unauthorized bot. However, the rate is low, and authorization is required for bots out of fear that they will flood a wiki. If a flood of edits appear from Rotlink, I would see no problem with any custodian temporarily blocking Rotlink pending investigation. Given this, the bot is performing two valuable services: identifying dead links and fixing them with links to archives (mostly archive.org). Some wikis have authorized Rotlink and/or Rotlinkbot to operate as a bot. We could do that as well. --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)


 * While Bots says that bots must be approved, there is then a section, added much later after discussion at that has:
 * Bots running without a bot flag should edit at intervals of over 1 minute. Once they have been authorised and appropriately flagged, they should operate at an absolute minimum interval of 5 seconds (12 edits per minute).


 * This clearly allows bots to operate at low rate without approval. (The intention may be just for testing, though). That was taken from meta policy. Rotlink operates globally, and global edit rate is below one edit per minute. (i.e, you can occasionally find minutes with 2 edits, but that does not mean that the bot actually operated at an interval of less than a minute, in terms of submitting edit requests, because there can be delay in those.) And that is why, in spite of all the flap over Rotlink, there has been no global lock request. --Abd (discuss • contribs) 16:49, 30 August 2014 (UTC)

Rotlink operation discussion
From a custodial point of view I would prefer that bots be declared and approved. It ensures that the community is aware of the actions planned, it gives everyone an opportunity to inquire and verify what the bot will do (with documentation), and it allows the bot edits to be tagged as bot, removing them from the default recent changes list. Unlike Abd, I interpret the requirement at Bots literally, that 'The operation of a bot requires approval.' I have my own bot, and I followed these guidelines and waited for approval before using it. We also have a page for bot operation discussion at Bots/Status. -- Dave Braunschweig (discuss • contribs) 19:35, 29 August 2014 (UTC


 * I read the policy in both ways: literally and as to intention.


 * As to literal reading, one needs to read the whole policy. It is obvious that "bot" is referring to high-rate automated editing, that is what is being regulated. This is clear from the exception noted in our policy, that a bot may operate below 1 edit per minute without approval. (Personally, I think that's too fast. I would not want to look at recent changes and see 1200 edits for the last day. But the origins of these policies were in high-watchfulness environments. Rotlink is a global editor, and global sysops watch for stuff like that!


 * As to intention, bot policy has a purpose. Rotlink is not violating the purpose, AFAIK.


 * Dave, you have a bot for doing high-volume editing here. That's a very different situation from that of Rotlink. Rotlink has already demonstrated high reliability, with over 173,000 edits globally. If this were a real problem, Rotlink would be globally locked. It's not. However, if I saw Rotlink flooding this or any wiki, I'd be at SRG in a flash with a global lock request, and it would be granted pronto.


 * Bots are written and tested at low rate, already, before approval.


 * As I mentioned, one of our options is to approve the bot. Would that make you happy, Dave? --Abd (discuss • contribs) 01:31, 30 August 2014 (UTC)


 * But there's more to it than that. This discussion, and the alternatives being discussed are why there should be approval before a bot makes 173,000 edits, or even just 309.  As you point out, a 'low volume' bot could still have a very high impact in a relatively short amount of time.  If this functionality is something we want, then yes, one approach would be to approve Rotlink as a bot.  Although, that, too, would violate policy, because 'Bot operators must: create a separate account for bot operation'.
 * Based on the discussion so far, I would much rather have the functionality suggested by JackPotte than the functionality currently provided by Rotlink. In addition, if this is something we want, then it should be done at a higher volume so that it has an effective impact.  JackPotte has offered to set up his bot to do this.  I would be happy to do the same with mine.  But first we should all agree on (or at least come to consensus on) what it is the bots should do.  -- Dave Braunschweig (discuss • contribs) 02:28, 30 August 2014 (UTC)


 * That is really a separate discussion from this. Rotlink is operating, now, doing useful work. We don't have an alternative, yet. But as to what the bots should do, how about starting with exactly what Rotlink does? After all, maybe Rotlink will stop operating. That's a problem with depending on any private bot. Our page archive bot, here, stopped working eventually when the user apparently stopped running it. If we have something that does the same job, why should we prevent Rotlink from running? Only if we have something better and we want to stop the Rotlink changes, then, it's simple. One button push. When needed. Not now.
 * Dave, are you really proposing that we block Rotlink because, someday, Rotlink might start eating our pages? Running amok? But any user could run amok, any time, and any user could set up a bot at any time. Remember, as well, in the User:Abd/Augusto De Luca study, I tested manual editing at in excess of 6 edits per minute, sustained. Simple to automate that, if I wanted to run amok and get myself blocked and banned. Rotlink is not going to do this.
 * We don't issue preventative blocks absent harmful behavior. Some wikis apparently have decided to do that. I hope we don't start to imitate them. --Abd (discuss • contribs) 02:55, 30 August 2014 (UTC)


 * I haven't proposed blocking Rotlink. I have asked the community what it wants to do.  My personal concerns are already described here.  I believe it is a bot functioning without community notice or approval, and it directs users to a (potentially self-serving) third-party website not related to the link the user is seeking, and without any visual cue or warning.  From a computer security perspective, it is something I would instruct my students never to do.  This approach is how most current malware is distributed.  If a better solution can be found, I would be in favor of that better solution.  JackPotte's proposal would provide a clear indication to the user what site they are actually visiting.  Then, anyone who trusts archive.is is free to click on the link, knowing what they are requesting, and from whom.  From a design perspective, it also centralizes the management of archive options in the dead link template, allowing for easy additions and removals as archive sites come and go.  If I'm the only one who has concerns, then it's my problem.  If others are also concerned, then we can implement a better solution.  -- Dave Braunschweig (discuss • contribs) 03:36, 30 August 2014 (UTC)


 * Is there an action you do propose, Dave? This is an example where general concerns, which you have well expressed, may not apply to the particular situation in front of us.
 * Less than a quarter of the links are to archive.is. If we want to stop Rotlink from adding links to archive.is, that is an action that could be taken, except there seems to be no necessity for it, and the one opinion that is global consensus (with a few exceptions) is that archive.is links are okay, so there is only the technical concern.
 * Rotlink is not going to distribute malware. He'd be shooting himself in the foot. We'd lock the account, take all the links down, lickety-split, I don't care how many there are, there are many bots that could go into gear quickly, and we'd blacklist the site. It would be an emergency and would be handled as such.
 * I'm not concerned about archive.is. I would be concerned about a bot operating like Rotlink operates, if it had not demonstrated a track record of good edits. This is the point: RandomEditorDoingYouAFavor placing links to some unknown site, absolutely, stop it now. But Rotlink automated something that was already happening, with many editors adding links. One of the problems in the Wikipedia RfC and in the meta blacklist discussion is that no analysis was done of how many edits were added by regular editors, first, before Rotlink pushed the issue.
 * Meta blacklisters care much more about "conflict of interest editing," per se, regardless of whether or not an editor is doing good work, I studied this extensively with an editor adding links to lyrikline.org, which was blacklisted for years for no good reason, all the links were good. The editor was blocked to boot, and antispam volunteers vandalized many projects, it's reasonable to call it that. (One nearly was blocked on de.wikipedia.) If he could have gotten away with it, there is a meta admin who was itching to blacklist archive.is. But there was no community support. Basically, to blacklist, you have to remove those links first. There was talk of using a bot to remove them on en.wiki, that went nowhere. Then an admin used the edit filter to disallow any new archive.is links, and few editors understand the edit filter, much less know how to challenge an action like that. I was one. That's why I'm no longer editing Wikipedia! (Successful challenges of administrators don't make one popular with the administrative community.)
 * In any case, thanks, Dave, for being concerned about security. This is not, however, a security issue, just because in some alternate universe it might be. I think we should edit the bot policy to more accurately reflect what we need. Rotlink isn't a problem. An unapproved bot editing at 1 edit per minute for more than short bursts, could be a problem. Actually, though, any busy editor can make a huge mess if nobody is paying attention. An eye-opener for me was looking at all those Rotlink edits. We have a lot of work to do to organize Wikiversity! The time we spend hassling dead links is time we don't spend on the organizational task. --Abd (discuss • contribs) 04:27, 30 August 2014 (UTC)


 * I'm in the universe where Russian hackers recently amassed 1.2 billion usernames and passwords using a botnet. One way botnets are built is using a combination of browser vulnerabilities and URL redirection.  Through Rotlink's activities, we're providing the URL redirection.  I would propose a 'truth in advertising' type of approach where the redirection is clearly labeled.  The link could be displayed as 'archive.is: title'.  I already do the same thing in other external links I create, so that users know the site they are visiting before they click.  JackPotte's proposal would also work, as users would see the name of the archive they are visiting before they click.  I particularly prefer JackPotte's suggestion that it be done through a template, so that any future updates can be performed on a single template rather than what will become thousands of pages and external links.  -- Dave Braunschweig (discuss • contribs) 14:03, 30 August 2014 (UTC)
 * That's an alternate universe, i.e., part of the complete universe that does not apply here. The Evil Plan described requires this: archive.is is a massive undertaking to provide archiving-on-demand-or-need of many, many web pages. That undertaking operates for a long time (it's been at least a year) and many incoming links are created. Then archive.is switches to a malware site. How the Russian hackers collected 1.2 billion usernames and passwords using a botnet has anything to do with Rotlink, I don't know, there appears to be no "botnet" involved. How a malware site that is loaded by following an archive.is link here ends up collecting usernames and passwords is entirely unclear. Perhaps they pretend to be Wikiversity requesting log-in? How one would fool a billion users this way is, again, obscure.
 * So there is a huge investment of time and energy, as represented by archive.is, all to create something that would be shut down within hours of starting to implement the Evil Plan. It makes no sense at all.
 * As to improved process, yes, I agree that links to archive sites should be identified. Many of the archive links do identify, because the original link was bare or visible, so the replaced link is. Hidden links, though, are not identified, but, then again, they were not identified before, necessarily. We have no policy that requires external links to be identified. It also looks like we may have a bug in Special:ExternalLinks, which is a serious problem.
 * We do not create content with policy. One of the major wiki problems is policy formation or central decision-making that is an "unfunded mandate." That's what the archive.is flap on en.wiki showed.
 * Bots can be used for maintenance, if there is something very clear that a bot can do, and if it is possible to review bot operation and stop a malfunctioning bot. Bots can be low-rate, allowing this review. (There is no particular need for a bot to be fast, unless there is a huge and very clear task to be undertaken.) In any case: until we have practical alternatives, Rotlink is performing a needed service. "Unauthorized Bot" is a red herring here. The real issue is content, how we want links and broken links to be handled, and, then, review of Recent Changes by those of us who watch it. I don't mind Rotlink because the rate is low enough here (averages on peak days, which are rare, one edit per hour) to allow monitoring of Rotlink activity and, in fact, to look at those links and pages and make links work better, I've already done some of that.
 * As I've said, if I see some drastic alteration in Rotlink behavior, I expect to look at it immediately, review global behavior, and possibly be at meta requesting global lock within minutes.
 * Of course, we could not stop a mal-intentioned bot in the hands of a sophisticated user, because the user would simply register new accounts. Most locked spambots are not sophisticated. They might as well wave a big red flag, "spambot." In this case, the behavior would not be distinguishable to ordinary users as spam or malicious linking. We'd whack archive.is quickly, though, with the global blacklist, if we found malware hosted there (i.e., directly by the site itself, there could be something problematic on an archived page, as there could with archive.org).
 * See where the page was a mess. I left the links as raw, which is common in print publications (where the whole link must be visible!), and we'd want that for a book compilation. Ugly, but useful and clear. Instead of focusing on Rotlink, who is helping, we may much more profitably focus on what we actually want for Wikiversity, and how to get from here to there. --Abd (discuss • contribs) 16:21, 30 August 2014 (UTC)
 * I've considered requesting the bot flag for Rotlink, but given that Rotlink is not particularly communicative, I'd rather not. I'd rather have those edits be reviewed, even though they are boringly good. It creates a kind of "random page" review, so there are other benefits as well. --Abd (discuss • contribs) 16:24, 30 August 2014 (UTC)


 * I have created a notice on User:Rotlink that Rotlink should be treated as a bot, and immediately blocked if operation moves out of established limits. No warning is needed (nor would warning be useful). I report there that consensus is, at present, to allow Rotlink to operate, based on the usefulness of Rotlink's work. Users should feel free to revert any Rotlink edits if they are harmful, and to Request custodian action if the bot begins to operate outside of safe limits. --Abd (discuss • contribs) 14:19, 6 September 2014 (UTC)

Rotlink operation conclusions

 * as proposer. --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)
 * I've checked the last ten link repairs by Rotlink, nine are between 8/25-8/29/2014 and the tenth is from 9/9/13. Each link was dead, either from the web accessible literature (8) or from Wikipedia (2), each was replaced with a link to archive.org, and each repair was correct. If Rotlink begins to operate outside this norm creating vandalism or unusable links then I believe a custodian should halt the bot or human pending investigation. --Marshallsumter (discuss • contribs) 00:30, 31 August 2014 (UTC)

Allow links to archive.is?
The general consensus on this has been yes, everywhere. Archive.is and archive.today are not globally blacklisted, are not blacklisted on en.wiki (where the close of the RfC mentioned above indicated a "weak support" for blacklisting), and AFAIK, are not blacklisted anywhere. Shall we allow links to archive.is and archive.today? (The alternative is blacklisting.)

Archive.is conclusions

 * as proposer. --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)
 * it might help the readers. JackPotte (discuss • contribs) 21:01, 29 August 2014 (UTC)
 * if the only links I can find for a dead link happen to be on archive.is or archive.today, I will use them. If Rotlink systematically shifts from archive.org to archive.is and this archive starts charging for downloads, then I believe solicitation has occurred and the bot or human should be halted, pending investigation. --Marshallsumter (discuss • contribs) 00:36, 31 August 2014 (UTC)

Allow Rotlink to add links to archive.is or archive.today?
Rotlink apparently has a Conflict of Interest with respect to those two domains. We have no policy prohibiting COI users from linking to their domain, if the domain is useful and otherwise allowable. Given that the Rotlink edits only flag dead links and provide a suggested solution, which any user may improve or revert, should we allow Rotlink to add such links? --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)

COI discussion
Noting that User:JackPotte's proposal above would resolve the conflict of interest issue. Users would be free to choose which archive they would like to view, and the visual clues make it clear that they are not visiting the original source. -- Dave Braunschweig (discuss • contribs) 21:25, 29 August 2014 (UTC)
 * JackPotte is free to implement his proposal, our approval is not needed, it would only be needed if that bot operates at high rate. I am not was not judging which is better, not having seen any example of what it does. Taking a hint from myself, I look at the French template and what links there and find permanent link, the template is used in Note 4. While that looks decent, the archive.org page that opens up is the URL page, not a specific link, and both snapshots are broken. The only link that seems to work at all is from wikiwix.


 * I don't see this as superior to what Rotlink is doing, but Rotlink probably doesn't check wikiwix. The problem here is that work is being set up for editors to do, and what happens on Wikiversity is that, too often, nobody gets around to it. Most readers would see that link and not know how to fix it. So, to see what is involved, I look at the wikitext. To start out, I click on the references section and, of course, don't see the code. I have to click on the ref number to find where the code is. I know already where this is heading! So I click on the number and I don't find the ref. That's because it's hidden by the donation request. (These ref jumps put the ref at the very top of the page, where they will be nicely hidden by the site message.)


 * Plus I need to know what to look for. It's a tiny highlighted ref number in the template on the top right. How the hell do I edit that? Okay, I have to edit the whole page. I was going to need to do that anyway, because it's the only way to see a prevue of references, and I don't want to blow my one chance to make a good edit to fr.wikipedia. Of course, the note numbers don't show in the wikitext, so, having been through this crap before, I first copy the date referenced, then edit the page, and search for it with my browser Find function, which didn't work. It's a template, so Find doesn't see the date, it's spread out in fields. Still, it wasn't a lot of text to search through.


 * About to substitute the wikiwix link for the broken link template, I realize I haven't checked to see if it supports the information. It does not. That link was to a googlebooks page that allowed searching a book by page, and the page was given. No wonder archive.is didn't have it! Total waste of time.


 * Summary: not so good. Better than a broken link and "dead link." Maybe. Not better than what Rotlink does, but maybe this was unfair as a test. (Rotlink would not have edited this.) Perhaps someone else can find a better example, I just looked at the first I found.


 * And then I found the original edit. No wonder nobody had fixed it since JackBot placed it January 1, 2013. Yes, JackBot=JackPotte. The bot appears to have mangled the note information, losing the actual book citation and I had to look way back to find it. I haven't seen Rotlink do anything like that, but nobody's perfect. I fixed it. --Abd (discuss • contribs) 02:29, 30 August 2014 (UTC)
 * And then, armed with the full book information from the original, and even though the page is now missing from the Googlebooks preview, I found a way to display the needed information, and added a link.. Old trick from an old dog. I did not put this link into the citation template, because, then, a bot might break the thing again if Google changes something, which they often do. Rotlink looks for dead links and replaces them, without changing anything else. In fact, I found one place where this wasn't optimal, but it was harmless, just didn't look good, the real problem was the original was an external link with the entire URL then placed again for display, so Rotlink replaced both. Rotlink really should not replace display-only text. But this is so rare that it's probably not worth pinging him. --Abd (discuss • contribs) 03:44, 30 August 2014 (UTC)

COI conclusions

 * as proposer. --Abd (discuss • contribs) 18:45, 29 August 2014 (UTC)
 * see my comments in the above support vote. --Marshallsumter (discuss • contribs) 00:39, 31 August 2014 (UTC)

Nomination of Dave Braunschweig for Permanent Custodianship
Candidates for Custodianship/Dave Braunschweig. All users are invited to comment there.

In the past, we have often site-messaged PC nominations. Dave, as a probationary custodian, should not edit the site message to show that the process is happening, my opinion, because of conflict of interest. However, another custodian, seeing this, may decide to site-message it, or, if there is community approval here for a site message, Dave could then go ahead and action it. So if you approve or disapprove of this being site-messaged, please comment! --Abd (discuss • contribs) 16:16, 29 August 2014 (UTC)

Shall the approval process be site-messaged?
It is commonly done. --Abd (discuss • contribs) 18:50, 29 August 2014 (UTC)

Site-message conclusions

 * as proposer, --Abd (discuss • contribs) 16:16, 29 August 2014 (UTC)
 * , this is usually done, but does this require a permanent custodian or bureaucrat? --Marshallsumter (discuss • contribs) 03:02, 30 August 2014 (UTC)
 * No, any custodian may edit the site-message. Ideally, the custodian is neutral. However, with a consensus here, or at least this proposal and no reasonable opposition in a decent period (I'd say three days), Dave could also site-message this. That message would be neutral and could actually attract negative votes, though I doubt that will happen. I'm just suggesting caution about recusal. In my view, it's better if this is in the site-message, but I don't want to start pinging all the custodians. Some might be on IRC. I don't do IRC, I'm allergic to it, I break out in hives. --Abd (discuss • contribs) 03:07, 30 August 2014 (UTC)


 * --guyvan52 (discuss • contribs) 02:56, 31 August 2014 (UTC)
 * ✅ - Thanks, Abd for getting this happening and discussing. -- Jtneill - Talk - c 02:47, 5 September 2014 (UTC)

Grants to improve your project
Greetings! The Individual Engagement Grants program is accepting proposals for funding new experiments from September 1st to 30th. Your idea could improve Wikimedia projects with a new tool or gadget, a better process to support community-building on your wiki, research on an important issue, or something else we haven't thought of yet. Whether you need $200 or $30,000 USD, Individual Engagement Grants can cover your own project development time in addition to hiring others to help you.
 * Submit your proposal
 * Get help: In IdeaLab or an upcoming Hangout session PEarley (WMF) (discuss • contribs) 16:09, 3 September 2014 (UTC)

Echo and watchlist
Special:Notifications & Special:Watchlist substantially overlap in functionality, except the former also contains extra (some non-public) events and doesn't provide with passive usage options (means to turn off web-nagging or email-nagging and to just keep visiting the page whenever I'm free), while the latter doesn't provide with options of active web-nagging notifications (but already provides email interface). Partly, in my personal view, the Echo/Notifications project was driven by low usability of watchlist; comes to mind. It's also perhaps worth noting that Echo users aren't exposed to Special:Notifications unless thy have JavaScript disabled — in which case it's their only means of reading the notifications.

I'd like to get this done: Thoughts on both, please?
 * 1) Merge these two pages into one.
 * 2) To remedy large inflow of information, introduce multiple levels of importance of the web-nagging notifications (red for mentions, orange for thanks, blue for new watchlist items, etc and configurable in your settings).

--Gryllida 02:23, 9 September 2014 (UTC)
 * By experience I'm sure that even if you post it on MW:Talk:Echo (Notifications) it will be complicated, unless you develop it yourself. JackPotte (discuss • contribs) 11:17, 9 September 2014 (UTC)
 * This is a "would we like it to be done?" discussion; means to get it done are a different beast. -Gryllida 22:07, 9 September 2014 (UTC)

Change in renaming process
''Part or all of this message may be in English. Please help translate if possible.'' Single-user login (SUL) finalisation's goal is so that every Wikimedia editor has a single, recognized global account with one username across all projects. As you may know, after a long delay, it's now underway as an effort between bureaucrats, stewards, and Wikimedia Foundation engineers. This will also allow for development of cross-wiki tools like global notifications and watchlists. There is no set date for the completion of single-user login finalization at this time.

The process involves changing all rename processes into one global renaming process. The ability for local bureaucrats to rename users on this wiki will be turned off on Monday, 15 September 2014, as one of the first steps. Global renamers are in the process of being created to make sure projects and languages are represented by the time this occurs. I sent a note to every bureaucrat about this process three weeks ago with an invitation to participate and many have begun requesting to be a part of the group. Together with the stewards, the global renamers will be empowered to help editors work through the often difficult process of getting a global name.

In parting, visit Special:MergeAccount to unify your account if you have never done so. If your local pages about renaming still need to be updated, please do so and consider pointing people to m:SRUC for future rename requests, especially if this project does not have bureaucrats that hold global renamer permissions. If you have any questions, you can read more on the help page on Meta. You can also follow the technical progress on mediawiki.org. Contact me on Meta any time with questions as well. Thank you for your time. -- User:Keegan (WMF) (talk) 9 September 2014 16.22 (UTC)

Is this a good naming policy?
I have been adding quizzes to Wikipedia and am wondering how to name them in Wikiversity namespace. One idea I had was to name it as a subpage to the article's name in Wikipedia. For example, Wikipedia has the article Saros (astronomy), so I put the quiz under  Saros (astronomy)/Quizzes , even though  Saros does not yet exist on Wikiversity. Is this strategy good/proper/permissible? --guyvan52 (discuss • contribs) 16:00, 15 September 2014 (UTC)
 * Well, it's not complete. Here are my thoughts. First of all, we probably don't need the disambiguated name, with "(astronomy)," here, but it's fine as a redirect to Saros, which should have a special note below the redirect link that it exists to match the Wikipedia name (otherwise it may be deleted with no local incoming links). On Saros create a stub, if nothing else, it would have a link to the Wikipedia article and then a link to the Quizzes subpage.
 * I don't think Saros is a big enough topic for a top-level resource here, I'd be happier if this were a subpage under a larger resource. This is an astronomy topic, a detail. I'd place this underneath a more general astronomy topic, if not Astronomy itself. Marshall may have some ideas. Right now, there is Astronomy and a link there could go to Astronomy/Saros which completely handles disambiguation.
 * We also have a stub on Orbital mechanics, which is a large enough topic for a stand-alone course in a university, hence, my thinking, fine for a top-level resource here.
 * So, based on the above thinking, create Orbital mechanics/Saros and move other similar resources to the same subspace. Then you'll have Orbital mechanics/Saros/Quizzes and the like. A link should then go to Orbital mechanics from the Astronomy resource. --Abd (discuss • contribs) 16:21, 15 September 2014 (UTC)


 * Once there are educational resources here, on a topic covered on Wikipedia, there should be sister wiki links placed there. Place w:Template:Wikiversity on the page, following the instructions in the Template documentation. --Abd (discuss • contribs) 16:29, 15 September 2014 (UTC)
 * I'm glad I asked. In a few hours I will build from the stub Orbital mechanics exactly as you suggest.  It's always good to turn a stub into something that is not a stub.--guyvan52 (discuss • contribs) 23:29, 15 September 2014 (UTC)
 * I moved all Saros (astronomy) pages into Orbital mechanics and its subpages. It's still chaotic and I will organize it better in a few days.  I also need to deal with the antikythera and ecliptic pages.  --guyvan52 (discuss • contribs) 04:06, 16 September 2014 (UTC)

Closing the custodianship vote for Dave Braunschweig
Candidates for Custodianship/Dave Braunschweig

Wikiversity Custodianship "proposed policy" stood as policy for most of the history of Wikiversity, the status as standing policy was removed in 2011, the argument being that there had never been a vote. Yet as a description of actual practice, the page has been accurate since as far back as I know. The core of it was in the original draft, by Sebmol, in August 2006.. The page was marked policy by JWSchmidt in February 2007 after extensive discussion. It is obvious that this designation was accepted by the community. In any case, it's what we have, and what we have been following since 2007 or before.

The page also provides for a one-month probationary period, which can be extended if necessary.


 * Dave was approved as a probationary custodian since 4 September 2013‎.

As there was no mentor recommendation forthcoming, after repeated requests and a year passing, I informed the mentor, Jtneil, that I would go ahead and make a recommendation. He thanked me for that, and the voting was opened 29 August 2014

The "policy" page provides for one week of evaluation of the vote for permanent status. It provides that a bureaucrat make the final decision. (The original draft policy had five days as did the version first marked as policy. I changed it to one week when it was still policy, and that change stuck.)

The vote was site-messaged on 5 September 2014‎.

On 12 September, I messaged all bureaucrats requesting a close, it having been two weeks since the voting began. There has been no response.

Community consensus is clear in the voting. 10/1, stable for 10 days after the last vote, is more than enough on every wiki to gain administrative status. Consensus, in fact, trumps policy, and we don't have policy except as, effectively, a proposal and a tradition, and the absence of bureaucrats was not contemplated. I am, accordingly, closing that discussion, even though I'm involved and am not a bureaucrat. I am claiming no special authority, only acting to implement community consensus. --Abd (discuss • contribs) 00:36, 19 September 2014 (UTC)

Dedicated Programming Compiler
This is to inform all those who may be interested in participating at any level that I have created a proposal for the Grants:IEG initiative advertised above. The proposal is entitled: "Dedicated Programming Compiler" and can be found at url=https://meta.wikimedia.org/wiki/Grants:IEG/Dedicated_Programming_Compiler#Measures_of_success. Please feel free to include yourself at any participatory level. Also, if you personally know a WMF dedicated mainframer who has sufficient background installing or firewalling computer langauge compilers, please let us know so that we can see if they might be interested. There may by real money in this! --Marshallsumter (discuss • contribs) 18:59, 20 September 2014 (UTC)

I teach a college physics course called How things work. Most of my students are non-science majors who are required to take a certain number of college courses. "Back in the day", such courses were referred to as Physics for Poets, but students these days are more career oriented. I have a lot of freedom to choose topics, but we are supposed to include at least one topic related to computer technology. Learning a simple program would be perfect. And there are advantages to doing it on an obscure language like Ada because it would be the second (or third) language for some of the more savvy students, but a first language for most. This puts them all on an equal playing field (to a point). Anyway, I am interested. You might have noticed my Wikiversity activities on Python and MATLAB. That was in part due to "How things work".--guyvan52 (discuss • contribs) 21:05, 20 September 2014 (UTC)


 * Great! Thank you for responding. There are three sections in the proposal where you are welcome to add your user name depending on your personal preferences and likely time constraints: Budget, if you'd like some cash for your hard work, should the proposal be accepted; Participants where your bio here would fit in well, and/or Endorsements, where the points you've made here for having such a dedicated compiler are most welcome. --Marshallsumter (discuss • contribs) 23:01, 20 September 2014 (UTC)
 * I can endorse as an educator who believes strongly in open source teaching and uses it almost exclusively in every course I teach. I can participate by writing educational material for students using the on-line compiler.  I can also task college engineering students with developing simple and well-documented codes to college engineering students  (I like to learn a new language by starting with such codes.)  Leave a note on my user talk page if you need me urgently.  Otherwise we should perhaps continue the discussion here in order to generate more interest within Wikiversity.--guyvan52 (discuss • contribs) 11:12, 21 September 2014 (UTC)


 * Would it be okay with you if I transferred a copy of your entry above [beginning with "I can"] to the proposal's Endorsements section. The only modification I would make is a v: in front of "User" so that meta knows it's from Wikiversity. --Marshallsumter (discuss • contribs) 15:43, 21 September 2014 (UTC)


 * BTW, do you know of any good online compilers for Python and/or GNU Octave (the MATLAB clone)? I could sure use one.--guyvan52 (discuss • contribs) 11:12, 21 September 2014 (UTC)


 * The web site Download Python lists several. You've hit on an important part of why the above proposal (and program) is needed. There is also Nuitka a Python compiler. It's fully compatible with Python 2.6, 2.7, 3.2 and 3.3. And, if you have CPython, Python source code is automatically compiled into Python byte code by the CPython interpreter. Then there's PyCompile, a mini python compiler which allows you to forget the command line. And of course, they have separate forms for Windows PC and Imac. Just the beginning. The age of your operating system is also a factor. --Marshallsumter (discuss • contribs) 21:51, 21 September 2014 (UTC)


 * For anyone interested, I've copied the current discussion at meta regarding this proposal into the discussion page. Comments, criticism, suggestions, welcome. --Marshallsumter (discuss • contribs) 02:38, 23 October 2014 (UTC)


 * Evaluation and comments on this proposal from the committee have now been received. These along with their numerical scores (1-10 point range for each criterion) are now posted on the the discussion page. Comments, criticism, or suggestions are most welcome. --Marshallsumter (discuss • contribs) 22:47, 16 November 2014 (UTC)

Lots of Questions on how to start
I've started me a little project, have a basic page ready but with some issues. How and where can I ask in an approved manner some help to solve them? ChrisVanBommel (discuss • contribs) 09:59, 21 September 2014 (UTC)
 * Somebody please delete this if I am wrong, but I think a good place to start is to create the page User:ChrisVanBommel/sandbox. Just click that (red)link and start writing! --guyvan52 (discuss • contribs) 11:19, 21 September 2014 (UTC)
 * I would prefer placing that context into your userspace. Such as User:ChrisVanBommel/[Your Project Name]. Myself, I wouldn't do a sandbox, since that is for testing wikitext and some other stuff related to that. But if you don't know the name of the Project yet, go ahead and do it in the sandbox. Have fun! --Goldenburg111 13:27, 21 September 2014 (UTC)
 * As others have mentioned, the best way to start is by starting. The best place to ask questions for now is here.  Once we know what your questions are, the discussion may continue here, on your talk page, on someone else's talk page, or on the talk page for the project itself, depending on factors more related to whomever helps than any approved manner.  What issues do you have?  -- Dave Braunschweig (discuss • contribs) 14:22, 21 September 2014 (UTC)
 * Thanks for the replies, I'll start with it in my sandbox https://en.wikiversity.org/wiki/User:ChrisVanBommel/sandbox ChrisVanBommel (discuss • contribs) 20:27, 21 September 2014 (UTC)

Learning project 'work smarter'.
I would like to start a new learning project with the name: 'work smarter'.
 * Who wants to join me?
 * What best practiced could be described?
 * I put some info on my user page: https://en.wikiversity.org/wiki/User:Timboliu/To-do/work_smarter. What are the minimum requirements before we can open up a learning project in top level.

Tim, Timboliu (discuss • contribs) 06:08, 23 September 2014 (UTC)


 * It seems to me that 'work smarter' could have many different interpretations, depending on the subject matter. There are some efficiency techniques that would apply to a variety of fields, but others are much more specific.  The video referenced on your page is very specific to software development and agile methodology.  If that's the 'work smarter' that you have in mind, then it should probably be a subpage of System development.  It could also extend Agile software development.
 * What I personally look for in a top-level project is something that would comprise a course or at least a lecture or activity on a specific subject. Something that helps me define a top level project is to look at Wikipedia titles as a guideline.  In this case, there is no 'work smarter' article on Wikipedia, so that topic either isn't specific enough as to what it will address, or it isn't the accepted terminology for that subject.  Alternatives in this case might be 'personal effectiveness' or 'business process management', depending on the direction a 'work smarter' project might take.
 * At the same time, it's important to note that Wikiversity isn't Wikipedia. Wikipedia is an encyclopedia of content.  Wikiversity is based around either learning projects or research projects.  See Learning projects and Research for more information.
 * One other thought. It is preferred to use internal links for Wikiveristy pages, such as User:Timboliu/To-do/work smarter rather than external http or https links.  Using internal links helps identify 'What links here'.  -- Dave Braunschweig (discuss • contribs) 13:54, 23 September 2014 (UTC)
 * Hi Dave, thanks for your feedback. I agree that when you look at the Spotify video you would say that this learning project could be placed in learning project system development. I added some other subjects to explain that learning project 'work smarter' is broader then just system development. About external linking. You say that it is preferred to use internal links, but this information is not internally available. Timboliu (discuss • contribs) 13:56, 27 September 2014 (UTC)
 * That's correct, Timboliu. How about making it "internally available"? I.e., it is available here, but not in some guideline. Whose job is it to put it in a guideline? User:Somebody else, too often. Using internal links is an aspect of working smarter here, but where would this information go? While "work smarter" could apply to many fields, the topic is a common one in business. I.e., business system development. In general, I'm looking toward the organization of Wikiversity so that it will be relatively obvious -- compared to now! -- where to place resources and learning circles or projects. While it is possible to connect anything with anything, keep it simple. Techniques for working smarter could be useful in personal development, in daily life, etc., but what will be the bulk of the study? We don't want the same page duplicated everywhere, or do we? I.e., underneath many fields could be a "work smarter" learning circle. Alternatively, there can be cross-links. But start somewhere! What I don't want to see is a hundred learning circles that years later we see went absolutely nowhere. When you start a learning circle, start learning! Share what you find. You have tended, in the past, to share almost nothing. I.e., "contacted so-and-so at such-and-such a company." Fine. What happened? Was the contact a conversation or you just sent an email and called that a contact? Are there facts you can report? Opinions communicated? Just that you lifted a finger doesn't create something to share. Results! Please! If you watch that video, how about summarizing it on a subpage? If you learn, yourself, and document it, you are creating something of value, starting with value for yourself. Documented learning is generally deeper learning. --Abd (discuss • contribs) 16:01, 27 September 2014 (UTC)
 * Hi Abd, thanks for the feedback. Work smarter is now a resource under learning project Business. I also added a to-do to make an abstract of the video. Regards, Timboliu (discuss • contribs) 08:02, 28 September 2014 (UTC)

1900
A number of the resources having a year title have a boilerplate as their only entry. A few have red-linked subpages. These are part of a project by KYPark. The user hasn't edited since October 2013. While I sincerely hope the user continues to contribute, what should we do about the empty or red-linked only boilerplate pages? I put three up for speedy delete or rescue, but there may be a lot of these. Random hit on 1900. Comments, criticism, help requested. --Marshallsumter (discuss • contribs) 14:56, 28 September 2014 (UTC)


 * i've reviewed a few of these now. It appears to ultimately be a single learning project on literature (author, year published, sometimes details or comments) with separate pages for each year.  I would be inclined to move all of it to subpages of the Literature project and have a subpage for 'Literature/By Year' that uses the existing template to link to the other pages at the same level.  -- Dave Braunschweig (discuss • contribs) 22:09, 28 September 2014 (UTC)
 * Thanks for bringing this up, Marshall. This, of course, never should have been in mainspace this way. It's possible that the study of literature by year could be of value to a student. So I'd propose moving all this project as suggested by Dave. As subpages are involved, this should be done by an admin and we might want cleanup by bot. So it would be, as an example, 1900 would be moved to Literature/By year/1900. There is no problem with redlinks, they indicate that KYPark intended to list those authors there. I have not reviewed the templates. I'm getting a bit cranky about templates that can make maintenance and organization more difficult. I removed the speedy deletion templates from several pages, we should simply organize this material. --Abd (discuss • contribs) 00:22, 29 September 2014‎


 * Sounds like a good plan! I checked my contributions and I put a speedy delete on four, just in case you only found three. If I'm counting them correctly there are at least 24 pages plus categories combined. --Marshallsumter (discuss • contribs) 00:33, 29 September 2014 (UTC)
 * Yes, thanks. There is work represented in collecting and placing the redlinks. They do no harm, so we keep them. I'll check those speedies, yes, I only saw three. Thanks for your work cleaning up Wikiversity. Mostly, we do it by organization, unless something is truly useless or harmful. --Abd (discuss • contribs) 00:41, 29 September 2014 (UTC)


 * You found all four. There were 1900, 1901, and 1902, plus 1900/input.


 * I put some of the pages Random finds up for speedy delete rather than fixing them myself because I like some one else to look at these. I hope that's still okay. Using random is fun of a sorts. But it also makes more work for others. I can suspend for awhile if you and Dave need a break. --Marshallsumter (discuss • contribs) 02:38, 29 September 2014 (UTC)


 * I'd say to keep using random looking for things to clean up. But try using some of the other templates when appropriate.  For example, there's, , , and .  If a page has incoming links, it's often not a speedy delete.  Thanks!  -- Dave Braunschweig (discuss • contribs) 03:46, 29 September 2014 (UTC)

I've been looking into this and it's not going to be as simple as moving 24 pages. There are actually more than 400 pages involved, and they are intricately interwoven across themselves and several other learning projects. To get an idea of the scope of this effort, take a look at Special:WhatLinksHere/Template:Navigate20c and Special:WhatLinksHere/Template:Cite_plus. It's going to have to be a combination of moves and clean-up on each page. Moves by themselves will break everything. -- Dave Braunschweig (discuss • contribs) 03:46, 29 September 2014 (UTC)


 * The categories probably can remain untouched for now. Moving each year to Literature/By year/1900 or Literature/1900, eg, would leave redirects. The tough ones are like 1911/Welby which links to 36 other pages. Ideally, this should become Literature/1911/Welby. If I only move 1911 to Literature/1911, the redirects should work and the matrix should not be broken. Shall I try it? If I encounter any unused boilerplates with no red-linked names, I could put these up for deletion or a move according to this plan. What do you think? --Marshallsumter (discuss • contribs) 15:29, 29 September 2014 (UTC)


 * I'm not a big fan of redirects unless it's something the average person would search for and find helpful. Having 1911 in the search index and having it bring you to a literature page doesn't seem helpful to me.  Having 1911 redirect to the Category:1911 seems much more useful.  I think I like Literature/1911 better than Literature/By year/1911.  It's probably much easier for me do these moves and updates by bot than it is to do them manually.  In theory I can do the whole thing in under an hour once I figure out what code I need.  I also realized this morning that it will be far easier to update the template(s) than it is to update the pages individually.  I'll see if I can make progress on this later tonight.  -- Dave Braunschweig (discuss • contribs) 18:32, 29 September 2014 (UTC)


 * Outstanding job on these Dave! It looks like you and your bot handled this matrix with ease. Those put on a subpage for KY Park should facilitate his continued efforts. --Marshallsumter (discuss • contribs) 05:13, 30 September 2014 (UTC)


 * It wasn't too bad, although it took 1,100 edits or so. And at 500 edits my IP address was blocked for about 15 minutes.  It must come up on some screen a steward has to verify manually.  There are two related projects that still need cleanup:  Tables, and History of pragmatics.  -- Dave Braunschweig (discuss • contribs) 12:56, 30 September 2014 (UTC)

mul: interwiki now exists
It quietly happened in the mediawiki release last week (git #f0d86f92 - Add additional interwiki links as requested in various bugs (bug 16962, bug 21915)) that the mul: interwiki has appeared for the wikiversities and wikisources to be able to point to beta.wikiversity and wikisource.org respectively. So we can now do mul:Main page and it appears in interwiki links as "More languages". We will have a number of places, especially in the project namespace where we need to do some additions. — billinghurst  sDrewth  04:11, 29 September 2014 (UTC)
 * Yes. This means, instead of typing  you can type  . When you type   without the leading colon, you get a language link in the sidebar, labelled "More languages" and pointing to the relevant page on Beta Wikiversity. The latter feature was mainly designed with Wikisource in mind, so I don't know if you'll be using it here; however, you can look at WV:Community Portal to see a "More languages" link in action. This, that and the other (discuss • contribs) 11:02, 29 September 2014 (UTC)

What is this module?
Rene and me developed the MOOC module. It is a combination of a Lua module together with a lot of JavaScript code, in order to increase the usability. The module has a short documentation and is already being used in the first part of the Web Science MOOC.

At the moment this MOOC module can not be used by everyone, since you need the JavaScript files. We inject them by using a resource loader that you currently have to merge with your common.js manually, as described in the documentation. It searches the DOM tree for a certain node that contains the wiki titles of JavaScript/CSS files it should load.

The MOOC module is a prototype. There may be an extension in future, that will make most of the JavaScript code unnecessary, but this is the future. For now this is the only module that enables users to host a MOOC at Wikiversity in an appropriate way that we are aware of.

What do we want?
Our request is to merge the resource loader into the global 'common.js' in order to make the MOOC module work for everyone out-of-the-box.

Rene discussed this with Dave Braunschweig, who suggested to talk to User:Darklama. However, this user seems to be inactive at the moment since Rene got no reply to his request he made nearly 5 weeks ago.

What is necessary?
As mentioned, we need the resource loader to be available in the global 'common.js'. It would also suffice to have the MOOC script enabled per default. We hope a custodian is able to do so or to trigger whatever is needed. We are aware that the resource loader is a potential security risk and tried our best to take this into account: It is limited to the User:Sebschlicht namespace, but we may not be aware of a way to work around this limit. This has to be reviewed by someone who is more familiar with the vulnerabilities here.

The script files should be moved to another namespace other than mine. This makes it harder for us to maintain (change) the code, but ensures to have a stable and reviewed version running all the time. We kept them in the user namespace, because the quite fine editor is available in user namespaces only.

Where are the source files?
Lua:
 * Module:Mooc
 * Module:Mooc/*

uses templates:
 * Template:Mooc/Overview example page: start page of Web Science MOOC part 1
 * Template:Mooc/Lesson example page: ethernet lesson in Web Science MOOC
 * Template:Mooc/Unit exaple page: collision detection unit in Web Science MOOC

JavaScript:
 * User:Sebschlicht/mooc.js loads CSS and the MOOC interface that allows on-page edits in modal boxes and loads the discussion script.
 * User:Sebschlicht/moocDiscussions.js aggregates the talk pages of a MOOC unit and enables on-page discussion.

CSS:
 * User:Sebschlicht/mooc.css contains a lot of CSS code that makes the template look nice and keeps the design maintainable.

We appreciate your feedback and hope, that the community is willing to support this project. --Sebschlicht (discuss • contribs) 13:42, 25 September 2014 (UTC)