User:GrammarEnthusiast/Regarding VisualEditor

Note: This content was originally posted on my main user page 7 August 2013. It was moved here 11 September 2015.

Regarding the present and future use of VisualEditor on English Wikipedia

(The following comment also appears at the official Request for Comment (RfC) page regarding whether, when, how, and to whom VisualEditor should be displayed by default, under the section Discussion of anonymous editors' default state.)

I feel strongly that VisualEditor must be completely disabled for new (not-yet-confirmed) and anonymous users, at least until the major (page-breaking, and anything that leads to an intended edit messing up something on the page to the extent that the intentional constructive edit doesn't do better than merely balancing it) bugs are worked out. It doesn't matter if inexperienced editors use, like and/or prefer VisualEditor so long as edits they make using it are likely to damage the articles!

Yes, I know that the stated purpose of VisualEditor is to make editing easier specifically for new / newer users. (I created my account less than two years ago and have fewer than 100 edits, but I did enough IP editing previously that I probably shouldn't count as a "new" user...) With the software being as buggy as it currently is (I don't need to point to specific bugs, since many other contributors have done so at length, and VisualEditor's supporters are well aware of the mountain of bug reports) VisualEditor cannot meet its stated goal of encouraging and retaining new editors. As a compromise, new and anonymous / not-logged-in users could be informed that there is a VisualEditor option for regular users, and be offered a sandbox page to try VisualEditor out with; surely that would accomplish the goal of having VisualEditor encourage new users to make their first edit and then make future edits as well, without needlessly abetting those whose only intent is vandalism.

In my opinion, obvious vandalism, such as page-blanking or the insertion of obvious nonsense (e.g. "BOOBS BOOBS BOOBS BOOBS BOOBS"), is less of a problem, in terms of its impact on the average non-editing user of Wikipedia, than subtle vandalism that can easily be overlooked as vandalism by experienced editors and non-editing users alike. Consider, for example, an instance of vandalism I just discovered and corrected http://wiki.riteme.site/w/index.php?title=Socotra_Starling&diff=prev&oldid=567518200 which dated to February 2011 (before I even created my editor account) and which had survived ten intermediate edits http://wiki.riteme.site/w/index.php?title=Socotra_Starling&action=history by 8 legitimate editors (albeit mostly bots). The vandal's account was blocked shortly after their edit to Socotra Starling, and all their other vandalism under that account had already been reverted or otherwise corrected, but this one managed to fly under the radar (pun intended) for two and a half years. Vandalism of this sort harms Wikipedia's usability — and reputation — far more than more obvious kinds of vandalism; the latter type of vandalism is just as easily accomplished with VisualEditor as with standard markup, but the former in many cases would be significantly easier with VisualEditor, particularly on longer articles, in which such small, subtle acts of vandalism are more likely to go unnoticed. While it's true that a vandal motivated enough to come up with 'sneaky' ways of vandalizing an article is also more likely to be motivated enough to figure out how to do so via wikitext, such vandals are not necessarily motivated enough (or tech-savvy enough, though plenty of computer novices seem able to learn wikimarkup just fine) to do so. And I submit that dissuading those types of vandals is an important enough goal to be considered with equal weight alongside the question of whether VisualEditor access will encourage new users to start editing and continue editing.

For the non-new users among us, the current interface compromise (or at least what I'm currently seeing) shows

I have to agree with the voters above who pointed out that the terminology "Edit source" is both unclear and likely to cause confusion of Source text with Source code in many users' minds. I think that if both options are going to be presented at the top of the page for any users, better terminology would be

"Edit (VisualEditor^beta) | Edit (wikitext)"

(ordered alphabetically in this example, but I don't feel strongly either way as to what order the two options should be ordered in if both are to be shown). I considered calling the latter "Edit (wikimarkup)" but since there's a strong movement for VisualEditor to recognize at least some wikimarkup — particularly double-square-brackets for intralinks — that has the potential to be needlessly confusing.

Some experienced Wikipedians on both sides of the divide say they prefer to have only their choice of editing option show up at the top of every page. Others, especially those who see benefits to both options and would prefer to switch back and forth on a regular basis, will want both displayed at all times. The solution to this disagreement seems simple to me; regardless of whether a (logged-in, confirmed) user clicks on "Edit (VisualEditor^beta)" or "Edit (wikitext)" or simply "Edit", I propose a brief announcement at the top of the page to this effect:

"You are editing the page [article title] using the [whichever] interface. [link to the other edit interface|Click here] to use the [the other] interface. If you do not want to see this message displayed again, [link that will hide editor-selection announcement|click here]. You may also select your default editing interface and show or hide this message via your Preferences page."

Couldn't that please everyone without drowning the donkey?

FWIW, I have used VisualEditor to edit one article, back when it was abruptly enabled as default. I figured (unaware at the time of why VisualEditor had become the default or the controversy surrounding its introduction) I'd try it, and if it didn't work as expected, then I'd figure a way around it. It seemed to work well enough for the relatively simple changes I made, though it made a change to how one of the article's reference citations was coded for some reason; the change doesn't appear to me to have broken the citation, but I still haven't learned the citation system well enough to be 100% sure. But I vastly prefer the "good old" wikitext editor. To be fair, I should disclose that I also prefer to edit HTML "by hand" rather than with a WYSIWYG editor, but I'm only conversant with fairly basic HTML, and haven't learned javascript, CSS, HTML5 or any of the other fancy new stuff. My preference for "under the hood" editors (I use the HTML options on sites like Tumblr and WordPress as well) has at least as much to do with a desire for precision as with the ease of using what's familiar, however; I'm not opposed to change simply because it is change.

I welcome any users (registered, confirmed, anonymous or brand-new) to engage in discussion with any of the points I raised above, either on my talk page, or at the official Request for Comment (RfC) page.