User:PamD/VEessay

I've hitherto more-or-less limited my input to describing bugs - albeit many of them are what are termed "enhancements", ie requests to regain some of the functionality hitherto available which is lost in VE. I see I've been in the top 10 contributors to this page, so perhaps it's time for something of an overview.

I'm with the many others who reckon VE is not yet ready to be made the default editor for new and inexperienced editors, registered or IP. There are too many facilities missing. It seems likely to lead them into making bad edits, which other editors will have to fix after them. I'm sure that the problems already identified - including our "enhancement" requests - provide enough work for the development team for the foreseeable future, without the need for more eyes to find more bugs yet.

My ideal scenario at this point would be: new and IP editors use Edit Source by default, but are given a prominent announcement on the lines of "A new Visual Editor is under development but not yet bug-free: if you would like to use it instead of the long-established Text Editor, click HERE. If you have any comments on the new editor, please click on the FEEDBACK button." Existing editors get the choice but are told clearly how to choose whether VE or Edit Source is their default editor. Announcements about MAJOR step-changes in VE are made to all editors, using Echo (brilliant suggestion someone made above), so that they can make an informed choice to switch to use it at a future point when it's good enough.

I can see that "bugs" which will actually crash the system or totally mess up someone's editing session have to be a priority, but many important aspects of the editor experience are "enhancements": areas where VE does not yet match the editing experience we had with the old Text Editor plus our various gadgets, browser-dependencies, etc, and where the developers' response sometimes seems pretty negative.

Things where VE seems dangerous include:
 * Hidden comments are not visible to editors - things like "Please do not make significant changes to the lead without discussing them first on the article's talk page." at the start of London, and also the templated messages like Use British English.
 * It's very easy to delete content like infoboxes and templates, without seeing what you've done. Especially things which appear on screen as line breaks.
 * Adding references is appalling. We want to encourage new editors to add references: the old "RefToolBar" system was pretty good: choose book/web/newspaper etc, then fill in relevant boxes, then preview the ref, then add it. The VE system is a nightmare: nothing like enough guidance to point editors in the right directions. And when you add the reference... it doesn't appear in the article as displayed in VE, so you can't tell whether you got it right or not and are likely to assume it's all gone horribly wrong.
 * In various ways (including the above) VE is NOT yet WYSIWYG: it's disconcerting enough for an experienced editor when changes I make don't appear instantly online, or when the VE display of a page mangles its existing layout (eg coordinates which appear top right in the displayed page suddenly display, in VE, on a line near the foot of the article): the editor's reaction is "What's going on, what have I done wrong, how can I fix it, do I try this... that... oh heck... stress, stress....". Not good for new editors.

Some of the things which make VE editing sessions harder work than Edit Source, or mean that I make less good edits, are: No blank lines, two templates on same line, generally grotty to read and edit.
 * Edit summaries - no longer the autocomplete/dropdown menu (apparently provided by the browser while edit summaries were one-line). So editors who do a lot of repetitive work can no longer easily create long, useful, linking, edit summaries (eg "Stub-sorting ([[Wikipedia:WikiProject Stub sorting|you can help!"). In VE I just add "Stub-sorting", no longer the link to the project: too much like hard work.
 * I am accustomed to being able to mouse over a link and see the page it links to: invaluable in checking whether the links on a page go where they should, or to dab pages or worse. Particularly useful for those of us who work a lot with disambiguation pages. Navigation popups are a Gadget, so not included in the spec of VE, it seems.
 * I can't see the article while I'm adding categories or stub templates: both the "page data" and the "transclusions" boxes hide the article text. But if I'm adding stubs or categories, I need to be able to see the article (When was she born, for a birth cat? How do we spell that district name for a geo-stub?). Sometimes I'm resorting to opening a second copy of the article in another tab: ridiculous.
 * I can't see the categories while I'm editing in VE: I don't know whether I need to open "Page settings" to add one or more cats, or whether they're already there. I can't click on a category to work upwards from it to find the relevant stub template.
 * Red links show as blue, so some aspects of disambiguation page cleanup are going to be impossible - and there are other situations too where knowing that a link is red makes me look harder for, and often find, the intended target: can't do it in VE.
 * The En.Wiki rules of layout, as in WP:ORDER, are ignored by VE: stub templates don't appear after categories. This means that pages edited in VE are more difficult for later editors to edit manually, as stuff isn't where it's expected to be. Order of events at the top of the page is also pretty suspect: I doubt that the accessibility-oriented rule of putting all hatnotes above all other templates is within VE's capabilities. The sort of tidying up which was AWB "general fixes" or the good practice of Twinkle's habitual behaviour is now lost.
 * VE's ideas of layout in general are not what most human editors would do: I made several additions in one edit of Çilikya which ended up looking like:

I'm struggling on, using VE most of the time (but occasionally just going straight to Edit Source when I know it'll make life a lot simpler, or want to copy and paste a ref from another article, etc). I don't think VE is yet good enough to make it the default editor for our newest recruits. It has massive potential, but it's not ready. I'm doing my best to help, by carefully pointing out bugs, as are many other editors (though it's a pity that one of the sharpest minds on the project, who figures on the list of feedback page contributors, is currently banned for a month). But please concentrate on getting more of the bugs/enhancements fixed, bringing VE up to the standard we ought to be offering to new editors, and revert the decision to make it the default at for new and IP editors, postponing this until it's a better product. Pam D  22:03, 16 July 2013 (UTC)