Forum:Implement a visual editor

Forums: Index → Proposals → Implement a visual editor

This is a closed discussion about a failed proposed change on SmashWiki. It remains for archival purposes. The closing administrator adds: "New information on this makes this idea much harder to implement, to the point it outweighs the benefits. Serpent  King 18:44, May 2, 2020 (EDT)"

Many people in the Smash community have knowledge to contribute to the Wiki, however not all of these people are experienced Wiki editors. While most people have a working knowledge of Microsoft Word and other similar word document applications, only those who regularly edit Wikis will be familiar with the formatting language of Wikis. Furthermore, some may even be familiar with Wiki editors, but are not too fond of the skeletal layout of the text box and the constant need to refresh the preview in order to see what their contributions would look like. Some may have trouble visualising the Wiki article, for various reasons, and some may simply want an editing format more similar to the word document interfaces they know, love, and are familiar and comfortable with.

As a Wiki, it is in our best interest to accommodate our users, so long as we do not inconvenience other users or harm the Wiki in the process. Some Wikis, including the largest public Wiki on the internet, Wikipedia, have a feature which enables their editors to switch to what is known as a visual editor. Visual editors display the article being edited, as it is being edited, and allow editors to see what their edits to the article look like in real time. Different Wikis have different ways of implementing visual editors, but Wikipedia’s take is an excellent example of a completely non-intrusive and rather user-friendly visual editor option in action:

For new users and existing users who have not changed their editor setting, it will open up the traditional editor by default; however, in the top right of the editing window is a small box with a pencil in it, which can be clicked on to switch to the visual editor. The pencil remains at the top of the page, which can be clicked on to switch back. By default, the Wiki will remember what editor you used last, and load the same editor (and thus, for those who have never switched editors, it will always load the classic editor). By going into preferences, users can also manually set the Wiki to always load either the classic or visual editor by default, regardless of which was used last, and even provides an option for having 2 edit tabs appear at the top of the page (where the page, talk, history etc tabs are), so that editors can spontaneously decide which editor they want to be opened, depending on which editor they decide would be best for them to use for the edit in question, and switch between the two without clicking on the pencil.

In essence, the visual editor is an accessibility tool (an area in which SmashWiki is somewhat lacking) which some users may prefer over the default, and we believe that a Wikipedia-style visual editor should be implemented on SmashWiki as well. Before voting on this proposal, please consider the following points:

• The addition of a visual editor will provide editors with more control over their editing experience.
• It will help to accomodate for existing and potential users who for whatever reason may not be fond of the source editor.
  • This is not meant to discourage inexperienced editors from learning how to do complex edits through source editing, but is simply providing a way for them to get the same results they otherwise would without even needing to learn all that to begin with; if they wish to learn how to do complex edits through source editing, the option is still there for them. Otherwise, this is just an accessibility tool to make it easier for them.
• Its implementation will not in any way inconvenience or disturb editors who wish to continue using the source editor, as that will remain the default, and unlike a certain other group of Wikis, users will not be prompted or encouraged in any way to use the visual editor instead.

Users who are not familiar with what the visual editor looks like are also advised to please test and play around with Wikipedia’s visual editor BEFORE voting on this proposal. Furthermore, please do not let certain other Wiki groups’ poor implementation of visual editors influence your decision on this proposal, as Wikipedia’s implementation of the visual editor will be the basis for its implementation on the Wiki, should this proposal pass.

VoqéoT 12:39, May 1, 2020 (EDT) & Alex the Weeb 12:42, May 1, 2020 (EDT)

Update 1

Regarding the timing, and feasibility of the addition of a visual editor, Porplemontage (the Wiki's owner) has said the following:

"I'm going to take a look at implementing visual editor as a non-invasive option after we upgrade to MediaWiki 1.35 around September"

Please do not base your vote on the technical details of when and how the editor would be added onto the Wiki, as that will be handled by Porplemontage. Alex the Weeb 14:04, May 1, 2020 (EDT)

Support

1. As per above. VoqéoT 12:39, May 1, 2020 (EDT)
   2. As per above. Alex the Weeb 12:42, May 1, 2020 (EDT)

   New information has recently come to light which reveals that implementing a visual editor on the Wiki would be far more difficult than was originally thought. As a result, I am redacting my support for this proposal for the time being. It may be the case that at a later date implementing a visual editor will become more feasible, but for now... Alex the Weeb 15:35, May 2, 2020 (EDT)
   

2. At first I was neutral about this addition, but after playing around with it for a while on Wikipedia, I realized it could potentially be really beneficial to the wiki even for experienced editors. I find that editing and adding on to templates is a lot less time consuming with the visual editor. It would make power ranking, tournament, and smasher pages a lot less time consuming. Definite support from me. Pokebub (talk) 13:26, May 1, 2020 (EDT)

Oppose

1. The visual editor used on Wikipedia is currently in beta. When a full release occurs, I'll change this to Neutral, but for now oppose. DekZek 13:44, May 1, 2020 (EDT)

   This isn't a proposal to introduce the visual editor right this instant. Porplemontage has said he will look into the implementation of a visual editor when the Wiki is updated to the latest version of MediaWiki. The purpose of this forum is to establish whether or not there would be support for actually implementing it here. Do not base your vote on the timing of the forum. Alex the Weeb 13:52, May 1, 2020 (EDT)

   But that is independent of the visual editor itself being in beta status. For all we know, it could still be in beta in September. DekZek 15:00, May 1, 2020 (EDT)

   Wait, I misread your comment. If implementation is conditional about the editor itself being out of beta, then I'd change my position to Neutral. DekZek 15:50, May 1, 2020 (EDT)

1. People are going to complain that I sound rude and disparaging for saying this. But I am strongly of the opinion that there should be a skill floor for the editing of anything but basic text. If you want to learn to edit a wiki, you should learn to edit the wiki, not learn to edit a façade of the wiki. And if you don't want to learn, there's no learning needed to just edit the raw text of articles, which is the bulk of all content that a fresh user would be trying to change.
   In addition, I have absolutely zero faith in any sort of WYSIWYG interface. I am constantly fighting "smart" editors (not here, but elsewhere) to achieve specific results that a well-meaning editor that doesn't edit source blows away by accident. A non-wiki example (which is a bit extreme but still an example) is a thing that automatically fills empty table cells with <p><br/></p>, even if the user explicitly wants the cell to be without these things. It's not on. I do not trust visual editors to keep manually-kerned source untouched. Toomai Glittershine The Spark 17:16, May 1, 2020 (EDT)

   You evidently haven't read over the proposal and other comments very well, or you read them and ignored what they said. Aside from the fact that suggesting that people who aren't confident at using source editors don't have valuable information to contribute to the Wiki is quite frankly asinine, the purpose of the visual editor is not to circumvent learning to edit a Wiki, but to provide a second method of doing so. This has been explained multiple times, but you either haven't noticed any of these explanations, or are committing an ad nauseam fallacy. As for your concerns about editors inserting garbage strings into the source, from what I can tell Wikipedia doesn't seem to have this problem, and the visual editor program that would be implemented will be the same one that is used there. Alex the Weeb 17:28, May 1, 2020 (EDT)

   "suggesting that people who aren't confident at using source editors don't have valuable information to contribute to the Wiki" I have never said this, and you would do well to not put words in my mouth. What I actually said was "there should be a skill floor" and "you should learn to edit the wiki [...] not [...] a façade". I am of the belief that anyone who wishes to contribute to the wiki should demonstrate a base level of competency with it. Not a high bar by any means - literally just being able to edit text while leaving syntax alone is fine - but a bar nevertheless.

   "provide a second method of doing so" Correct me if I'm wrong, but the visual editor is designed for articles and simple templates. It is not capable (and nor should it be) of correctly handling the requirements of complex templates. Users who have learned the "easy" method will be forced out of their comfort zone and have to re-learn a lot of things if they ever wish to participate deeper in the wiki. I am of the opinion they should just learn the "complete" way the first time. Toomai Glittershine The Honcho 22:35, May 1, 2020 (EDT)

   "But I am strongly of the opinion that there should be a skill floor for the editing of anything but basic text. If you want to learn to edit a wiki, you should learn to edit the wiki, not learn to edit a façade of the wiki."

   Ok, but why? If the visual editor can achieve the same results source editing otherwise can achieve (such as Wikipedia’s visual editor), why is this a problem exactly? Both methods of editing can achieve the same exact end result, but the visual editor would simply just be easier to use for inexperienced users; you're basically just cutting out the middle man of needing to learn it to begin with, which I see no problem with so long as the results are the same. And again, nothing would stop them from learning how to source edit if they so please; the only difference would be that they wouldn't of necessity need to learn how to to begin with in order to contribute, and I have no problem with saying that. VoqéoT 18:09, May 1, 2020 (EDT)

   Let me give you an example of why I do not like these editors. I went to Wikipedia, opened the [[VisualEditor]] page, selected the infobox, and deleted 4 parameters that were already set to blank (changing nothing else). By switching back to source mode, it was revealed that the actual code was dramatically changed - no difference in the output, but a huge difference in the code's spacing and ordering that a non-savvy editor will never see (and those who review edits will have a harder time understanding and deciding whether this is a good edit). (The exact same change done to the raw code looks like this.) And this is just an infobox, not a more complex template such as those we use for building tables. I hope this helps you understand why I am deeply distrustful of these things. Toomai Glittershine The Honcho 22:35, May 1, 2020 (EDT)

   That is interesting how that at works like that, but I guess as long as the end result is the same, I don't still don't see this as a big problem if things aren't getting messed up. However, I'm not technically inclined on all this to know the implications of what that could entail, and perhaps this is something Porple can look into, if possible, if this does end up being a problem. But again, if the end result is the same, I just personally don't think it's a big problem. VoqéoT 05:06, May 2, 2020 (EDT)

   On reflection, I didn't give enough prominence to one of my points so I'll re-iterate it: One of the premises of this proposal is that experienced users do not need to care because they can keep doing their own thing. But imagine that most new users start doing this, and their edits suddenly become much more difficult to quickly judge - compare the four-line diff to the many-line diff, and think of how much more work it is for others to decide whether this edit is okay or it needs to be undone/rollbacked. So no, this proposal will not be invisible to experienced users, and will in fact increase the workload of those who patrol for bad edits. Toomai Glittershine The Eggster 10:50, May 2, 2020 (EDT)

2. I've edited Wikis for more than half of my life, and have always known how to use the editor. RickTommy 23:01, May 1, 2020 (EDT)

   As mentioned in the proposal, source editing would still remain as the default. The visual editor would simply be optional to whoever wishes to use it. If you don't personally want to use the visual editor, you would not be forced to at all. VoqéoT 05:06, May 2, 2020 (EDT)

3. Widely oppose Apparently the visual editor would require us to put metadata on every single one of our templates (infoboxes, nav templates, shortcut templates, moveset tables, layout templates, ...everything) describing what the parameters of it are. As Toomai said, there is a very steep effort-to-payoff ratio which seals the deal for me. This one is a no go in my eyes now. Serpent  King 14:56, May 2, 2020 (EDT)

Neutral

1. Quite frankly, I don't care. I don't think it hurts too much to have it on the wiki, nor would it hurt too much if we don't have it on the wiki. CookiesCreme 13:01, May 1, 2020 (EDT)
2. Echoing C&C. I've already laid out my opinions through Discord, and they mostly remain the same. But if things are easy to edit, even with the visual editor, then I'm not against it. Aidan, the Rurouni 14:44, May 1, 2020 (EDT)
3. For the reasons listed in my previous oppose comment.DekZek 15:50, May 1, 2020 (EDT)
4. I'll give a probably weak opinion and say I personally do not like it. I've used it and it just doesn't work out for me, I'd much rather use the source code. That said, I've only used Wikia's, which is a mess to work with (to go along with other messes Wikia goes through). I'm more leaning toward oppose due to personal taste, but if Wikipedia's is actually useful, then it might not be a problem. Alex95 (talk) 18:28, May 1, 2020 (EDT)
5. This is completely optional so I see no harm adding it in. If this is good and can help new editors learn how to edit then I will meant towards supporting. @Alex95 I don't believe using Wikia as an example is a good reason to oppose. We're not Wikia, we are our own independent Wiki so we can make ours the way we want. OmegαToαd64 • the Best Kαrter 18:35, May 1, 2020 (EDT)
6. I dont care if this is added to the wiki. I will say, that it will be easier to make edits on this wiki page. Although, vandals will be able to clear the page way slower than faster or not. So Its really up to the admins to decide this. S3AHAWK (talk) 21:13, May 1, 2020 (EDT)

Comments

VisualEditor does reformat template parameters to a specified format. Wikipedia generally uses block for their infoboxes, but it looks like if the format is left unspecified, VisualEditor will leave the param spacing alone. This got me down a different rabbit hole though, which is that if we were to implement visual editor, in order to do it "right" we would also want to implement TemplateData and add metadata to each template describing the parameters. The reason being is that without TemplateData, VE's template editor isn't nearly as user-friendly, and since the whole idea is to make things easier, it probably should be. I would want to either go all-in or forget it, having a feature implemented half-well isn't great imo. But adding template data to each template makes it a much bigger project, so now I'm thinking well who is going to want to add template data, that's a lot of work with an upside that most will never see. --Porplemontage (talk) 10:56, May 2, 2020 (EDT)

I was wondering if this sort of thing was a factor. I agree that requiring templates to have metadata has an extremely skewed effort-to-payoff ratio. Toomai Glittershine The Rainbow 13:01, May 2, 2020 (EDT)

Oh wow...yeah no that's a nail in the coffin for me. Serpent  King 14:52, May 2, 2020 (EDT)

I'm not an admin but I agree with you. S3AHAWK (talk) 16:25, May 2, 2020 (EDT)