VisualEditor/Průzkum 2015

From mediawiki.org
This page is a translated version of the page VisualEditor/Survey 2015 and the translation is 31% complete.

Tým Editování žádá o vaši pomoc při vylepšování VisualEditoru. Tento tým má svůj seznam problémů s nejvyšší prioritou, ale rád by se dozvěděl i o drobných problémech. Tyto problémy mohou editaci znepříjemnit, zabírat vám příliš mnoho času, nebo tvořit ono „stokrát nic“. Může se zdát, že malé problémy nestojí za to hlásit jako chybu, ale tým Editování chce znát a zkusit opravit i tyhle věci.

Kontaktují několik rozdílných typů uživatelů a žádají vás, abyste se podělili o své myšlenky kliknutím na tento odkaz.

Tento rychlý, jednoduchý anonymní průzkum provozuje Qualtrics. Můžete odpovídat ve svém vlastním jazyce. Pokud se průzkumu zúčastníte, souhlasíte, že vaše odpovědi mohou být využity v souladu s těmito pravidly. Tento průzkum provozuje Qualtrics a jejich užití vašich údajů popisují jejich pravidla ochrany soukromí. Pokud máte otázky nebo byste raději odpovídali na wiki, nechte nám zprávu v diskusi. Celý obsah průzkumu je uveden níže.

Informace, které poskytnete, budou použity k vylepšení funkcí a vzhledu VisualEditoru. Odkaz na konečné výsledky přidáme na tuto stránku.

Obsah průzkumu

Úvod

Tento průzkum provádí nadace Wikimedia, aby se dozvěděla více o zkušenostech členů komunity s VisualEditorem na Wikipedii a sesterských projektech. VisualEditor je editor formátovaného textu, který umožňuje editovat Wikipedii a další projekty bez předchozího studia jazyka wikitextu.

Tyto snímky obrazovky byly vytvořeny z článku „Galaxy“ na anglické Wikipedii. První snímek (vlevo) zobrazuje článek s otevřenou editací ve VisualEditoru. Druhý snímek (vpravo) zobrazuje stejný článek otevřený v editoru wikitextu.

Souhlasíte, že vaše odpovědi mohou být použity v souladu s těmito pravidly. Tento průzkum provozuje Qualtrics a jejich užití vašich údajů popisují jejich pravidla ochrany soukromí.  Informace o licencích a autorství snímků obrazovky najdete na https://en.wikipedia.org/wiki/Galaxy.

Otázky

  1. Pro začátek: do kterého projektu nejčastěji přispíváte?
    • Wikipedie
    • Commons
    • MediaWiki
    • Wikiknihy
    • Wikidata
    • Wikizprávy
    • Wikicitáty
    • Wikiverzita
    • Wikicesty
    • Jiný projekt nadace Wikimedia
    • Běžně nepřispívám do žádného projektu nadace Wikimedia.
  2. Jak často používáte při editaci článků VisualEditor? (na škále 0–100 %)
    • Nikdy (0 %)
    • (10% to 90%)
    • Vždy (100 %)
  3. Jak pravděpodobné je, že byste VisualEditor doporučil(a) novému přispěvateli? (na škále 0–10)
    • Vždy bych doporučil(a) editor wikitextu. (0)
    • (1 to 9)
    • Vždy bych doporučil(a) VisualEditor. (10)
  4. Zaškrtněte vše, co platí:
    • V minulosti jsem pravidelně používal(a) VisualEditor, ale teď většinou používám editor wikitextu.
    • Zkoušel(a) jsem používat VisualEditor, ale často jsem stránku opustil(a) bez uložení změn.
    • Někdy začnu editovat ve VisualEditoru, ale pak přepnu na editor wikitextu.
    • Poskytl(a) jsem na wiki svůj názor na VisualEditor na stránkách typu VisualEditor/Feedback.
    • Nahlásil(a) nebo sledoval(a) jsem chyby v Bugzille či Phabricatoru.
    • Nic z výše uvedeného
  5. Oblastí, která ve VisualEditoru nejvíce potřebuje zlepšit, je…
    • Přidávání nebo editace citací či referencí
    • Obrázky a další média
    • Obecné editování
    • Podpora pro jiné jazyky než angličtina
    • Přidávání nebo editace šablon
    • Výkon a rychlost
    • Vytváření a editace odkazů na jiné články a jiné webové stránky
    • Kopírování a vkládání
    • Diskutování s jinými uživateli
    • Přepínání tam a zpět mezi VisualEditorem a editorem wikitextu
    • Nastavování a měnění kategorií
    • Nabídnutí dvou editačních tlačítek namísto toho, aby si software pamatoval, které používám raději
    • Řešení editačních konfliktů, aniž bych musel vidět wikitext
    • Vizuální zobrazení rozdílů verzí a historie článků
    • Vzorce, klikací mapy, časové osy a další typy speciálního obsahu
    • Stabilita a spolehlivost
    • Jiné
  6. Rád bych v této oblasti (vybrané v předchozí otázce) viděl(a) zlepšení. Pokud bych v této oblasti mohl(a) změnit jednu věc, změnil(a) bych…
  7. Jak jste se se o tomto průzkumu dozvěděl(a)?
    • Zpráva v mé uživatelské diskusi
    • Zpráva v uživatelské diskusi někoho jiného
    • Z jiné stránky na wiki
    • E-mailem
    • Jinak
  8. Nepovinné: Pokud budeme mít dotazy ohledně vašich komentářů či návrhů, můžete sem napsat své uživatelské jméno, abychom se s vámi mohli spojit. Pokud chcete, můžete přidat označení wiki, na které vás nejlépe najdeme. (Například: Wikipedista:Příklad na cs.wikipedia.org)

Results

1,639 invitations were sent, including some duplicates due to contacting a few dozen people via e-mail (from Bugzilla accounts, which are not associated with wiki accounts). As of 17 April 2015, 508 surveys had been completed in Qualtrics and two on wiki.

First, to which project do you contribute most often?

There were 501 responses to this question.

Project Number Percentage
Wikipedia 463 92%
Commons 7 1%
MediaWiki 7 1%
Wikibooks 5 1%
Wikidata 6 1%
Wikinews 1 0%
Wikiquote 3 1%
Wikiversity 4 1%
Wikivoyage 1 0%
Another project sponsored by the Wikimedia Foundation 3 1%
I do not normally contribute to any Wikimedia Foundation projects. 1 0%

How frequently do you use VisualEditor when you are editing articles?

There were 452 responses to this question.

The mean was using VisualEditor for 40% of article editing, and the standard deviation was 30%. The answers were mostly evenly distributed, with clusters among people who almost never used VisualEditor and also editors who almost always used it.

# Answer Response %
1 Never 70 15%
2 10% 73 16%
3 20% 32 7%
4 30% 36 8%
5 40% 28 6%
6 50% 33 7%
7 60% 28 6%
8 70% 34 8%
9 80% 45 10%
10 90% 51 11%
11 Always 22 5%
Total 452 100%

How likely are you to recommend VisualEditor to a new editor? (on a scale of 0 to 10)

There were 449 responses to this question.

The mean was 60% with a standard deviation of 30%.

Answer Response %
0 36 8%
1 19 4%
2 21 5%
3 15 3%
4 22 5%
5 55 12%
6 32 7%
7 38 8%
8 71 16%
9 54 12%
10 86 19%
Total 449 100%

This answer correlated very strongly (p=<0.0001) with whether the respondent uses VisualEditor frequently.

How frequently do you use VisualEditor when you are editing articles?
Never 10% 20% 30% 40% 50% 60% 70% 80% 90% Always Total
Likelihood
of recommending
VisualEditor to
a new editor
0% 25 5 2 0 0 1 0 1 0 1 0 35
10% 7 10 0 1 0 0 0 1 0 0 0 19
20% 2 10 2 1 2 3 1 0 0 0 0 21
30% 6 4 2 2 0 0 1 0 0 0 0 15
40% 3 6 5 3 4 1 0 0 0 0 0 22
50% 8 17 3 8 6 6 1 4 1 1 0 55
60% 2 4 5 3 4 4 3 3 3 1 0 32
70% 3 3 4 6 3 3 5 5 1 5 0 38
80% 5 8 5 7 5 6 6 9 14 4 1 70
90% 1 3 0 3 3 2 3 4 11 23 1 54
100% 3 3 4 2 1 7 8 7 15 16 20 86
Total 65 73 32 36 28 33 28 34 45 51 22 447

Choose all that apply:

There were 453 responses to this question.

In the past, I used VisualEditor regularly, but now I mostly use the wikitext editor. 10%
I have tried to use VisualEditor, but I often left the page without trying to save my changes. 21%
I sometimes start editing in VisualEditor, but then switch to the wikitext editor. 61%
I have provided feedback about VisualEditor on wiki at pages like VisualEditor/Feedback. 31%
I have reported bugs or subscribed to bugs in Bugzilla or Phabricator. 21%
None of the above 17%

In VisualEditor, the area that needs the most improvement is...

There were 433 responses to this question.

The order of the items (except "Other") were randomized, so that different people saw different items at the top of the list. The most popular choice was "Adding and editing templates", followed by "Switching back and forth between VisualEditor and wikitext editor", "Adding or editing citations or references", and "Performance and speed". Responses were correlated with how much the participant uses VisualEditor. People who use VisualEditor for most of their editing were most interested in improved support for citations, such as the automatic reference filler Citoid, rich copying and pasting, and special content types, such as mathematical formulae. People who never or rarely use VisualEditor were more interested in performance and reliability as well as general or non-software issues; they were comparatively uninterested in template or citation support within VisualEditor.

Most important area for improvement
Answer Response %
Adding or editing templates 73 17%
Switching back and forth between VisualEditor and wikitext editor 48 11%
Adding or editing citations 46 11%
Performance and speed 45 10%
Other 43 10%
General editing 39 9%
Stability and reliability 35 8%
Formulæ, imagemaps, timelines or other special content types 21 5%
Support for languages other than English 15 3%
Copying and pasting 14 3%
Showing me diffs and the history of articles visually 12 3%
Images and other media 10 2%
Handling edit conflicts without showing me wikitext 10 2%
Making or editing links to other pages and other websites 10 2%
Setting and changing categories 8 2%
Having discussions with other editors 3 1%
Having two edit buttons rather than the software remembering which I like to use 1 0%
Total 433 100%

If I could make one change to this area, I would change...

There were 300 responses to this question.

Overall: Editors who use VisualEditor frequently have different concerns compared to those who use it rarely or never. Among those who commented, VisualEditor was considered superior for writing, copyediting, and editing tables (except for tables generated from templates).  Several complex workflows, especially involving templates and citations, were areas of concern for people who use VisualEditor dozens or hundreds of times in a month. Editors who rarely or never use VisualEditor are more interested in the quality of Parsoid's conversion of the HTML that VisualEditor uses into wikitext, in the ability to switch to the wikitext editor, or general performance. Some users would like to see it enabled in talk namespaces or in the project namespace. A couple of respondents used the suggestion field to say that other editors should not be allowed to use VisualEditor, or that using VisualEditor would block the personal growth of new editors.

User interface: Some of the issues reported stem from a lack of familiarity with VisualEditor's features. For example, although a majority of respondents said that they sometimes switch from VisualEditor to the wikitext editor, a few believed that this feature did not exist. A few others believed that it was not possible to insert or edit new templates. These features may not be sufficiently prominent, or editors may benefit from a GuidedTour of the interface to point out the location of these items, such as the one proposed in phab:T89074. Several editors said that making and editing internal and external links is confusing. The Design Research team has also identified this as an area of concern in user testing before the survey started, and improvements are being tested now.

Citations:  Most of the concerns about citations are being or will be resolved by the Citoid system, by better local TemplateData, and/or by improvements to the template system overall.  Some changes, such as the ability to name citations, would need to be made in VisualEditor.  Some other suggestions, like better support for page numbers and nested citations, may need to be handled through changes to the Cite extension. Requests included support for wikis' {{reflist}} templates. Templates that contain or display references can be inserted and viewed in VisualEditor. However, their contents do not update during an editing session, because this would significantly harm performance. Changes to citations immediately appear if the page is using MediaWiki's faster, simpler, native <references /> tag; however, that tag does not currently support column formatting. Respondents also recommended being able to edit a citation by clicking on the citation in the references list, a planned VisualEditor feature. 

Local issues: Some editors said that VisualEditor is less convenient because parameters they commonly use are not present in, or not correctly prioritized in, the TemplateData.  The options automatically presented inside the template dialog are controlled on-wiki by the community, rather than in VisualEditor. For example, in a {{cite news}} template, whether the newspaper parameter is added by default depends upon whether that parameter is marked as either "suggested" or "required" in that template’s TemplateData on the local wiki.

Performance:  Lost-work bugs and slow performance were significant concerns for some editors, especially editors who have older or slower computers, or who teach editing to people who do.  Performance has been improved significantly since the survey started, and more work is planned, which should address these concerns.

Other recurring requests:  

  • Better special character support, especially for inserting pairs (e.g., German-style quotation marks); met since the survey was started.
  • Better support for copying and pasting citations between articles
  • In situ editing of the contents of templates like infoboxes
  • Bi-directional switching between editing environments
  • Support for Internet Explorer 9
  • Fewer "dirty diffs" (awkward changes to wikitext that do not affect the page's appearance)
  • The ability to upload images to Commons, or the local wiki, within VisualEditor 
  • Section editing 
  • Built-in spell check

Except for the last two, most of these are already planned or are being strongly considered. Section editing does not reduce edit conflicts and paradoxically would not improve loading times in VisualEditor. Instead, the team is working on a long-term plan making it possible for editors to modify individual pieces of a page, like a table or a paragraph, rather than whole sections at once. The team is not planning to provide built-in spelling or grammar checking, as it is unlikely that a spell checker inside VisualEditor could work as well as or as quickly as the ones editors already have in their web browsers.

How did you hear about this survey?

423 people answered this question. 82% said that they received an invitation on their own talk pages. 9% received an invitation through e-mail. 6% found the link to the survey on someone else's talk page. 2% chose "Other", and about half of them said that they received the message either on wiki or in e-mail, e.g., e-mail messages to the translator's mailing list. One person found the link on another wiki page.

Languages

Almost all invitations were delivered in English. Almost 40% of the survey respondents took the survey in a language other than English. People who took the survey in English, French, or German were more likely to have used VisualEditor regularly in the past, but not currently. People who took the survey in English or German were more likely to have exited without saving or to switch to the wikitext editor during an edit. Participants taking the survey in English were much more likely to have provided feedback or reported bugs.

Outcomes

Newly tracked bugs

  • T97323: Improve tabbing in template dialog