An annotation tool (Community Wishlist/W416)
Why an annotation tool?
[edit]Many of our readers are students, or people who are trying to learn about something (is not always the same). One of the things many students make when they are studying is annotating text, and yellow markers are a symbol of the process of studying itself. The use of Wikipedia by students is very evident in smaller languages, where the peaks of visits happen during school days. Our own internal processes, where articles are constantly changing, make it difficult to remember where you read something, so learning from a dynamic web page is more complex than doing it with text-books. Nevertheless, many languages lack of text-books, and schools are using Internet, chromebooks and other digital assets more than ever. That's why creating an annotation tool could help students, but also casual readers, to learn more. Furthermore, it could boost the process of account creation, as it gives a perk for creating one. Nowadays, having an account doesn't give any inherent advantage over not having one, so we should test if we can give something to logged-in users that can't be accomplished by non-logged in users.
Currently, creating an account on Wikimedia is not something useful if you are not trying to edit Wikipedia itself. There are no perks, no benefits of doing it. That could explain why we have lots of accounts created, but few of them ever engage with Wikimedia itself. If we could have a tool that makes the Wikimedia experience better only if you are logged-in, that could be a good reason to create an account. We don't know if account creating would lead into more edits, but it seems a good assumption.
How to annotate?
[edit]Annotating the web is not so usual, but there are some tools to do it. The one we looked into when we designed this process was annotator.js, which is open code (but seems stalled now): http://annotatorjs.org/. This tool allows to simply select a text and saving the highlight. The solution seems simple, but the implementation is not so simple, as we couldn't find a way to create all the sanitized code that would be needed for Wikimedia deployment (and that's the main reason for our idea to be abandoned).
Where would we save the annotation?
[edit]Here comes an interesting tweak in this project. The obvious answer would be "our namespace". We could just copy parts of the text into subpages inside our namespace and that's it. But... wouldn't it be even better if we could create our own Notebook? And the good news is that we have a project just for doing that: Wikibooks. We could just save personal books at Wikibooks with our annotated texts, ideas and so on. We were thinking on that idea when we found that is technically impossible to save a page inside our own namespace in another project. Is to say, I can create a button to save something at /User:MY_NAME/Whatever, but I can't do it at OTHERPROJECT/wiki/User:MY_NAME/Whatever, because there's no way to send that via any magic word (and it would be great if there is a way and we didn't find it).
This idea would reduce the burden of subpages created on a Wiki, and would make the Wikibooks project more visible, especially for education, which I think is a good move.
What else could be done?
[edit]There are, at least, a couple of things that could be done.
- The first one is adding a button not to annotate, but to discuss a section of the text. This would make talk pages more visible (which are not so much used outside English Wikipedia), the process of article creation more transparent and "non editors" more engaged with the article. Once you are discussing the nuances of something, it seems that you could start to do *other things*. There could be some social problems with this, especially on pages with high traffic, but it's something that could be a good design experiment.
- The second one is a little bit more outside of our goal, but could be interesting for the AI team. Imagine that lots of people are using the tool, and you have sections highlighted by many users. We could know (not if we save it only at Wikibooks) which sections are more interesting, have more attention or are more relevant. This information could be used to summarize articles, by hand or automatically (I know lots of data is needed for this).
Unassigned
Readers