Talk:Collection Extension 2/Flow

Previous page history was archived for backup purposes at Talk:Collection Extension 2/LQT Archive 1 on 2015-07-10.

hi! This looks really great; thanks for pulling this together. The things you have outlined are really useful in thinking about how we could go about approaching this. A couple initial thoughts:

1. Branding/Labeling: I think it is important to make it clear that this tool is particularly of use for the those who need/want to read stuff offline. To me, "Read Later" doesn't necessarily signify that, but it could. I don't have a solution, but I want to make sure we are thinking about this and making sure it is clear. (Maybe this requires some research, see point 3.)
2. Functionality': I really like the new way it is all laid out! This is SO much cleaner than the current process, which, to be honest, I even at times navigate. One thing I would suggest adding to your wireframe is the epub/mobi functionality, which I think is important to offer. Also, for the "Reading List," it would be useful to have the option of sorting the article list alphabetically, too. I know for me as a user this would be important.
3. Research: The one gap that we really need to work on is knowing if this might be of impact. On the tracking side of PP, it is imperative that we at a minimum know the traffic and downloads that are being directed towards the collections extension, which currently isn't a system that is easily in place. This way we can see if changes are actually making an impact. On the UX research, do we know if these new designs would resonate with individuals? Would they be easier to use? I'm not sure if you did any research like this prior to launching the initial book creator, but I think it is really worth investing in this upfront so that we can truly ensure that this feature would be most helpful - for as you mention in the piece up front, we know that people WANT this tool!

Thanks again for all this, Christoph! Great work to you and your team; I truly believe this could make an impact on a lot of minds!

This post was posted by Peachey88, but signed as Jwild.

Hi Jessie, thank you for your comments. Before we start implementing, we can do some comparative user research. It will be probably quite easy to show an improvement, but it might be hard to quantify it. As a first step we could do some usability tests with click-dummies or paper prototypes. We could test some variations on button labels: "Read later/offline" might be an alternative worth testing.

I'm really really late to read this proposal, but I want to share my opinion about it with you:

• "Read later" seems also to me to be the best labelling in comparison to the others...
• I fully support the new workflow, nothing against.
• The main problem we have now, is that the tool is not prominent enough... so it has to be more prominent to be more used.
• Epub format support seems to me to be a must "have"

What is the status of this project? Kelson (talk) 22:33, 5 August 2012 (UTC)

Hi! The status was totally stalled. No one expressed demand, and so it wasn't budgeted for or prioritized :(

=(

Sad. This project is, like Wikipublish, necessary to improve or e-book generation toolchain. How much is needed for the implementation?... hope this project will not be forgotten. In any case, I won't. Kelson (talk) 08:30, 8 August 2012 (UTC)

Maybe ePub can drive interest for this project as well (It will become available soon on Wikipedia). Personally, I would love to continue with this.

Hi! I usually use the collection extension, and I think it's a great solution to save/print books or single articles. The drawback is that the extension is quite difficult to configure and to mantain; for example in theese days I upgraded my MW from 1.17 to 1.19.1 and Collection from 1.5 to 1.6.1 as explained in the conllection extension page but it stop working, and nobody support it neither answers the quistions. So, having a more complete (with ebook etc) and reliable Collection 2 would be really amazing. What should we do to encourage and support the developement of this project ?

Hi!

I see the analysis considered only (English) Wikipedia, but I think it is important to keep in mind that the kind of material (articles) produced on Wikipedia is quite different from what is created on Wikibooks and Wikisources (mainly Books). There are discrepancy also in the kind of readers and editors of these projects.

Considering that the first version of the extension focused on providing Wikipedias some way to create "books" (simple collection of isolated articles), the result was a tool which is not very integrated with the wikis where, by definition, the editors create real books (e.g. most of the reported bugs and feature requests in this direction are still open, partially because MediaWiki doesn't provide a way for users to "organize a set of pages as a book" - hopefully this will change soon).

On those (non-Wikipedia) projects, the need for "collections" of pages (as opposed to PDF versions of single pages) is a lot greater, despite the current lack of integration, and this need could make a difference in the numbers when comparing the proportion between PDFs of single and multiple pages.

When revisiting the "book metaphor", please take into account this Wikibooks' blog post where it is discussed the confusing nomenclature we had at hands (on non-English projects as well) when Collection extension was enabled on Wikibooks projects.

A new tool to export as EPUB has been created for Wikisource. It is called WSExport. It is running currently on the Toolserver (see discussion here). Probably it should become a Mediawiki extension or maybe share some base code with Collections 2. Is it anyone who could help to make an extension out of it the same way as happened with Extension:Proofread Page? The main developer says he has no time to do this right now.

This would be great! I will help spread the word to try to find someone...

Ok, I met with Howie on the product team and he had some good suggestions on where to start. Web placement for the "Read" button will be the toughest to figure out, but in the meantime, we could really start with (1) improving the single page PDF making experience and (2) the collections creation experience. (1) This page is quite cluttered on en.wikipedia; perhaps we could eliminate some of the items along the bottom (similar to what it is like if I were to use the PDF extension on mediawiki.org), and make the "Download the File" more prominent? I'm sure there are other things that would help (like making the "Back to article" more obvious), too. (2) The collections creation comes after #1, and you build out a good idea for this one! How long do you think these things would take? If we start with redoing just the PDF experience, it would be useful to see how much more use the tool actually gets (following all the way to downloads). Jwild 19:27, 1 February 2012 (UTC)