Page MenuHomePhabricator

Wikimedia Shop should be translatable to all languages
Closed, DeclinedPublic

Description

The Wikimedia Shop is currently only available in English. Ideally the shop would support translation of the product contents, but at least the interface should be internationalized and translatable.


URL: https://meta.wikimedia.org/wiki/Talk:Wikimedia_merchandise#Translation_of_the_shop

Details

Reference
bz37797

Event Timeline

bzimport raised the priority of this task from to Medium.Nov 22 2014, 12:27 AM
bzimport added projects: Wikimedia-Shop, I18n.
bzimport set Reference to bz37797.

It would be quite easy to translate it to many languages if it was opened to edit. putting it as depends on bug#37798 .

This is separate from bug#37798 because the current underlying software does not allow full translations (even if openly editable you would just be changing english to X language and english would no longer be available).

There are ways that we could get the 'normal' part of the shop to be translated but even with that hack are unable to translate the checkout at all (it can only be one language per store front period). I think that having half the store translated but not the payment pages would be an even worse experience then having it all in English.

We've been looking at other (open and PHP) options that would allow us to translate it in a more complete way but requires moving to new store software (or building our own). I don't think this is at all out of the question but we're going to need to wait at least another month or two and test out the demand on the public before we can justify the resources to do that.

Obviously with our culture it's incredibly important long term to be able to handle multiple languages (as many as possible) and that's going to remain important especially before we roll out any kind of international publicity. As we're starting out however we're going to focus on the English version and not pushing it past the English projects because of this. We aren't looking to do this "a year from now" though we're hoping to make any changes needed much sooner as soon as we can justify it and get the resources.

This appears to be a very English language centric view, James, that is hard to connect to 4 out of 5 strategic priorities of the Wikimedia Foundation: Increase participation, Improve quality, Increase reach, Encourage innovation.

Are the plans still to not look into this before the fiscal year 2013-2014?

I apologize if my statement was un clear, I think it made much more sense in my mind with my intonation then it did on paper. I did not mean to imply we would not be looking at it before fiscal year 2013-2014 I was attempting to say that we would DEFINITELY attempt to do it well before that. The current software challenges make it difficult to do it at a large scale (as many languages as we would like as quickly as we would like) but there are lots of options that we are looking at including multiple interconnected shops and additional software options and leveraging the chapter store connections that we want to make as we expand as well.

My personal goal at the moment is to have at least 1-2 new languages up and running before the holiday season (at the same time as a European fulfillment center).

(In reply to comment #4)

The current software
challenges make it difficult to do it at a large scale (as many languages as we
would like as quickly as we would like) but there are lots of options that we
are looking at including multiple interconnected shops and additional software
options and leveraging the chapter store connections that we want to make as we
expand as well.

I'm curious, how abstracted is the payment processing/handling of the site from the browsing part? If it's reasonably abstracted, setting up a wiki might be easiest here. Wikis support a proper Vector skin, they support galleries of products, they support internationalization, they come with revision control and access control, etc.

I know we've discussed using MediaWiki (and possibly an extension) for the Wikimedia Shop, but I'm curious just how much work is involved (roughly). Is the code that runs the current Shopify shop public? How easy is it to take data from a standard MediaWiki wiki and pass it to Shopify's shopping cart? What are the other hurdles?

Or to think about this another way: does Shopify have other customers who have their own sites where Shopify only acts as the backend?

Also, can you share the assessments that you surely have produced to choose it in the first place, to understand it better? They don't need to be polished in any way, not to be made public if you don't want to, as long as they're available for those interested (chapters come to mind for instance).

James: Any chance to answer comment 5 and comment 6, please?

[I'm decreasing priority as fixing is definitely something to achieve, but not the very first item on a list of urgent things.]

Thanks for the poke Andrew,

@MZ: The shopify code is sadly not completely open or available. Pieces of it (https://github.com/Shopify ) including much of their payment process ( https://github.com/Shopify/active_merchant ) is but it really only lets us write a new payment processor for them which they can then integrate if they want.

In the long run I would really like to use either a Mediawiki extension or something like Magento which is an open source php shop system (both free and paid/hosted open source but the free version is still pretty darn good) . If we use Mediawiki I think the easiest payment system is going to be piggy backing onto our fundraising processing software that already integrates well and is fairly scalable for something like that. Both options would be much better for translations.

@Nemo: Yeah, the original decision to start with Shopify was done before my time in the role but I have very unpolished notes that I made near the start and since then that I'm happy to share in writing and/or phone/email conversations about my thoughts. The simple version is that there are multiple other options that could be available (including some that I would prefer like Magento and MW) but the ease and simplicity of starting with shopify meant it made sense to continue along this path for now given our resources. They are very good for 1 language (any language really) and are very plug and play.

I think we need to increase sales to really justify the resources that would be required to do something more complex then shopify but I'm working on that and heading in that direction. I think it will happen.

On a side note: I expect to have at least one language opening up next month. I'm going to start the translation process on wiki as soon as I can (I have not yet finalized what that first language will be but have it narrowed down to a couple at least).

(In reply to comment #9)

I think we need to increase sales to really justify the resources that would be
required to do something more complex then shopify but I'm working on that and
heading in that direction. I think it will happen.

This is a comma 22. [Chapters are not allowed to make shops on the grounds that the WMF does, but] the WMF doesn't sell to those countries because it sells only in English, and will not support those languages because their countries are not buying enough.

Can we please stop finding excuses to provide services only in English? It's getting old.

Proper product selection and thinking ahead goes a long way. This story makes me really sad. More sad than would be appropriate to write down here, frankly.

(In reply to comment #11)

Can we please stop finding excuses to provide services only in English? It's
getting old.

I started some notes here: [[m:Best practices in evaluating new software]]. I think documenting the expectation here will help.

Just an update on this bug — I have updated the merch FAQ to show our current stance, and welcome further discussion on the talk page. I'd like to recommend we close this bug.

Thanks all for your attention and input.

(In reply to Caitlin Cogdill from comment #14)

I'd like to
recommend we close this bug.

I personally would prefer leaving it open. Priority, whiteboard, and other fields can be tweaked to reflect your prognosis/plans. (which you can then sort/filter on)

Other bugs are sometimes kept open for extraordinary periods. (see e.g. Bug 1, Bug 189)

Where are the higher priority items you mentioned tracked? (e.g. redesign, shipping improvements) I guess some of those deserve their own bugs if they don't already exist.

This bug will stay open. When the "in-house design that should make it more easy to display alternate languages" happens, we'll close it.

Okay, that's fine. I'd just like to ask the conversation moves to Meta so it can more exposed to the community at large and contained under the other shop discussions.

Okay, but seriously, what's wrong with using MediaWiki for this?

As I understand it, in order to accept donations to the Wikimedia Foundation, we've already built a robust payment processing system that includes full internationalization support. Why can't we use that system to accept payments for merchandise?

The gallery of goods is then a simple series of wiki pages on Meta-Wiki, which we can make transparently available via shop.wikimedia.org/wiki-page-name, if somebody wants that URL structure instead. And, hey, wiki pages are easy enough to translate these days, so we've then resolved this bug.

The next step here is probably an RFC on mediawiki.org with a full plan of attack.

It looks like this might be the lucky month of the shop-answers-giveaway (https://lists.wikimedia.org/pipermail/wikimedia-l/2016-March/083495.html ), can someone ping Lisa?

Can we please stop finding excuses to provide services only in English? It's getting old.

Proper product selection and thinking ahead goes a long way. This story makes me really sad. More sad than would be appropriate to write down here, frankly.

Agreed. Especially considering the WMF locks out other affiliates from this space.
And doubly so because the Wikimedia Store has a custom terms of use and privacy policy, which are English only.

It would be very interesting to see aggregated high level demographics data of paying customers of the Wikimedia Shop, to assess whether the English-only language is currently resulting in mostly purchases from English speaking countries.

@MBeat33, do you have timeline to fix this bug?

@Nemo_bis The current luck of proper multilingual support sucks for shopify as has been covered above. Although promised improvements to its customer base have been made over the last half decade to change this, they never materialised. It's certainly one of our desires for merchandise to be more accessible across both language and geography. But that would require us moving software platform and at the moment we don't have a timeline as to when that may or may not happy. There is nothing as yet planned for this financial year (which takes us up to at least June 2017)

Nikerabbit renamed this task from Wikimedia Shop is only available in English to Wikimedia Shop should be translatable to all languages.Mar 14 2017, 9:26 AM
Nikerabbit updated the task description. (Show Details)

Closing the Wikimedia-Shop project in Wikimedia Phabricator, as Wikimedia Phabricator is not used to track Shop /Store issues. The preferred feedback method for Shop / Store issues nowadays is contacting merchandise@wikimedia.org according to the shop managers. If the issue reported in this task is still a problem nowadays, please see the Contact section on https://store.wikimedia.org/ . Thanks for your understanding and sorry that the issue reported in this task has not received earlier feedback / action.

This is a valid task for WMF-General-or-Unknown: the shop is on an ever-changing software, so whatever is the current software's issue tracker matters little, but the overall task remains the same.