Grants talk:PEG/UG ELiso/Content Translation Campaign 2015

A decision has been made on this grant. Comments are still welcome on this page.

GAC members who support this request

1. --DerekvG (talk) 14:08, 2 October 2015 (UTC)
2. --Ilario (talk) 19:50, 7 October 2015 (UTC)
3. I don't support the idea of translating to Esperanto as an end, however it seems quite clear that this will have a benefit in making translation to other languages, particularly European languages, much easier once the content is there. Given that it's only a small sum of money, it's worth experimenting with. Craig Franklin (talk) 06:24, 10 October 2015 (UTC).
4. -- ♪Karthik♫ ♪Nadar♫ 17:55, 23 October 2015 (UTC)

Updates

1. After more consultations I have updated budget (additional 15 EUR) as diploms / thanks letters should be printed on some more representative paper. --KuboF (talk) 21:41, 7 October 2015 (UTC)

Community comments

We, organisers, would like to receive your recommendations about these issues:

Recruiting new users

ELiSo is world-wide diaspora organisation and it wouldn't be effective to make physical edit-a-thon or similar meeting to promote the competition. We will use Central Notice, Village Pump on all Esperanto language Wikimedia projects, mailinglists, social media, personal invitation and communication channels of our partners. And some revue articles after the competition. How we could grab new users during the competition and keep them active? --KuboF (talk) 22:38, 2 October 2015 (UTC)

Rules about machine translation

Usual translation competitions can give more points to human translation than to computed-aided (pre-translated by machine) translations. But we want to encourage users to use the tool in effective ways to create more quality articles in future; and if they will use machine translation, it is OK. On the other side, if human translation would be equal to machine translation, in practice would be preferred machine translation (now from 4 languages) with risk of creating or extending content gap. Quite paradox; any suggestions? --KuboF (talk) 22:38, 2 October 2015 (UTC)

Tony1

As you know, I don't generally support the funding of Esperanto-related projects, even though I had a go at learning E. as a teenager. But grudgingly I'll mildly support this. It's a modest budget, and the project is well designed. Care has been taken in framing the measures of success—I like them. Two suggestions: I'd slant the spending towards one or two big top prizes, since that has significant recruitment and publicity value (in relation to your first issue above); questionnaires for feedback—you might pass your draft through grantmaking/evaluation staff, who are very good at advising on optimising the utility of data through their design ... it's not easy to do, in my experience. I suppose you'll be notifying the various Esperanto societies around the world, too. Tony (talk) 09:29, 8 October 2015 (UTC)

Thanks for comment, Tony! Yes, it is planed 2 tablets as first and second prizes to attract more attention. Esperanto organisations are in our inform list too. Thanks for advice about grantmaking/evaluation staff; I only thinking about checking of questionnaire in Esperanto but we will solve it somehow. --KuboF (talk) 18:57, 8 October 2015 (UTC)

WMF comments

Hi KuboF, thank you for the work you have done to prepare this request and respond to community comments. It is great to see a project that will encourage contributors to use the Content Translation tool while also expanding the Esperanto Wikipedia. In a grant discussion last year, a strong point was made that Esperanto Wikipedia makes information accessible to users who have small Wikipedia's in their local languages and find Esperanto easier to read than English. Please see the comments below, and let us know if you have questions or need additional planning support.

• In the impact section there are a few lines that suggest that help pages and instruction materials for how to use the the Content Translation tool will be created as part of this grant. This kind of documentation could be very useful for others - it would be good to have a plan for where these instructional materials will be posted so that they are easy for users to find.
• 800KB of content for 60 translations would mean translated articles would be very short stubs. Would such short articles improve the quality of Esperanto Wikipedia?
• Do you have a plan for how you will measure the number of contest participants who are still using Content Translation tool three months after the contest, i.e. can you collect that information using a tool, or will you need to survey participants to find out?

We look forward to your responses! Cheers --KHarold (WMF) (talk) 00:10, 21 October 2015 (UTC)

@KHarold (WMF):

• Good point! For the competition there will be brief introduction and link to full description on the competition page. ContentTranslation help will include this links as well. For the time after competition we should think more about it. Now I think about adding link to ContentTranslation's description in Beta features and maybe into ContentTranslation itself. It is a thing to think more!
• 800 KB for 60 articles means 13,3 KB per article, what is quite good. It is about 7 times more than mean article on Esperanto Wikipedia (as of February 2014). Definitely, such articles would improve Esperanto Wikipedia's quality. On the other hand, the main goal of the competition is to motivate users for useful using of ContentTranslation, not the one-time improvement (this program is planned as kick-up for middle-to-long term impact).
• It is planned to use Wikimetrics for users activity tracking. I personally asked Amire80 (ContentTranslation's product manager) and he accepted to eventually help with special statistics.

--KuboF (talk) 20:55, 22 October 2015 (UTC)

Hi KuboF, I must have made a mistake with my calculation of article size, you are correct, 13.3 KB articles would not be stubs at all. We are pleased to approve this request with the understanding that in addition to growing the content available on Esperanto Wikipedia, there will be secondary benefits of increasing documentation on how to use the Content Translation tool, and encouraging editors to begin using the tool. We look forward to learning whether this is an effective strategy for increasing the use of the Content Translation tool, and hopefully, learning new strategies for reaching audiences on many different projects. Cheers! --KHarold (WMF) (talk) 21:13, 26 October 2015 (UTC)

Planning resources

Amical Wikimedia, WMEE and WMES have run content translation contests that required outreach to communities on many different language projects, you might consider contacting them directly if you do not find helpful information in the links below.

• You can learn a bit more about WMES's outreach plan in their WLM 2014 Report.
• Amical Wikimedia shared some of the challenges they have encountered promoting contents to the international community in their APG 2014-15 Interim Report.
• You might consider reaching out to organizers of WMEE's annual Translation Bee to learn how they publicize the event internationally.

You can find resources to plan and execute the surveys described in the grant request in the following learning patterns:

• Learning pattern: Qualtrics survey tool
• Learning pattern: Asking the right questions