This page is for gathering feedback on the Campaign Organizer framework. Please share what you think about the platform! Want to discuss with other people? Give us feedback here on this talk page or Wikimedia Space.
Are there gaps?
What gaps do you see in this framework? Are we missing something? Are there any additional steps that are not included here?
It's mostly nice. Some points:
- For compiling the lists of things to do, there should be several different available sources:
- manually written lists, and they must be easy for community members to update, without having to deploy configuration changes through Gerrit
- Output from tools such as GapFinder
- Wikidata queries
- (perhaps other sources)
- Tracking the progress and reporting the results are closely related. There must be tools that generate progress reports and results reports from the earliest stage.
- A way to share a campaign on various external and social networking channels must be built in as early as possible. E.g.: Send a link to a campaign on email, Twitter, Facebook, WhatsApp, etc. Such a link must lead the user as directly as possible to making a relevant contribution, even if it's a brand new contributor who never edited Wikimedia projects and doesn't have an account.
- This also means that the campaign and the contribution page to which it leads must include as few unnecessary steps as possible: adding templates, signing talk pages, adding userboxes, etc.
- Whatever works for a campaign can also mostly work for single users or very small groups of users who work on lists of tasks: articles about all the books by a certain author, articles about all the villages in a certain province, Wiktionary entries for all the Aramaic words the begin with the letter Mem, etc. The scenario of a single user or a small group should be taken into account when developing such a tool. It's not scope creep because the changes are probably minimal or even non-existent.
Most of these things, as well as a lot of things on the main Campaigns/Organizer Framework page are already addressed in the designs for the Translation list project. It was designed for Content Translation, but despite the title, it can be almost completely reused for any content creation campaign with minimal design changes. These designs were extensively tested with users, and there are detailed reports about it. Generally, the response was good, and there is great demand for it among users and affiliate organizations. Unfortunately, the project was never implemented, but the designs are still 100% relevant. I hope you find them useful (if not, I'd love to know why). --Amir E. Aharoni (talk) 19:47, 23 September 2019 (UTC)
- @Amire80: We agree on the list building options, and I haven't actually found a good hierarchy for these so I am writing some blogs on it. I agree entirely on the phabricator project you describe their -- listbuilding is so central but so fundamentally not understood as a "core" part of the behaviour of the movement. I think I commented on that task earlier in the year about the interest -- unfortunately that is the kind of roadmapping thing that I don't have a ton of influence over. I will keep bring up the opportunities when I get a chance though! Astinson (WMF) (talk) 21:54, 26 September 2019 (UTC)
It is a very nice guide to campaign design. Great work! I would add a few notes:
- I think before gathering resources, agreeing on the timing of the campaign is important. In different communities, having multiple similar campaigns happening at the same time could be detrimental to all of them. In my experience, this happened at the beginning.
- I agree with Aharoni on the tracking progress and reporting tools. Reporting as it is right now is a huge burden on the organizers. There must be a better way to go about it. I actually even applied to talk about this at Wikimania this year to brainstorm what could be done to alleviate this burden.
Finally, I would suggest going over old reports (example: our report) from different campaigns to get an insight into what people have done and their feedback on the process. I understand that it might be a lot of work, but I am just suggesting it in case not many people contributed here.--Reem Al-Kashif (talk) 00:05, 24 September 2019 (UTC)
- Hi @Reem Al-Kashif: -- we have added a bit of information about conflicting and scheduling -- however timing is contingent on a number of things, especially if you are working with partners. For example, I am working with the UNCHR partnership -- and the date of the campaigns was determined because of the partnership, not before. We clarified a bit that the stages are not "steps" persay but components of each broader section of the workflow. Hopefully that will help describe how some of these can be more concurrent.
- As for the tracking: tracking and building worklists (also something @Ahoroni: identifies above) seem to be the most "painful" part of organizing these right now, which is also in the complete control of the Movement to improve -- so we think our initial experiments will be focused on those areas.
- We have been reading a fair amount of the documentation from each of the campaigns that we could identify, and did a data collection push in the Indic-language communities to find additional documentation/experiences -- but of course there is always a risk of missing things. Thanks for flagging it, and anything else that you find inspiring would be useful as well!Astinson (WMF) (talk) 14:28, 25 September 2019 (UTC)
- @Reem Al-Kashif: Did you end up doing the session at Wikimania? If you have notes or outcomes, it would be great to talk about it! Astinson (WMF) (talk) 14:29, 25 September 2019 (UTC)
Should we improve something?
Is the documentation clear? What is confusing or hard to understand?
What should we focus on?
What do you want to learn the most about? If we created training materials or other documentation, what do you think would be the most useful thing to learn?
Any other tools?
What other tools do you use while organizing campaigns or contests?
Unless I missed it, there are no reference to maps in the document − it is somewhat touched upon in the following section with Monumental, but I think it would deserves a clearer entry:
…the Wiki Loves Monuments Community has developed two simple interfaces for contributing to this content: Wikipedia-driven lists with clear upload buttons using Commons Upload Campaigns for images, and the Monumental tool helps communities running a Wiki Loves Monuments Campaign to create an engaging and simple access point for contribution.
Wiki Loves Monuments had been relying on maps-based tools (built on top of the monuments database) long-before the Wikidata-based Monumental came to be − the first one was probably https://tools.wmflabs.org/heritage/map/ (now dead) ; then we had https://tools.wmflabs.org/wlm-maps/ (and its Swedish fork) ; a more recent addition is https://tools.wmflabs.org/map-of-monuments/.
(I don’t have it at the tip of my fingers, but fairly sure there were some maps for Wiki Loves Earth and Wiki Loves Public Art)
The maps-based approach also became the 'canonical' contribution path for two “beyond-WLM” campaigns: Poland’s WikiVacations with http://wikiwakacje.pl/mapa/ ; Austria’s WikiDaheim with https://wikidaheim.at/.
While a somewhat different project (focused on article-writing and not photo-upload), France’s Ma Commune Wikipédia also used a map https://macommune.wikipedia.fr/ as main entry-point (map-of-monuments was also built as an article-writing tool).
So, basically, it the subject of the campaign has geo-coordinates attached, then maps have often been a natural entry-point, sometimes over-taking/replacing text-based lists.
- @Jean-Frédéric: Good point -- I think there is something about the function of map tools to "generate a worklist" so to speak -- that will be better covered in a followup documentation project I am doing on list building. However, I think I will add something in the "Build worklist and design participation tools" section to highlight how different modes of engagement can appeal in different ways. Astinson (WMF) (talk) 15:11, 30 September 2019 (UTC)
Lists from pdfs
i've been recently adding items from print / online sources. a nice tool would be to make upload of lists into wikidata from spreadsheet, and then running tabernacle, to prepare lists for content drive. Slowking4 (talk) 14:27, 3 October 2019 (UTC)
- @Slowking4: Thanks! I think list building is at the center of a lot of the skills we need -- and we should definitely plan on building better tools, practices and training around this. Astinson (WMF) (talk) 12:48, 9 October 2019 (UTC)
Do you want to help?
Are you interested in trying to learn more? Can we reach out to you for further conversations? If you are interested, please describe what you are interested in, and add your username!