Grants talk:IdeaLab/Future IdeaLab Campaigns
Future IdeaLab Campaigns
Thanks for sharing your thoughts on future IdeaLab campaigns. We are looking forward to learning from you and incorporating your input where we can. Discussion is welcome in any language.
Feel free to use this page for open questions and comments including:
- Suggesting / commenting on campaign ideas
- Queries about what specific campaign ideas represent
- Questions about campaigns in general
Proposed IdeaLab campaign topics are also available for review.
- 1 How do you plan to make this happen in the different projects and different languages?
- 2 The language in the survey
- 3 I haven't read the front page, but
- 4 Definitions
- 5 Category tree (model of the world) = map+navigation
- 6 On a related note to preceeding making Wikidata work across different cultures based on my polyglot-ish experience
- 7 Score
- 8 Harassment
- 9 how to mitigate editorial imperialism
How do you plan to make this happen in the different projects and different languages?Edit
I expect you to go to the communities and gather input there, not wait for them to go here. How do you plan to do this?
The Wikiversum is a decentralised entity, with a service agency (WMF) in SF, and all things have to be done decentralised, if they will be deemed valid. Centralisation is working active against the project as the whole. Centralisation may be fine in some aspects, but in general it's plain wrong. Grüße vom Sänger ♫(Reden) 13:02, 5 December 2015 (UTC)
- @Sänger: I'm preparing a brief message for community notification pages for which I'll request community translations, and post them to the appropriate places once they're complete. They will be directed to the meta pages regarding the consultation and include a link to the AllOurIdeas survey. That said, I have no issue with folks uninterested in participating on meta and who would rather discuss campaign topics within their local project. I can monitor those discussions on those local community notification pages and incorporate their campaign ideas into AllOurIdeas as they surface. I JethroBT (WMF) (talk) 20:07, 6 December 2015 (UTC)
In my opinion the Wikivers is so decentralized at this time that there is no transparency. Between reams of guidelines, protocols, technical jargon and many conflicting ideas, it is like being lost in a maze, with too many cracks to hide in. Ideally there should be a balance between anarchy and order, but what it is I don't know - only that I feel the Wikiverse isn't there yet. Worese, I feel that it runs the risk of losing touch with itself and its stated goals. All the Best, Shir-El too 13:26, 6 December 2015 (UTC)
The language in the surveyEdit
Unusual words like "ameliorate" that probably should be dumped in all registers in English: why not keep it simple ("improve")? How unnecessarily hard is that for second-language speakers?
Too many terms I was unprepared for. This significantly reduced the quality of the data I was able to offer. What are "microcontributions" in whatever context was intended? What is ORES? Couldn't there have been warnings about these? ("First, read our short explanatory notes about some of the terms and concepts we'll present you with.") Or links from the actual choices on the form. Perhaps I might have given ORES lots of thumbs-ups, but I just don't know, so it scored can't-decides and don't-knows, or I voted for the other one. This distorts your data.
Who are "project volunteers"? Off-wiki affiliate-type volunteers, or online editors? If you say both, I'll say that's a low-value question. If you say one or the other, I'll ask why this couldn't have been clarified.
- @Tony1: I can speak to the question about use of the term volunteers, which refers to anyone volunteering their time to contribute to a Wikimedia project in any capacity; it therefore refers to both. IdeaLab campaigns are intended to cast a wide net though, so I think it's OK if it refers to both online and offline volunteers. The other choices you've mentioned were submitted by participants of the survey. You make a good point about making sure the language is simple; I'll go ahead and language in the instructional green text during at the top to encourage participants to use such language. I can also point you to ORES here if you want to learn more about it. I JethroBT (WMF) (talk) 06:03, 7 December 2015 (UTC)
- Thank you, but only one issue of several is resolved. Why are you not liaising with editors who suggest issues to simplify and straighten out the language used before it goes live? Linking me now to information about ORES is not the point—once you start the survey it's too late, without bother and time-wasting. The offline/online volunteers are very different: it yields no useful data when the categories are fuzzed up. This is a most unsatisfactory instrument. Tony (talk) 07:54, 9 December 2015 (UTC)
I haven't read the front page, butEdit
...as I have been politely asked to share my thoughts, I am willing to do so. Indeed, there are some topics that I'd like to see discussed in meta. My favourite two are:
- content maintenance instead of the paradigm of growth in numbers: How do we keep the content which we already have up-to-date? How do we go sure that content is really correct, which is much more than "not obviously wrong"?
- problematic accounts who more than occasionally provide useful content: How do we integrate them (keep them integrated) in a way which does not impair the editing experience of others and which prevents (further) trollification?
- Thanks for your thoughts here, Man77. Your second idea sounds similar to a topic already submitted, Strategies for engaging with and motivating project volunteers, albeit this idea is much more general. Your second idea sounds great; I've tried to summarize it in the following way:
- Developing strategies to maintain accurate and up-to-date project content
- If you want, you can add this or your preferred phrasing into the survey directly. Thanks again, I JethroBT (WMF) (talk) 20:51, 7 December 2015 (UTC)
First of all, I request you people to properly define new term such as AllIdeaslab, IdeaLab and such. It is possible that these new terms may be misinterpreted and by the basic laws of logic we will have many meanings of the same term by many readers. That will create only confusion. So please, give your authentic definition of these new terms so that we may participate properly and contribute effectively. After that I wish to participate in that since, as I feel from my interpretation of these terms, The idea is very useful for exchange of views. Pathare Prabhu (talk) 12:51, 10 December 2015 (UTC)
Category tree (=model/picture of the world) - a "map". The current philosophy of categorization - too huge number of categories is prohibited. Therefore, we have a "map" where "scale"="1:10000000..." ("children's", primitive, very very simple graph), but not "1:1" (real, as it is). To have a "1:1 scale", we need to: 1) allow editors (or group) to create alternative (individual authors) trees categories. "Survive" will be the most "adapted" (corresponding to the rules), as in natural selection. 2) To work with a large number of categories will need the appropriate tool. Big "map" ("1:1") is useful for people/intelligent systems navigating, queries to the knowledge base (category tree) like this: a) Who / What + b) By whom / Whereby + c) carry out the process + d) The result of the process. Search can be by a, b, c, d or (a + (b + (c + (d)))), like semantic search, navigating. Question: how campaigns can help solve the problem 1 - big category tree, adding to the MediaWiki possibility of creating alternative category tree (small, big; scientific, philosophical, religious and so on) and 2 - a navigation tool for big category tree, search, edit, display, etc.? --Fractaler (talk) 11:33, 18 December 2015 (UTC)
Someone will have to figure out Wikidata meshes into the different cultures. Here is the background: I am something of a (poor) polyglot and have started to attempt translations of a few articles related to my interests. There is a big difference between the wiki language versions. The German version, in particular, is "run with Germanic precision" as newspapers' popular newfangled idiom puts it these days. Google machine translations into German seem particularly problematic, so I was rather surprised my contributed translation was mistaken for one. Someone should copy the "don't bite the newbie prospective translators" campaign over there. A bigger issue (with the Germanic precision) is that inter-wiki links, even if appropriate marked, are frowned upon (true of many wikipedia localizations, but it seems especially there). From experience, more content, more quality links, make a page more useful. Hyperlinks are a major reason why Wikipedia is superior to paper encyclopedias. Taking a subject that might be a long article in a newspaper or paper encyclopedia and re-writing it concisely, but full of appropriate links, seems to be the true art of Wikipedia. It leverages the work of other editors into your article. If you have foreign-language links (and particularly if the article is already a translation), IMHO it makes sense to take advantage of them: more links/content are better than less, and readers presumable have machine translation turned on. Here is how this relates to Wikidata: data theory says you want many views of a centralized (or at least cached) data to improve data quality. So one infobox pulling from Wikidata that gets translated into 20 languages, not 20 different conflicting infoboxes managed independently. (Especially important with biographies where you don't want to police 20 versions of them, including monitoring the Old English version.) This seems the essence of what Wikidata is trying to do. So far so good. Now... if you have that auto-translated infobox pulling from Wikidata (which I have already seen working quite nicely on the Spanish site), sometimes someone's alma matter in the infobox, say, will not have an article in Spanish. Then you get the (en) link. But, the German wikipedia, at least, currently considers these evil and seems to revert them immediately (at least if they came from a wiki other than Wikidata). It seems the Swedes have the right idea with "Ideanator", a 3rd party (?) site pulling from Wikidata. If the article is not available in Swedish but has rich Wikidata, search on the Swedish site shows you an "article" --- infobox, really --- written in pseudo-Swedish that it writes using the Wikidata source. (In practice, Wikidata labels are translated into the "nearest" available language, ideally based on user-preference language priorities set in the browser but I doubt this is implemented). I think this is the future. (The superintelligence is apparently not yet at Google given how bad machine translation still is.) How would this work in a localized Wikipedia article? If the linked article is not available in the target language (Swedish?) you'd get the Ideanator link, or something similar (maybe a Wikidata icon) instead of the (en) link now generated by some Wikidata templates. You click the Wikidata/Ideanator icon, and it brings an infobox translated from Wikidata, together with a list of the article's available languages. For many things --- institutions, etc., the Ideanator infobox may be good enough for a foreign language speaker to get a strong gist of the concept and decide if reading a full article machine translation is worth the trouble. On some Wikipedias, this will require a change of culture to embrace the inter-wiki link, or at least the link to Wikidata/Ideanator article. Rich links are the essence of hypertext & Wikipedia. IMHO links to quality content are a presence that strengthens the usefulness of articles rather than weakens it. Certainly, don't bite the newbies if they add inter-wiki links to Wikidata, please. Just my 2 cents. Prot783 (talk) 06:34, 22 December 2015 (UTC)
- Hey Pharos. The lists of ideas can be sorted using the table from highest to lowest score, but careful interpretation is needed here. While a higher score does mean a topic increasingly likely to be preferred, the score does not take into account the relative frequency with which that topic was actually voted for (or against) compared to other topics. The result is that some topics with more extreme scores resulted because they were voted on so few times, so those scores are less representative of the communities we sampled. This is a limitation of the surveying method I chose; it allows for new topics to be submitted by participants during a surveying period, at the risk of lowered participation (though, many submitted topics received a substantial number of votes). It was more important to me to ensure participants were able to submit their own ideas in an easy and accessible manner in this case. Bit of a lengthy answer, but does that clarify things a bit? I JethroBT (WMF) (talk) 23:24, 2 February 2016 (UTC)
On Category:IdeaLab/Ideas/Inspire/Addressing harassment, I'm surprised to read that «This campaign is happening because it's one of many ideas for campaigns that the community requested». Could the page explain the outcome and what reasoning led to the conclusion that the community requested said campaign? In my reading of the results, harassment was the least-desired topic. Nemo 15:39, 5 June 2016 (UTC)
how to mitigate editorial imperialismEdit
The curent (harrassment) campaign refers to mechanisms being “gamed” by editors to remove ideological opponents. The 'gaming' by editors is, in my view, one of the most significantly prevalent and destructive forces currently at work within the wikipedia projects. I have repeatedly witnessed areas of contention being controlled by a cadre of like-minded advocates of an ideological polarity and, in many ways, the current editorial mechanisms systemically reinforce such behaviours.
In fairness, as some articles mature, many polarised themes have given way to more balanced (and better researched) content. However, and particularly when there are academics with a lot of time and many papers to hand, years of editorial contributions can be swept aside in a matter of days, with significant shifts made in the content and interpretation of the topic; often (and unfortunately) much of the nuanced editorial content is 'tidied' into a presentment of the topic from only one academic domain.