Hello, everybody! If you'd like to discuss anything with me, please post new topics at the bottom of the page and sign the posts with 4 tildes (the ~ key) in a row. Thanks!
our state library some ten or twenty years ago used to have a music listening area (long gone now), and on the notice boards were fliers for feldrenkrais and alexander technique practitioners with their intent to attract musicians requiring attention. JarrahTree (talk) 09:41, 17 February 2024 (UTC)Reply
oh dear the SUL work around failed for a while, it is like dancing with a drunken cat, I think I have found a way in again...
I am not sure I have really got a handle on the lengthy discussion as an Australian reading the talk page at the Australia article, maybe I would not be much help, as a 60+ native born west australian, some of the assertions look like crap -give me walk in Manhattan, or Ceduna anytime than offering generalisations about Australian transitions in culture in the last 50 years... Trust all is well. The more I think about that lengthy discussion, I am trying to prevent myself from adding anything further sigh.... JarrahTree (talk) 07:37, 9 March 2024 (UTC)Reply
BTW my apologies - I had not given a link to the background info on what I call SUL - probably more correctly called Unified login...
Latest comment: 5 months ago10 comments4 people in discussion
I'm a veteran Wikimedia editor but new to Wikivoyage. Can you explain why it is difficult to link recent Wikimedia Commons files on Wikivoyage? Thanks, Buaidh (talk) 03:03, 19 February 2024 (UTC)Reply
If it is, that's something new to me. Have you been using double square brackets and putting the filename inside? For example: [[File:Filename|thumb|Caption]] Ikan Kekek (talk) 03:41, 19 February 2024 (UTC)Reply
In that case one should see a deletion log entry or something (the filename doesn't hint on anything that could have been censored). It isn't on Wikipedia in English either. –LPfi (talk) 20:40, 19 February 2024 (UTC)Reply
The reason why the example above produces a redlink is because the file name is misspelled (the correct file name starts with "Olde" like in the link, not just "Old"). --Ypsilon (talk) 05:02, 20 February 2024 (UTC)Reply
If you feel it's necessary, go ahead, but I'd couch it as ttcf and Be fair#Political disputes. The main point is that Wikivoyage doesn't seek to score political points but does provide information people who may be traveling somewhere should know. Ikan Kekek (talk) 23:20, 4 March 2024 (UTC)Reply
Political leanings in Ukraine
Latest comment: 4 months ago14 comments6 people in discussion
Admittedly, we wouldn't know for sure how the population thinks now, but if you look at electoral maps of Ukraine prior to Euromaidan in 2014, the eastern and southern provinces of Ukraine (including Crimea), which were mostly Russian-speaking, mostly voted for pro-Russia candidates, while the central and western provinces, which are mostly Ukrainian-speaking, mostly voted for pro-US/EU candidates. It was covered in this lecture by University of Chicago professor John Mearsheimer, who is quite well-known in political science circles as one of America's foremost geopolitical scholars today. The dog2 (talk) 20:51, 5 March 2024 (UTC)Reply
Mearsheimer is now considered to hold fringe views on Ukraine, but this just highlights how we should avoid getting trapped in these discussions. If it is controversial, let's avoid the discussion and focus on travel information. Wikivoyage does not need to take a stand on these issues. Readers should get their information on this sort thing from better sources than a travel guide. Ground Zero (talk) 21:15, 5 March 2024 (UTC)Reply
The other thing is that every region of Ukraine voted for independence in 1991, though by a lower percentage and with more absenteeism in Crimea than in other regions. It's a fallacy that being Russian-speaking equates to wanting to be part of Russia. Odesa and Kharkiv are or at least were Russian-speaking until they were repeatedly bombed. What do you think most of their inhabitants think about Russia now? But Ground Zero really has the best response. Ikan Kekek (talk) 22:12, 5 March 2024 (UTC)Reply
This is a bit of a tangent, and I am in no way suggesting that we include any of this in any of our articles, but Mearsheimer's views may not be as fringe as you think. According to opinion polls, less than 30% of Americans are in favour of sending American troops into Ukraine to fight the Russians, but more than 50% of Americans are in favour of sending American troops to Taiwan to fight the Chinese should war break out there. And Nikki Haley has in fact promised that if she gets elected president, her first order of business will be to go war with China. The dog2 (talk) 00:47, 6 March 2024 (UTC)Reply
Cite the opinion poll and provide the wording of the questions. Also, Haley will not be elected, but come on, she said she'll attack China on Day 1 in the alternate universe in which she won? Ikan Kekek (talk) 01:21, 6 March 2024 (UTC)Reply
Here's an article citing one of those polls: [1]. In this particular one, it shows that 60% of the American public support U.S. military intervention in Taiwan. The dog2 (talk) 02:23, 6 March 2024 (UTC)Reply
Thanks. Their methodology seems problematic:
"To obtain the sample, YouGov used a sample matching approach to recruit respondents and generate a
representative sample of the American population.
YouGov sample is generated through an opt-in survey panel, composed of approximately 1 million US residents
who agreed to participate in YouGov’s Web surveys. Panel members are recruited using multiple methods to help
ensure diversity in the panel population. Recruiting methods include Web advertising campaigns (public surveys),
permission-based email campaigns, partner-sponsored solicitations, SMS-to-Web recruitment (voter registration-
based sampling). YouGov invites panel members by email and/or SMS, depending on how the panelist has opted to
be contacted. All invited respondents opted-in to receive survey invitations from YouGov.
To recruit a representative sample, YouGov recruited panelists to this survey by their previously provided answers
to YouGov profile questions. At the recruitment stage respondents were invited based on their fit to interlocking
variables of (gender x race x age x education). To match survey participants to the YouGov population frame, more
interviews than needed are collected. The final sample of respondents to this survey were matched to a more
complete population frame, selecting the closest matches to the population. Completed interviews were weighted to
the sampling frame using propensity scores. Population weights were normalized to equal sample size."
I have no idea how reliable their results are. I'm very skeptical that 50-60% of Americans want U.S. troops to die in the defense of Taiwan, which is not to say that I oppose a U.S. commitment to help Taiwan defend itself. Also, you just can't rely on a single poll for anything. Ikan Kekek (talk) 02:32, 6 March 2024 (UTC)Reply
┌────────────────────────────────────────────────────────────────────────────────────────────────────┘
As web surveys are much less expensive than phone or snailmail ones, and you probably get high response rates from opt-in panels, an approach like this is common. By comparing sample profile statistics with population statistics, one can test for bias, although not necessarily reliably.
Extending this to giving more weight to answers from people from groups under-represented in the sample (or choosing a sub-sample with a less biased profile) lessens the bias in the profiles, and is a commonly used strategy.
However, there is no guarantee that a person from an under-represented category represents all the category well. What about Democrats reached through Trump's X feed? Democrats are certainly under-represented there, but giving greater weight to such odd Democrats might give worse results than not using such weights and just have fewer Democrats in the sample (and admitting the sample isn't representative of the population at large).
Mostly the survey institutes try their best to get good samples, but there are no statistic measures that can test the biases introduced by the sampling methods (other than indirectly by comparing survey results with similar surveys with a better sample, or by comparing relevant variables in the original sample with those of the general population).
Here, they tried to get a "better diversity" (essentially less bias, I assume) by using different methods of recruitment. Adding a bad method to the other ones can give more diversity, but might increase the bias. For the several-methods approach to work, you need methods that complement each other. I don't know whether web advertising campaigns and permission-based email campaigns reach different-enough populations to motivate not just using the better one. I hope those responsible did test the profiles of these different populations to be able to conclude they got a better sample this way (that's up to their professionalism, both in skill and in wish for correct results, especially as commercial survey institutes seldom publish a thorough discussion on their methodologies). Anyway, I assume any opt-in panel will include more extremists than an unbiased sample. –LPfi (talk) 12:58, 6 March 2024 (UTC)Reply
An opt-in panel is likely to bias results in that their opinions may well be different from everyone who would refuse to participate, regardless of any other demographics. That's a big problem in polling now. Ikan Kekek (talk) 17:14, 6 March 2024 (UTC)Reply
My understanding is that the reason why most Americans oppose sending U.S. troops into Ukraine to fight the Russians is because of fear over nuclear escalation, given that Russia has the most number of nukes in the world. I wonder if the average American is aware that China also has nukes, albeit about only one tenth the number that the U.S. and Russia have. After all, a war between the U.S. and China over Taiwan is very likely to go nuclear, especially if China manages to sink an aircraft carrier, and the American people start calling for Chinese blood to avenge the deaths of their soldiers. And not to mention, such a war is also very likely to escalate into World War III, given that Japan, South Korea, the Philippines, India, Canada, Australia, the UK, New Zealand, France, Germany and Italy are all likely to join the Americans in fighting for Taiwan. If the poll numbers are accurate, perhaps it reflects that the average citizen in Middle America thinks of China as a backward country that the U.S. military can easily overwhelm, and isn't aware that China could potentially destroy New York City and Los Angeles with their nukes, and all those other countries that join the American-led coalition might see their cities get nuked too. The dog2 (talk) 17:30, 6 March 2024 (UTC)Reply
Again, quite apart from methodological issues with this poll, basing anything on a single poll is senseless. Yet you are engaging in flights of fancy. I don't think the Chinese dictator is any more interested in getting his country nuked than Putin is. Can we please end this discussion? Ikan Kekek (talk) 17:50, 6 March 2024 (UTC)Reply