Wikivoyage talk:Listings

Active discussions

Archived discussions

Embassy and Consulate listingsEdit

Swept in from the pub

While I agree that this should be kept as it is useful for travellers who want to get visas or renew their passports, in the interest of usefulness to travellers, I think we should perhaps agree on a set of rules as to what we should list to keep the lists from becoming too unwieldy. For instance, some embassies have separate chancery and consular locations, so in that case, I think we should only list the consular sections, since there is probably no reason for you to visit the chancery unless you are some important government official (and if you are, your government will make all the arrangements, and you probably won't be relying to Wikivoyage). In the Canberra article for instance, I don't think we should be listing the U.S. embassy since it doesn't provide consular services. Residents of Canberra would have to go to the consulate in Sydney if they want to apply for a U.S. visa, and likewise, U.S. citizens based in Canberra need to go to Sydney to renew their passports. And in most cities with a Hong Kong Economic and Trade Office (with the exception of Shanghai), that office does not provide consular services, and you'll have to go to the Chinese embassy/consulate if you want to apply for a Hong Kong visa, so it makes no sense to list the Hong Kong Economic and Trade Office.

There is one complication though. In Singapore for instance, there is a separate visa application centre if you want to apply for a Chinese visa, and the embassy does not handle visa applications. On the other hand, if you are a Chinese citizen based in Singapore and want to renew your passport, that is handled by the Chinese embassy itself. In such cases, how should we handle the listing? The dog2 (talk) 17:34, 17 January 2020 (UTC)

Yes, good points. There's no point listing embassies that don't offer consular services to the general public. I would hope diplomatic staff do use Wikivoyage during their postings, but not to find their own office!
Issues like the China Embassy in Singapore can be handled on a case-by-case basis; both offices in that case are useful to travellers, so they should both be listed. --ThunderingTyphoons! (talk) 17:50, 17 January 2020 (UTC)
I agree that we should only list offices that provide consular services and visa application centres. This is not the place for trade office information. Ground Zero (talk) 17:53, 17 January 2020 (UTC)
Consular services for residents of a country are not just routine stuff like passports, but also support for those in prison, or support to business travellers. So I think it is OK to list a consulate if it doesn't handle the routine stuff but would be useful to a small business looking to export.
I think we should also think about how we arrange the listings. In most cases travellers are either interested in their country's consulate to get help or a particular country's consulate to apply for a visa etc. Travellers are not going to look at the list of consulates in a city to decide which to visit - it would be more useful to have a page which showed where all the consulates of a particular country are on the map - a set of pages like Consulates of Australia. AlasdairW (talk) 18:34, 17 January 2020 (UTC)
In the scenario you mentioned, I guess the Hong Kong Economic and Trade Office would assist you if you own a business and want to expand into Hong Kong, because that falls under the umbrella of trade relations. On the other hand, if you are a Hong Kong resident who is arrested in a foreign country, it is the Chinese embassy/consulate that will provide you with consular assistance, not the Hong Kong Economic and Trade Office.
And yes, I think it might be useful to have pages listing the consulates and embassies of each country, but we need to figure out what articles like that should be classified under. The dog2 (talk) 19:54, 17 January 2020 (UTC)
Speaking of which, I'm not sure how it works in different countries but in Singapore, even though Denmark, Norway, Sweden and Finland each have their own embassies, visa applications for all the Nordic countries (including Iceland) are handled by the Danish embassy, though each of those individual embassies will provide consular services for their own respective citizens, so I guess that is an additional complication. I've tried to reflect that in the Singapore article, so let me know if you think this is an acceptable solution. The dog2 (talk) 20:19, 17 January 2020 (UTC)
The Singapore list looks good. The small size of the country does eases one aspect of this - there is no question of do I look in Perth, Adelaide or Melbourne for the nearest consulate if I am in Coober Pedy. AlasdairW (talk) 23:16, 17 January 2020 (UTC)

Tours againEdit

I'm not entirely happy with the policy at Wikivoyage:Listings#Tour_listings. I think it misses some types of listing that should be allowed, though I'm not sure how to allow them without starting down a slippery slope.

I'm mostly an independent traveller & almost never take tours, but there are exceptions.

  • One is for complex attractions like the Forbidden City, D-Day beaches or Angkor Wat; sure I could go on my own & WV should provide info for that, but I'd prefer to have a local expert guiding me & I'd like WV to tell me where to find a good one.
  • The other is for attractions that are inconveniently rural; for example, on Bohol many people want to see the Chocolate Hills and/or the tarsier sanctuary. Sure you can get to those by jeepney or rent a motorcycle & ride to them, but most people go on small tour buses. I'd like WV to tell me which of the many bus operators are good: safe drivers, competent guides, good air conditioning, ...

I also often travel with Filipinos who are far more into tours than I am; e.g. I recently circumnavigated Siquijor by traysikel because that's what the gang I was with wanted to do. Should we provide info for people like them? Pashley (talk) 12:04, 20 March 2020 (UTC)

I don't have a problem with doing so in principle, but how would we design things so that only particularly good tours are listed? That didn't work before we banned tour listings. Ikan Kekek (talk) 00:26, 5 May 2020 (UTC)

Yelp is not a reliable source for phone numbersEdit

Swept in from the pub

See https://www.vice.com/en_us/article/wjwebw/yelp-is-sneakily-replacing-restaurants-phone-numbers-so-grubhub-can-take-a-cut Apparently, Yelp and Grubhub have inserted fake phone numbers, to drive business to Grubhub instead of local restaurants. Please keep an eye out for that if you're looking up contact information. WhatamIdoing (talk) 19:13, 23 May 2020 (UTC)

Thanks for posting this. Is there any website where we could find a list of all phone numbers owned by Grubhub? We could then see if any are listed on Wikivoyage and replace them. —Granger (talk · contribs) 19:26, 23 May 2020 (UTC)
@WhatamIdoing: Thanks a lot for this but please don't use Google AMP. I've changed it to the proper URI and hope you don't mind. I recommend https://addons.mozilla.org/en-US/firefox/addon/amp2html/?src=search But yes, definitely do not trust Yelp and especially GrubHub for phone numbers. —Justin (koavf)TCM 19:30, 23 May 2020 (UTC)
When I create a listing for a place I have not personally been to, I always base the information on their own website. Where else would I get accurate information? Hobbitschuster (talk) 21:39, 23 May 2020 (UTC)
Not every place maintains a website. In that case, a web search is the normal way to find contact information. WhatamIdoing (talk) 23:14, 23 May 2020 (UTC)

Google's "Black-owned" label: Should Wikivoyage mention owner ethnicity?Edit

Swept in from the pub

Google has recently added a feature which marks a business as black-owned. [1] The labeling is controversial. Would there be any circumstance where the owners' ethnic origin, nationality or skin color would be relevant in a Wikivoyage entry? /Yvwv (talk) 22:28, 7 August 2020 (UTC)

My immediate thought is that travellers might want to seek out a place that's locally owned—I think "locally owned" is a perfectly fine descriptor in a listing. For mom-and-pop shops, the nationality of the people who run it can be part of a description of what the establishment is like ("A quiet three-room bed-and-breakfast set back in the woods, run by an elderly German couple" or whatever). Ethnic origin can sometimes be relevant for restaurants ("authentic Bai cuisine"). I'm not sure one way or another about "black-owned", which seems more controversial. —Granger (talk · contribs) 23:10, 7 August 2020 (UTC)
Sometimes the ethnicity of a business's owner is irrelevant, sometimes it's downright essential. As an example of the latter: in Buffalo, the African Heritage Food Co-Op is a neighborhood grocery store that also functions as a business incubator operated by a local not-for-profit to help local African-American entrepreneurs get set up in business. They host a Black Business Bazaar every week (or they did before COVID) with pop-up vendors, and their official motto is "Anything Less Than Ownership Is Unacceptable". I can't imagine how it would be possible to create a listing for that place that is complete without mentioning that it's black-owned. -- AndreCarrotflower (talk) 23:41, 7 August 2020 (UTC)
In tourist towns like Dumaguete we mention nationality for some foreign restaurant owners & chefs. Also for one place "Run by a Filipino chef who trained in Thailand. If you like spicy food ..." More controversially, in Tibet some tourists seek out locally-owned businesses, trying to avoid giving money to colonisers. Pashley (talk) 23:58, 7 August 2020 (UTC)
Depends on what. For instance, you might want mention that a soul food restaurant is black-owned, because some tourists might want to support black-owned businesses and African-American culture. The dog2 (talk) 00:51, 8 August 2020 (UTC)
It is information for the readers. Many readers won't care. Others readers will care. I think we provide the information and let readers decide for themselves whether it matters to them. I don't think that we should take the position that readers should not care about whether a business is Black-owned or not, which I think is what a policy against identifying Black-owned businesses would be saying. Ground Zero (talk) 01:01, 8 August 2020 (UTC)
I understand the sensitivity about these kinds of designations because, for example, in the past, some Jew-haters have avoided or attacked shops because they were Jewish-owned, but in non-genocidal situations, I agree with Ground Zero - it's simply information that readers might or might not care about. Ikan Kekek (talk) 01:12, 8 August 2020 (UTC)
Most of the time it would not be relevant. Sometimes it would be. You could mention other characteristics of the business owner, who may be a woman, somebody from the LGBT community, a veteran, someone who is paraplegic, blind, on the autism spectrum, etc. but there has to be a relevant connection between the aspect of their identity and the business. Knowing the ethnicity of the business owner also doesn't tell you where all of your money is going. The employees, landlord and suppliers could all be of a different ethnicity. Gizza (roam) 01:37, 8 August 2020 (UTC)
I recently bought wine by a Black-owned Ontario winery to support Black-owned business, and because someone in my wine club said it's good. I haven't tried it yet. There is no connection between the race and wine, at least in this case, but I think it should be my choice whether I want to support Black-owned businesses or not. I make no judgement about those who choose not to use that information making purchases. Ground Zero (talk) 01:55, 8 August 2020 (UTC)
I generally agree with what others have said. Our job on WV is to write a lively description of places. Sometimes, mentioning background details about a shop's proprietors adds some nice flavor or context to the description, and I don't doubt we already have a number of listings that do this. I also think that a liberal interpretation of WV:Boring implies that of course we should look for and recommend minority-owned businesses. I don't think, however, that we need to try to flag every business we list that's black-owned, women-owned, LGBT-owned, veteran-owned, or whatever. Every business is owned by somebody, and the attributes of who owns it are not always important or interesting, even when they are a minority. A reader who's looking exclusively for that is looking for an index, not a travel guide (or they're looking for "Blackivoyage, the free worldwide travel guide to black-owned businesses that you can edit" :-P, which Wikivoyage is not). --Bigpeteb (talk) 02:04, 8 August 2020 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── If you go to New Orleans, Willie Mae's Scotch House and Dooky Chase Restaurant are two soul foods places that are not just mere restaurants but institutions precisely because of their connection to African-American history and the Civil Right Movement. Both businesses were owned by African-American women who lived into their late nineties, and are still run by the same family. So in the case of both those businesses, the race of the owners is absolutely relevant to people who are interested in African-American history.

Anyway, I think we should take a sensible approach to this. Not every black, Latino, LGBT or whatever-owned business needs to be specifically identified as such, but I would not support a blanket ban on mentioning it. The dog2 (talk) 02:43, 8 August 2020 (UTC)

I feel like that's the right balance, too. It's optional to mention and might be useful information to some reader. Ikan Kekek (talk) 02:50, 8 August 2020 (UTC)
I agree with the overall consensus, and add that "Black" – that is, the whole idea of seeing the entire world as divided into four races, so that a Cameroonian person and a Jamaican person and an African-American person are basically 'the same' – seems to mostly be an American thing. Other parts of the world focus more on sub-groups, so that "Cameroonian" or "Black" is less important than whether you are a Fulani vs a Highlander. With that in mind, I think that any labels provided ought to be relevant to the destination. "Black-owned" is potentially informative for US destinations, but it wouldn't make sense for many destinations. WhatamIdoing (talk) 04:00, 8 August 2020 (UTC)
Yes to most things said above. I'll add that it often is difficult to know who really run a place. Like DaGizza said, not only ownership matters. And there are people trying to ride on people wanting an authentic experience or wanting to support a special group. How do we know who owns the place? Do we check the business records? It works for small places, where you spoke to the owner and he or she told you about the business, but trying to put this label on places beyond that, you will often get fooled. Tell it where it is an important aspect and you know, but don't try too much. --LPfi (talk) 05:18, 8 August 2020 (UTC)

Suggestion for improvement: "Dynamic opening hours"Edit

Swept in from the pub
 
OsmAnd screenshot with opening hours

One update could be quite convenient for travellers: Instead of showing the opening hours in the old school "Lonely Planet" way (e.g. Mo-Fr 9:00-18:00), wouldn't it be nice, if the website could parse it to something like live opening information? Examples:

If the place is still closed

  • "Closed. Opens in 26 mins" or
  • "Closed. Opens on Monday at 8:00" (in case it only opens further in the future)

If the place is open

  • "Open"
  • "Open. Closes in 20mins" (in case it closes soon)

OpenStreetMap-based OsmAnd, the navigation app I am using, is able to do that (please see screenshot on the right). There seem to be libraries to parse the "opening_hours" OSM tag. So the effort should be moderate.--Renek78 (talk) 11:31, 10 August 2020 (UTC)

Dynamic hours would be great for people using Wikivoyage while travelling, but static hours are better for people who are planning a trip: I don't want to go to Leipzig on a Monday if all of the museums are closed then. Ground Zero (talk) 11:54, 10 August 2020 (UTC)
Yes, I too think it'd be good if readers would have the option to toggle between the live information and the opening hours for each weekday. They're not necessarily going to the place they're reading about immediately. --Ypsilon (talk) 12:31, 10 August 2020 (UTC)
I am really irritate on places that provide only dynamic info. I want to plan when to go based on what is open, which makes me have to guess at a million dates and then check each place (and transportation) for all of them. Tell me what you regard off season! A related observation, which probably should be told in a suitable travel topic, is route planners and booking engines claiming there is no service on days for which they don't have information yet (which in corona times sometimes is next week) – or when they don't cover the provider offering the service. –LPfi (talk) 12:47, 10 August 2020 (UTC)
Some of our information is out of date. We can't be certain enough of opening times to authoritatively say "opens in 25 minutes" or something of that sort, unfortunately. It is a good idea that we could implement one day when we have enough editors to keep all business listings up-to-date. --Comment by Selfie City (talk | contributions) 12:57, 10 August 2020 (UTC)
No. We should not say "opens in 25 minutes" when there is a "closed today" sign on the door. We cannot have editors check every shop and sight daily. This is another thing I really hate about dynamic info. The dynamic web page can continue giving such exact times as long as the web contract subsists, never mind the business went bankrupt last year. If it says "daily 10–15" and the clock says 9:35, who needs the "in 25 minutes"? (Let alone the "in 3 minutes" when they opened early.) When it would be useful is when there are umpty different times depending on day of week and the phase of the moon. Good luck parsing that info. –LPfi (talk) 15:03, 10 August 2020 (UTC)
But sorry, I did not check the map nor read carefully. In that context the dynamic feature is worthwhile. You could colour code or mask listings on the map depending on whether they are open (now or at a specified time). Also without the map, if you have a list of several listings, quickly seeing which ones are open, open soon or are open long enough would be useful. We don't necessarily need updated info on them all to do that. –LPfi (talk) 15:13, 10 August 2020 (UTC)
Good points. (The other issue LPfi mentioned should be mentioned in the Aggregators article, I'd say.) —Granger (talk · contribs) 15:08, 10 August 2020 (UTC)

I can definitely see this being a useful feature, and something to help us keep on par with other sites and apps. But how do we implement it? I would be a little wary of basing it strictly off of the hours field in our listings. For one, that field only has a semi-structured format currently, and there are a number of listings that only make sense when read in English such as unusual non-regular hours, listing times for tours rather than opening hours, bars that are open until "close", etc. More importantly, I think having a live "open/closed now" leads users to expect a higher level of accuracy; how do we ensure the data remains accurate? It's quite annoying when multiple providers have conflicting data for opening hours (Google Maps, OSM, Yelp, etc.), and we might just be one more source of incorrect data.

I'd suggest we consider whether the best implementation might be to not store opening hours ourselves, and instead obtain it from another data source. Since we already integrate with Wikidata, perhaps that's the best option. This would require each individual store/restaurant/bar to have its own Wikidata item, and I'm not sure if that's within the goals of Wikidata or not. However, it would bring a lot of advantages. Wikidata would force the data for opening hours to be stored in a structured format. It allows sources to be linked, which would make it easier to verify information and keep it updated. Wikidata items can be linked to other sites such as OSM, so an OSM POI could be used to auto-fill missing data in Wikidata, and updating one could prompt a corresponding update to be made in the other.

Unfortunately, a very quick search suggests that Wikidata doesn't yet have a property for the hours that a business is open. --Bigpeteb (talk) 17:17, 10 August 2020 (UTC)

And that property would need a complex data structure, so I wouldn't hold my breath. Items for each restaurant/whatever is indeed not regarded in scope at Wikidata at the moment. The worst problem however is how to make the updates. Would passersby and business owners update the wikidata data? Then we need to provide an easy to use user interface to that complex data structure. How do Google, OSM & al get the data? –LPfi (talk) 17:33, 10 August 2020 (UTC)
Unless there are some bots that are searching for potential changes (and OSM is historically rather anti-bot), I would guess that in OSM it's being entirely crowdsourced. :-/ (OSM editors are really dedicated to have done so much work manually.) In Google Maps, it scrapes websites and probably some other data sources that Google knows how to deeply index to made educated guesses that "there's a business here" and what its hours are, but a business can also claim ownership of a listing to set its info, and it's also partly crowdsourced as anyone can submit corrections to a listing. --Bigpeteb (talk) 17:52, 10 August 2020 (UTC)
Technically, you don't need to bother OSM with bots, since wikipedia sync's the maps into kartographer daily. However, AFAIK there's one unsolvable blocker - OSM doesn't have stable IDs for anything outside relations. So basically there's no way to reliably refer to a restaurant in OSM, even if it had the opening times. You'd have to do it the other way - create WD entry and then store the WD ID to the OSM entity... That could be semireliable, but basically you'd have to duplicate half OSM into WD to do this :-( -- andree.sk(talk) 20:31, 10 August 2020 (UTC)
I consider this an inessential feature, so if it can't be done now, no big loss. If you don't know what time it is where you are or where you plan to be and don't realize you can do a simple web search to find out what the current time is anywhere, if indeed your phone or computer doesn't show you because you're already in town, you have bigger problems than can be solved by this kind of feature. Ikan Kekek (talk) 21:19, 10 August 2020 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────

Dynamic opening hours can be a useful mean for travelers. But it is not easy to realize it.

  1. The up-to-date information must be stored in the article or at Wikidata (I am not sure if we can access to OSM data at the WMF's map server). This info had to be stored in a standardized uniform manner -- which is to specify -- and to be translated at non-English wikis. From my point of view the current way of storage at Wikidata (P3025 -- P3028) is not convenient. There should be the same listings as used at OSM and in the listing templates. So we had to discuss the property change at Wikidata. This will take time and needs some additional programming by the wikibase programmers team.
  2. At German Wikivoyage there is a tool named listing info with a button beside the listing edit button. This info shows additional notes for instance on online travel companies and is planned to show opening hints and features of the establishments.
  3. Calculating times is not an easy task. We need the current time (from the WMF web server -- I hope that it can be done with JavaSript), the uniform opening hours, and some country- or region-dependent data on time zones and daylight saving time. Some structures like country data are already prepared at the German Wikivoyage and can be expanded. These data are placed in the html code which gives fast access to JavaScript scripts.

-- RolandUnger (talk) 06:18, 11 August 2020 (UTC)

National taxi servicesEdit

Swept in from the pub

User:Vkem has been adding taxi listings, which I in some cases have found redundant.

In Finland most taxi businesses are small, and until recent years you either called them directly or, in cities, through a common call centre. Now most of the call centres have expanded operations, competing with those in other cities. Many now cover much of the country (including the countryside), but any one taxi car usually has an agreement with just one of them (or none), which means local market shares matter. Most of the call centre businesses now use smart phone apps and web forms in addition to the (premium fee) call centre.

The question then is, should the national call centre businesses (and their apps) be mentioned in all articles about destinations where they claim coverage, in region articles or only in the country article? What about rural areas with few taxis, should the apps be listed although taxis commonly are called directly (we might not have the numbers, but any local bar or lodging would have them)?

(I would also like to have more of a discussion in By taxi, on how the apps compare to other ways of finding a taxi, and any quirks and pitfalls to keep in mind when choosing whether to use them.)

There are now discussions started at

I suggest we keep the discussion here, until it is down to national details.

LPfi (talk) 06:55, 23 August 2020 (UTC)

Vkem made a point about recommended the apps: "I think they would give additional advantage for Wikivoyage users compared to having only phone call numbers of local taxi companies, regarding that many Wikivoyage users do not have English as their first language, and may have trouble pronouncing local language street addresses". –LPfi (talk) 07:02, 23 August 2020 (UTC)
Here is my longer explanation why I think that national taxi services info would be a good thing. At least in Finland, national companies have in practice usually a good coverage of the regions they claim to have coverage and I think that mentioning them in those specific regions that they serve is useful. But when it comes to having a coverage in whole province, I understand that mentioning them in all municipalities of that province may not be necessary. I suggest that there would be a link to taxi section of province in the municipality and town pages, to help people find the provincial operator of taxi. Regarding the national taxi operator issue, I think that a discussion what we should do with them would be good. Pros of national operators are that they often have handy smartphone apps helping to order them, being easier than calling the number of small local taxi company, phone calls may be difficult if English is not a native language and addresses are in local language. It would be in interests of local taxi companies to give them preference, but I think we should think practically and keep the interest of travellers first, and not prioritize small local taxi companies if large companies have better ordering process and equal level of service. The apps made to be similar than Uber make the taxi order process clear and give right away the price information of the trip. So I think that the national operators are useful to be mentioned if they have some specific cities they claim to serve and there is no reason to doubt that they would not work in practice. If necessary, I can do test searchs to proof the actual working of the national taxi service apps. I see little harm mentioning the national taxi services, especially if no information about local services is available at all in city or town page, I did not understand what harm would be done if they are mentioned to give additional possibility. National taxi operator info has been largely lacking in Nordic countries Wikivoyage pages. Also I would point that now when taxi fares are unregulated, taking taxi outside of app may cause unpleasant surprises regarding taxi prices, as far as I know we do not have yet fare cap to limit the fares to reasonable levels and it is a thing why knowing the fixed price right away when ordering the taxi through the app is good I think. --Vkem (talk) 09:20, 23 August 2020 (UTC)
I have made test searchs using Menevä app and 02 Taksi app and rural service is acceptable in most cases, they probably use local taxi companies when they seem to have short order times even in small municipality centers. For example in Huittinen, tested it yesterday and it gave arrival times 5 minutes in town center, 20 minutes in village, the service offered was regional operator Taksi Länsi-Suomi. If you like, you can try 02 Taksi app and Menevä app to do some tests how fast the taxi should arrive and see if the service is goof enough for Wikivoyage users. For Sweden similar services are TaxiKurir app (claims to cover also Norway and Denmark) and SverigeTaxi app. --Vkem (talk) 09:36, 23 August 2020 (UTC)
I have two main concerns:
  1. Having information in many articles means it has to be maintained in many articles, meaning severe workload (if not done by bot) and risk of obsolete information remaining.
  2. Listing the national operators but not local businesses in an article gives the impression these are the ones to use. Editors and readers will make less effort to find the local ones. Of course, the local businesses can in many cases be used through national operators, but if only a minority of local businesses can be reached that way, it might have been better to let the traveller ask locally instead.
(There is also the issue of furthering centralisation, but I suppose ttcf overrides that concern.)
My view is that the national operators (call centres) primarily should be described in the country article, alongside the info on how the taxi business in general works in the country. If only some of the national operators are usable in a region, there could be some discussion, but (depending on whether the operators' coverage correlates with region borders) also this could be on the country level ("X operates mainly in the capital region and in some big cities elsewhere (A, B, C, D)", "Y originates in the surroundings of C and is the dominating operator there, with some presence in the E and F regions", "Z strives to cover all the country, but often has only a few cars in each town, except B, where Z dominates").
If the national operators have good websites (eh, usable), the travellers themselves can check details on local level, without us having the burden to keep the information updated. Some are mostly in Finnish, so a few lines on how to find the relevant info may be needed – still easier to maintain.
I think info should be included on local level only when information is stable or the local taxis for the place are hard to find otherwise, or there are other significant quirks; old local firms in rural areas tend to be reliable (unless the legal reforms changes the playground), as the market is stable and drivers have chosen their profession.
On the language issue: should we advice travellers to offer specifying addresses by SMS if there seems to be a risk of misunderstanding in a voice call to a specific business/driver? These tend to be normal GSM numbers (in contrast to the call centre ones).
LPfi (talk) 07:57, 28 August 2020 (UTC)
How different is this from a national chain store? Do the national call centers feel more like McDonald's (in which case, we'd exclude them from almost all articles and add only local taxis) or more like a public inter-city bus service (in which case, we'd include them and their competitors)? WhatamIdoing (talk) 19:27, 28 August 2020 (UTC)
@WhatamIdoing: For Finland (I don't know the other countries well): The independents are usually drivers from the old tradition, i.e. as mainstream as those you get through the call centres (novelty taxis are a different matter). I'd say public bus-service is the relevant comparison – but regional bus service. Yes, you can often use an intercity coach for regional transport, but if you rely on those, much of the region is unreachable. If you list just the intercity coaches in Get around, readers may believe these are the only options. In the countryside I suppose a significant part of the local taxi businesses (all of them at some destinations) count on being reached by phone and prefer avoiding the non-trivial expenses (thousands of euros) of making an agreement with a national operator. –LPfi (talk) 07:45, 2 September 2020 (UTC)

New editing tools for generating article contentEdit

Swept in from the pub

I've created two new editing tools to help make it easier to generate new listings and "Go next" content for articles. I'd be grateful for feedback to determine if they are useful to others and worth spending further time developing.

  • Nearby Attractions - This tool will generate Wikivoyage listings for a destination by fetching Wikipedia articles for places located within a specified radius of the destination and auto-generating Wikivoyage listings from those articles. The resulting listings can then be reviewed, edited, and copied to Wikivoyage. While not all Wikipedia articles merit a listing, I've found it to be a good way to quickly generate a list of candidate listings since Wikipedia has articles on museums, famous landmarks, and other places that should be included in Wikivoyage but are often missing.
  • Nearby Destinations - This tool will take as input a destination name (using the Wikivoyage article name) and then return all nearby Wikivoyage articles formatted as a {{mapgroup}} that can be included in the "Go next" section of an article. See Culver City#Go next for an example. Currently most Wikivoyage articles lack useful information about neighboring destinations, so this tool attempts to make it easier to provide that information.

For both of these tools manual review and editing is still required. In most cases the auto-generated listing description from Wikipedia needs to be updated to make it more relevant for travelers, to avoid SEO issues due to the same content being used by both sites, and to remove invalid markup. For both tools, editorial judgment is required when determining which of the results generated by the tools are appropriate for inclusion in Wikivoyage.

Bug reports and suggestions for improvement would be greatly appreciated. The tools work for me, but I've only tested using the Chrome browser. Also, I couldn't immediately find an appropriate article in the Wikivoyage namespace to mention these tools in, so are there suggestions of where to list editing tools like these, as well as other tools like WV:AWB, or if not, would there be any objection to adding a new page like "Wikivoyage:Editing tools"? -- Ryan • (talk) • 05:11, 29 August 2020 (UTC)

I think it's a great way to find listings though as you say it needs to be edited heavily. Also many of the listings captured by the tool are not see. I used it on a article where it found a stadium (do) and airport (get in). Presumably it would capture shopping malls which would be buy and all other kinds of non-see listings. The maximum radius for the "Go next" tool may need to be increased. 20 km (12 mi) doesn't seem to be enough for rural or sparsely populated areas. It hasn't returned anything on a few articles I tried. Thank you for plunging forward and making these tools. They have lots of potential. Gizza (roam) 05:32, 29 August 2020 (UTC)
Thanks for the feedback, I'm glad to know that at least one other person thinks it has potential. The maximum radius values are unfortunately limitations of the underlying Mediawiki APIs that the tool uses, so rural areas will be challenging - I'll see if there are ways to work around that. I've also been debating ways to change the generated listing types for results so that a park or stadium can be changed to a "do" listing, but didn't want to pursue that before first soliciting feedback; I'll add it to my TODO list. -- Ryan • (talk) • 05:40, 29 August 2020 (UTC)
These are useful tools. I used Nearby Attractions to add a couple of listings to Coll. It would be good if it added the lat/long to the listing in addition to the Wikidata field - this is useful for the full page map. Of course, Wikipedia has loads of articles of no interest to us - I tried for Helensburgh and one suggestion was a football club disbanded in 1924 and another was a railway station that closed in 1964. The tool copies some templates which don't work here.
Nearby Destinations is also useful - I have used it to expand Go Next in Helensburgh. In this case i am tempted to remove some of the destinations, which although geographically nearby, are a long distance by road due the the river estuary - 10 miles for birds can be 100 miles for drivers. For editors (not readers), I would be interested in a tool that did something similar showing nearby WV articles in other languages. AlasdairW (talk) 21:41, 29 August 2020 (UTC)
Nearby Attractions uses WP article names for the search, and so isn't great for looking for attractions in huge cities - I couldn't search for a district. The maximum results is far too small - for somewhere like Edinburgh it needs to be 5000, as there are 148 articles in w:Category:Category A listed buildings in Edinburgh alone. AlasdairW (talk)
Thank you for the feedback, I'll try to get the lat/long included and also increase the maximum number of results within the next few days. Regarding the issue with districts, instead of using a Wikipedia article for the target lat/long I should be able to switch to using the Wikivoyage article, which would then allow you to search using Wikivoyage district articles; I'll post an update when that change is made.
As to templates in the resulting code, it may be tricky to try to strip that out - if I just remove it then useful information will go missing in cases where the original content was something like XYZ park is a {{convert|100|acre}} park in...; my inclination would be to leave it as-is and rely on editors to update the text before copying to Wikivoyage, but other suggestions would be appreciated. -- Ryan • (talk) • 22:40, 29 August 2020 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────These are fascinating tools! They should be used with care, because many articles for defunct businesses, buildings and stations come up, for example, in a search for Wikipedia articles related to w:East Village, Manhattan. You also get many false positives in Nearby Destinations searches for big cities. This one is for New York City: Manhattan/SoHo Manhattan/TriBeCa Manhattan/Chinatown Manhattan/Lower East Side Manhattan/Financial District Manhattan/Greenwich Village Manhattan/East Village Manhattan/Gramercy Flatiron Manhattan/Chelsea Brooklyn/Downtown Manhattan/Midtown East Hoboken Manhattan/Theater District Brooklyn/Gowanus and Red Hook Brooklyn/Williamsburg

As you can see, all but Hoboken are neighborhoods of New York. One useful tool might be to start the Nearby Destinations x-number of km outside the city (maybe at least 5, in this instance) and be able to extend them to any desired extent (such as 100 km, maybe more), or to somehow enable the elimination of articles about district articles from a search of destinations close to the parent city article. Ikan Kekek (talk) 00:04, 30 August 2020 (UTC)

@AlasdairW: The "nearby attractions" tool has been updated to use the Wikivoyage article to look up the target coordinates, so it should work better now for finding listings within a Wikivoyage district. The maximum number of results is a limitation of the Mediawiki API, so that's not something I'm able to easily increase, although if the API is updated I'll update the tool as well. The other suggestions that people have made are all much appreciated and currently on my TODO list. -- Ryan • (talk) • 04:52, 31 August 2020 (UTC)
@Wrh2: Thanks. It is useful to be able to use Wikivoyage names, particularly for districts and articles where we use different syntax in disambiguation. It is a pity that the API has such a low limit. I found that I got 49 listings with a radius of 0.7km in Glasgow, Edinburgh/Old Town, Edinburgh/New Town and Washington, D.C./National Mall. Trying rural destinations was much better, and in several I got less than 49 within 10km. A possible work-around would be to allow an offset in the lat/longs, to make 1 or 2km "jumps" around a city centre, but I realise this is extra work that might not be used by many editors.
To follow up on an earlier point, I was not expecting the tool to automatically strip out templates which don't work, but the instructions could point out that most templates will need to be removed. Thanks for creating useful tools. AlasdairW (talk) 21:51, 31 August 2020 (UTC)

I like the {{mapgroup}} effect, didn't know that template existed before. I thought that the consensus on en-wv is to not have map markers for Go next items, and some of my edits were already reverted based on that argument. However, the {{mapgroup}} template does generate map markers, so doesn't that conflict with the consensus/policy? 87.74.196.200 10:11, 31 August 2020 (UTC)

@Wrh2: Perhaps it might be useful to exclude any Wikipedia articles in categories within the w:Category:Former entities hierarchy. I don't know how practical that is with the API, though. Powers (talk) 18:22, 3 September 2020 (UTC)

Return to the project page "Listings".