Wikivoyage:Roadmap/Enable listings reviews

Pop up visualization

Review popup draft

I've created a mock up of what a review popup would look like. Users would click on the "review!" text next to a listing, this editor would pop up, and they could then select their star rating and leave a review, as personalized and as detailed as they like.

Reviews would then be listed on an external "reviews page," which would not be part of our normal, anyone-can-edit wiki. The reviews page would be a resource to other travelers looking for more detailed (and possibly more trustworthy) information, a resource to wiki editors looking for material and recommendations for the guides, and a way of personalizing the Wikivoyage experience for users, who might otherwise not be drawn in to our fairly impersonal looking project.

Users can also tick a box to alert Wikivoyage that the place has closed. This is hugely important. We could set up a "recentchanges" of sorts just dedicated to reviewing this alert, to help keep our guides better up to date. We lag way behind dedicated review sites in removing closed listings, and being up-to-date is supposed to be our strength. Users are often hesitant to remove listings (to remove others' work), but wouldn't feel nervous about ticking an alert box.

You might notice that the review pop up looks a lot like the listings editor that I mocked up. I'm hoping that a similar interface might allow the listings reviews (which have a low barrier to entry) to act as a gateway drug for further editing on the wiki.

Thoughts? --Peter ^Talk 20:37, 1 February 2013 (UTC)

I was not a big fan of this concept when it was introduced, but your mockup strikes me as appealing. LtPowers (talk) 21:13, 1 February 2013 (UTC)

Yes simple and non intrusive. Travel Doc James (talk · contribs · email) 22:35, 1 February 2013 (UTC)

Looks good. Very simple is what we need. So we are not planning to add all the other options that TripAdvisor does, like "type of traveller", month stayed, separate star ratings for location, staff, facilities, value, etc? Also, are we planning on adding a link to the pre-existing reviews from each listing, or at least a visualisation of the number of stars (which could be a link to the list of reviews)? JamesA ^>talk 04:24, 2 February 2013 (UTC)

I'm personally of the opinion that the broken down ratings on TA aren't of very great value, and would place more value on simplicity—the reviews themselves should explain what factored into the rating decision. I'm still not sure how would be best to visualize the review content in the travel guide, but yes, the rating and link to reviews is a must. The "review" link with the popup reviews editor could fill the roll of the latter, since it has a direct link. We could possibly have the star ratings popup as hovertext over the listing name. Other ideas? --Peter ^Talk 19:05, 2 February 2013 (UTC)

The mockup looks good. I agree the broken down ratings frequently aren't very useful, but I find the headline star rating on its own can be suspect and I don't always want to wade through a lot of reviews. An idea I had was to provide a short list of features and allow reviewers to check 1 or 2 that they most liked about the place. For a restaurant, it could be: Food, Service, Price (or Value), Ambiance, Location. I'm not sure if this would work or if it would be an improvement, but I thought I'd toss it out there. -Shaundd (talk) 05:34, 6 February 2013 (UTC)

Review page visualization

Example reviews page

By the way, I'll work on a mock up of what the separate reviews page would look like. I intend for it to have a basic, clean look, that would distinguish it as much as possible from the messy ad-ridden hell holes you see on other review sites. --Peter ^Talk 20:38, 1 February 2013 (UTC)

I'm not in the slightest bit wedded to this mock up, but I figured it would at least get the ball rolling. I'm not a web site design guy in any way shape or form. My critique would be that it looks a little too similar to our main site, and that something strikingly different might make sense, and better emphasize the distinction between our travel guide content and auxiliary content like this. --Peter ^Talk 23:12, 1 February 2013 (UTC)

I think this mockup is great! The user avatars could pose issues, as WM wikis don't have this feature and it'd have to be hard-wired across the site. Another thing I'd like to see is the reviewer's location/country below their avatar, like on TA. This can be easily set in your preferences. In terms of differentiating this review "area" from our main editing site, you could use background colours. I notice you've already changed the colour of the page title, which is a good first step. JamesA ^>talk 04:24, 2 February 2013 (UTC)

I have now messed around with the background colors, and added a "reviews" caption under the logo. I also put the listing details up in the header, with an edit button, so casual reviewers could help keep the details up-to-date while leaving their message. I envision that edit link opening the form listings editor popup. I also added reviewer's location (which should probably link to the travel guide for that place) as well as their number of reviews, which is a key piece of information for judging their trustworthyness! I added a "helpful?" button to the reviews, to allow readers to up-vote, down-vote each others' work, and then allow reviews to be sortable accordingly. --Peter ^Talk 19:02, 2 February 2013 (UTC)

Registered users

I think this feature should be limited to registered users. Upon clicking the "review" link, anons would get a popup that offers a link to existing reviews, and asks that they register an account to allow them to begin creating their own. That would help us monitor who exactly is writing reviews, while also encouraging people to establish a broader tie with the site. --Peter ^Talk 20:44, 1 February 2013 (UTC)

I totally get where you're going with this, but it's a little strange to think of letting IP users edit whatever the heck they want in our guides, but not leave a simple star-rating. LtPowers (talk) 21:12, 1 February 2013 (UTC)

My one concern is that the low barrier to entry could make it easy for business promoters to game the system. On wiki, there's enough challenge in travel writing to quickly expose promoters. Star ratings, on the other hand, could become meaningless, and impossible to track if gamed.

On the flipside, the other benefit I'm hoping for is to rope in new contributors by introducing more of a social atmosphere on the reviews side of things. --Peter ^Talk 22:03, 1 February 2013 (UTC)

We could simply present the data from registered users much more prominently or require that they be reviewed by a confirmed logged in user before counting. We want everyone to be able to contribute at least just a bit and this will not change WV itself.Travel Doc James (talk · contribs · email) 22:29, 1 February 2013 (UTC)

I completely agree with Peter's concerns regarding IP reviews, and would also prefer registered reviews. They do this on TA, and that's worked pretty well. And it's a good way to get people prepared for proper editing. JamesA ^>talk 04:24, 2 February 2013 (UTC)

We could trial both options initially. If anonymous reviews are useless we can than restrict to only registered users.Travel Doc James (talk · contribs · email) 07:21, 3 February 2013 (UTC)

Or maybe restrict star ratings to registered users? Since star ratings involve the least personal effort, they are the easiest to game anonymously. One other problem with anon reviews, though, is that # of reviews is one of the best ways to judge a user's reliability—if they just have one glowing review for some business, there's more room for suspicion that they are a paid promoter. Similarly, a lone attack on a business could be a disgruntled employee. --Peter ^Talk 18:22, 3 February 2013 (UTC)

I'm fine with restricting reviews and star ratings to registered users for the reasons mentioned above. It does concern me a wee bit that it will reduce the number of reviews (possibly good, as well as bad) -- and volume of reviews is what these rating systems really need to be useful. -Shaundd (talk) 05:40, 6 February 2013 (UTC)

Integration with Wikidata

I think its imperative that any revamp of our listings system, also involves a gradual migration of listings to Wikidata, who knows, it might even attract developer assets from the project once 1.0 is out the door, since it would be a pretty significant opportunity to see it used, and used well. Sertmann (talk) 22:04, 3 February 2013 (UTC)

Could you explain in greater detail how this would work? Having our listings information on Wikidata is something I've seen proposed elsewhere, but I don't understand what it means, or how we would go about it. --Peter ^Talk 22:28, 3 February 2013 (UTC)

Wikidata is intended as a repository for cross-wiki information, much as Commons functions as a repository for cross-wiki images and files. In theory, using Wikidata would allow us to share listing ratings across all language versions of Wikivoyage. Note, however, that I said 'in theory'. At the moment, Wikidata only has the capability to store interwiki links for Wikipedias; additional functionality is well in the future. LtPowers (talk) 00:57, 5 February 2013 (UTC)

Additional functionality is being rolled out fairly quickly. Further details are here [1] Travel Doc James (talk · contribs · email) 02:38, 5 February 2013 (UTC)

Gotcha. Sharing listings content between language versions is absolutely something we should do. Syncing details like phone numbers, open hours, closures—this would be extremely useful. --Peter ^Talk 02:49, 5 February 2013 (UTC)

I see three problems. Sharing the listings requires them to be very well standardised; free text cannot be shared across languages. And the user interface of Wikidata is very different from ours, which means a barrier for Wikivoyagers to change listings. I am often not even able to log in on Wikidata (javascript issues?). Interwiki links are largely maintained by bots, but changes in listings require as broad a range of contributors as possible. Will changing opening hours and telephone numbers be easy for casual visitors? And what about watch lists? We probably cannot trust Wikidata folks to notice vandals changing such data. --LPfi (talk) 08:59, 6 November 2013 (UTC)

random thoughts on reviews page

In my early years with Wikitravel I voted agressively for having a place for reviews--but now I'm not sure :-)

1. Why would it be easier / more attractive for an average user to post a review to us, instead of Tripadvisor (or Yelp in the US)? Given that they not only already have so many reviews, but also a convenience of mobile app, which is always easier to use than a [mobile] web site. I can understand if ours support offline writing/adding, but it's a challenging goal. And not sure we have enough app development talent for creating a competitive review writing experience.

2. Another option to differentiate is to provide a set of pre-defined questions (like "Food", "Service", "Ambience", "Value for money") and some predefined answers (like "boring", "formal", "experimental" etc)--all to make leaving a review much faster and easy experience compared to writing a coherent text (on mobile!). It is easily reproducible, but can give us some initial edge if done right.

3. Foursquare becomes more and more a competitor here. Both with its very ubiquitos coverage, "tips" which are (a) more popular on top, (b) much easier to write compared to a traditional review -- and agressive attempts to derive "rating" from various factors (FSq have many, we should admit). So the question is the same: why most people should write to us, not to FSq? --DenisYurkin (talk) 19:36, 6 February 2013 (UTC)

I don't think we're trying to compete with Tripadvisor. We're just trying to give users fewer reasons to leave and resort to Tripadvisor while they're reading Wikivoyage; it's an added bonus-feature onto our carefully curated travel guides. LtPowers (talk) 18:27, 16 February 2013 (UTC)

We won't compete right away (and this is a reason to get moving quickly), but I think we offer two big advantages. One is freedom/ownership of content, and the other is that our travel guides are valuable independent of any reviews—we already have a hook to draw in reviewers, if they're already reading our guides.

By the way, since you mention Foursquare, we're not the only ones working on this basic plan: see "How Wikimedia and OpenStreetMap can help us build an alternative to commercial Web 2.0 services" [2]. --Peter ^Talk 20:30, 20 February 2013 (UTC)

I think many of us are here because it is owned by a not for profit rather than a for profit. I think that will draw in contributors. Travel Doc James (talk · contribs · email) 06:15, 21 February 2013 (UTC)

Turn on article feedback

Do we want the article feedback tool turned on on WV? We can add it to pages as we see fit [3] This will be just to all can see how it works and get comfortable with it. If there is consensus I will push to WMF to make it happen.

Support

1. Yes yes I know it is not a vote but a consensus. Makes it easier to present something to the WMF though. Travel Doc James (talk · contribs · email) 18:23, 26 May 2013 (UTC)

;Oppose

Discussion

Erm, let's not do the support oppose thing. I think our decision was to add {{consensus}} to the top of threads that produce an actionable consensus that needs WMF help. Where did we discuss this before? I thought we already had a consensus? --Peter ^Talk 18:47, 26 May 2013 (UTC)

I don't know whether we have a consensus, but I am using article feedback on WP. There are serious issues, but it can be very useful occasionally.

• It should only be turned on if you are going to watch it, or it is a complete waste of effort. Feedback that no-one looks at is an insult to the person who takes the trouble to give it.
• At present there is no way to put it on a watchlist, so you have to keep your own list of which pages you are watching for feedback. This is a pain in the butt.
• A lot of the feedback is really poorly expressed, because the tool does not direct the user how to give useful feedback. As a result more is useless than should be the case. Most users don't have a clue what would be useful, and here I am referring to the ones who actually want to give useful feedback.
• The feedback analysis page or whatever you may call it, does not provide all the options you need to classify the feedback.

A great idea with poor implementation, but not entirely useless. If they are still doing any development, it doesn't show. • • • Peter (Southwood) ^(talk): 11:16, 3 August 2013 (UTC)

We can use the feedback tool in two different ways 1) to allow people to give feedback on the article itself (agree that this is only useful if watched and that is how Wikipedia uses it, feedback is only on pages were people request it). 2)allow people to give feedback on hotels and restaurants. Could be useful if watched or not. Will take time to build up. Could stimulant more people to get involved. Travel Doc James (talk · contribs · email) 08:57, 5 August 2013 (UTC)

I've got the impression the feedback tool was developed to get readers to participate, not to get useful feedback. At least the evaluation was much along those lines. I hate such abuse of visitors' altruism. Unless we really can use the feedback, we should not ask for it (feedback on discussion pages is usually acted on, so no problem there).

How would the feedback on listings work? Is the tool useful for that? How does it compare with the reviews suggested earlier?

--LPfi (talk) 09:12, 6 November 2013 (UTC)

The feedback on listings could be useful for removing bad restaurants/bars/hotels that get a large number of bad reviews. Since we have a limited number of businesses that can be placed on any list, this might be useful for places that regular editors are not familiar with. I like the idea. Altaihunters (talk) 08:01, 28 November 2013 (UTC)