participedia / api

Website and API for Participedia V3
https://participedia.net
MIT License
20 stars 14 forks source link

redirect hidden and/or converted articles to new urls #665

Open ascott opened 5 years ago

ascott commented 5 years ago

slack convo reference: https://participedia-team.slack.com/archives/G1ERG6UKH/p1562090390003900

is there a way to redirect this page https://participedia.net/organization/4902 to this page https://participedia.net/method/5397? I hid the former and converted it to a method (to conform to PPedia definitional standards). this is one of the most accessed pages according to Google Analytics, and we might be putting people off by returning the 'this page does not exist' message. I'm going to try and find and change all the links, but having a redirect would save me a lot of work (and it would be great if we could do this for the other entries I've hid & converted)

todo:

scottofletcher commented 5 years ago

Question: can you folks replace the 'N' with a 'Y' in column E so I know which ones are completed? I'm adding more as I go along

ascott commented 5 years ago

@scottofletcher i will do these all at once, so let me know when you are done adding them.

scottofletcher commented 5 years ago

this is an ongoing thing though - there are always going to be entries that need 'deleting' (either because they were entered in the wrong format - ie. a case instead of a method - or because they were a duplicate of an existing entry) and redirecting

ascott commented 5 years ago

hmm, ok, i think we should only add the redirects for the urls that are being requested most often. if it's a url that doesn't get many hits from google or other referrals than i think we can let the user use the search to find that entry. @jesicarson do you have any thoughts about this, is there a better solution? should we prioritize this?

jesicarson commented 5 years ago

I'm not sure I fully understand what the problem is. Is it: a) that "hidden" entries are being indexed by google and bringing new visitors to a 404? or, is it: b) that there's a possibility of people having saved an older entry url and returning to it and getting a 404? Or something else? thx for clarifying before i weigh in on priority level.

ascott commented 5 years ago

my understanding is that both a and b scenarios are happening. @scottofletcher can you confirm? can you share the numbers from google analytics showing how many requests we are getting for hidden/deleted url's?

scottofletcher commented 5 years ago

yes, that's correct, there are 3 things happening:

1) hidden entries are indexed by google 2) people have linked to now-hidden entries in the components fields, methods/tools fields, and organization field (for example, Martin King entered duplicate entries on Sciencewise and then linked half of the 30+ components to the hidden one, and half to the visible one) 3) people have hyperlinked to now-hidden entries both in ppedia entry narratives and in external documents (eg. that recent piece we reprinted by Selen and John linked to an entry that we had hidden. I had to un-hide the entry so people wouldn't end up with the 404)

This is likely going to keep happening as people (including myself!) enter duplicate entries, or entries on extremely similar content (eg. Dynamic Governance is the exact same thing as Sociocracy, so it would only stand to reason to redirect to the more common term and include something like "Sociocracy - also known as Dynamic Governance - etc." in the narrative. Wikipedia does this for similar content as well.)

there are probably other use cases that I'm not thinking of right now, but my feeling is that this would significantly increase user satisfaction with the website. no one likes to hit a 404 and it can be off-putting if you encounter it more than once (which I think might end up happening - as you can see, the list of hidden entries is already very long and will continue to grow).

not that this is an emergency, but if we could figure out some way to streamline this, I think it would be a great idea. I don't suppose it's easy enough for an idiot like me to understand? if you think I could figure it out, I'm willing to learn and then save you the hassle

jesicarson commented 5 years ago

Ok given all this it seems that some thought should be put into addressing the workflow/ problems resulting from "hiding" entries. I would say this could be moved up to P1, but after we implement faceted search for methods and orgs. I'll add it to the to do list. thanks folks!

scottofletcher commented 5 years ago

works for me! and let me know if I can help out - I'd be happy to integrate this into my content cleanup workflow

ascott commented 5 years ago

i've set up some more reporting around 404's and saved a report to show the most common 404 pages.

the quick overview is that we are seeing ~2% of pageviews as 404 pages. the report also shows a bounce rate for 404 pages as 64%, while the overall bounce rate is 74%. this tells us that users landing on the 404 page are either navigating to the home page or using the search to find what they are looking for and staying on the site at a higher rate than other traffic.

given this, i would suggest we don't prioritize setting up redirects at this point.

also, i set up some additional rules for the google bots so they don't index old urls anymore.

jesicarson commented 5 years ago

This sounds good, thanks Alanna.

scottofletcher commented 5 years ago

ok, I agree that we should de-prioritize this, but I still think it could be handy if we start to see more duplicate or miscategorized entries