Last modified: 2014-02-12 23:47:26 UTC
Somewhere in the last few days, changes landed on MobileFrontend that make the "random page" feature very confusing. There seems to be some kind of error message, as though it's failing to load the page and then offers two choices: Read this article? [Try again] [Yes] "Intellectual growth should commence at birth and cease only at death." Albert Einstein "Try again" appears to attempt to fetch another random page, but equally failing -- only the page title is shown, just as the first time. "Yes" appears to actually try to fetch the page (again? first time?) and loads it. The Einstein quote never seems to change, and is a bit confusing as to its purpose and context.
This appears to be commit 73ed319dbbaba465a3943f80e19ee375e3b1affa by jon. "hijack random button and load content dynamically in beta this makes random much more playful and fun" Since the content doesn't actually get loaded until you decipher the buttons, I'm not 100% on board with the fun part. :)
I suspect it is confusing as the behaviour has slightly changed from what you suspected :) See bug 30842 for the problem it is trying to solve. The random api only returns the article titles. Ideally it would return the summary as well. I wanted to do this to show holes in the api. I think this behaviour is still better than the previous as it at least allows you to flick through article titles more quickly to find something interesting and actually reduces load on server by not loading entire pages. I agree it could be better but for time being it's just beta ;-)
Sounds like we can improve the behaviour/UI but don't think there is a bug here.
*nod* I'd recommend a couple language changes: "Try again" => "Random" "Yes" => "Read full article" If we could pull the summary instead of the Einstein quote, so we're showing at least part of the intro, that might be super awesome. But will take a second API hit at this time.
Agreed - when we can pull the intro instead - the read full article button would then also be redundant! Guess we should bring this up in the next iteration planning session.
