tl;dr Everyone will need an API Key starting Feb 1 Feb 7th in order to use TV with any script. They can be claimed here for free. The first month is free for every user. Once that first month ends the price is discounted at $5/month for users that sign up before Mar 1 and $7.99/month otherwise. NOTE: If you have submitted at least 10 reviews on TV (as of 1/20/19), you will get a permanent key for free, this covers hundreds of users who helped build TV & is also extended to a # of community members who have contributed their time/expertise elsewhere. ___________________________________ Background The original post for the new API changes is here. This post is going to serve as an update on the plans laid out in that one and to explain some of the process/decision making that came about when actually testing those plans on real users. As I dug into MTS, Forker, & updating TVJS, I realized that my original plans for TV's API weren't going to work as intended. There was no good way to split the # of requests for a portion of a day across the scripts that needed it, given that scripts request data in different ways, cache it differently, and have different loads (MTS takes much more data than Forker, etc etc). All things TV itself can't affect or change and will be highly variable across different use cases. On top of that, and most importantly, in testing partial day access it became clear some of the very problems the API is meant to alleviate (what MTS/Forker users currently suffer with outdated/missing data) is going to reoccur every day. Once a free daily quota limit is reached users will have some portion of data cached but not receive any new data. This means that throughout the day the information previously sent from TurkerView gets progressively worse. This leads to a poor experience for the worker & puts script authors in the position of being forced to explain why data inconsistencies manifest themselves. It's confusing, unreliable, and it'll lead to bad decisions on HITs/Requesters, never mind the technical knowledge burden that places on users. So, better ideas aside (they're welcome if you've got them): The Full Plan Since this is a divergence from the original plan I've done this: anyone who (as of 1/20/2019) has submitted 10 or more reviews already on TV has permanently upgraded free API keys. I'd absolutely appreciate it if they chose to support the site's infrastructure, but it won't be required for the API and all access will remain free. February 1st February 7th will mark when you need an API Key in order to access data through the API. TVJS will update today (1/21) & Forker tomorrow (1/22, TPG pending). MTS will get code pushed to it after I get some reasonable load data returned so I can make sure there is enough infrastructure in place to support it. Once/if MTS' code is accepted we'll transition. Again target date is Feb 1st Feb 7th, this is a 'best laid plans' scenario and should an issue arise I will delay auth token enforcement to preserve worker experience if need be. Instead of doing small daily quota limits, we are going to provide users a month of access for free to start out without any of the complicated limits previously planned. It'll work for all scripts & users wont need to worry about caps as long as they're using any of the apps currently supported in normal ways (if you're not in MTurk PRE purgatory, you'll be fine). If you claim your key before Feb 1st you'll be good until Mar 1 (so over 30 days) The subscription price will stay at $5/mo until Mar 1st, but along w/ the trial window period will be subject to what is actually sustainable after that (most likely you'll see it revert to the original intended prices Mar 1st & go to 2-weeks w/ a large discount + time extension for folks participating on the platform). For now, that's nominally known as the "price of a cup of coffee" on the internet, I think it's fair for both what TV represents & provides. The API is a convenience feature, there's nothing you get from it you can't really find on the website itself & I plan on adding a lot of extras/perks for supporting the website in the future (almost guaranteed before Mar even hits) Return Reviews, at least for the foreseeable future, will not be included in any of this. I think they're more useful to newer users & I'd like to keep them open access if possible. This is simpler for all parties. It takes care of newbies (many of whom don't make it to a month on the platform anyway), old folks returning to the trenches, and gives everyone the same fair shake to figure out if TurkerView is useful & worth supporting for them.