Hacker News Reader

Comment by mellosouls | Hacker News Reader

Where is the evidence/documentation for this claim; ie that it's a general policy? I'm not disputing it but its just a tweet that doesn't provide context.

View on HN

Here is how you verify it yourself,

    curl https://twitter.com/zarfeblong/status/1339742840142872577

Still possible to curl, just looks like it is a bit more obnoxious now and requires parameters like your browserAgent and cookie to be set. Look in your network tab for requests to an endpoint like this[1] with a lot of query parameters after.

1- https://twitter.com/i/api/2/timeline/conversation/<twitter status id>.json

Any HTTP request you can make with a browser can in principle be made with curl. The issue is that seeing content now requires you to jump through lots of hoops, whereas it was previously openly accessible by default.

That's true. I agree it is a worse change and obnoxious for twitter to do (unless there is some easy way I/we just aren't aware of that lets you easily get the tweets as before).

As pointed out by https://twitter.com/magusnn/status/1339833122456662017, you can impersonate GoogleBot and get an eye-watering 500kb of HTML to read a single tweet.

Turn off JS then go to twitter.com. There's the evidence/docs.

Everyone is missing the question asked here. The poster isn't asking for evidence of the bug. The poster is asking about the assumption that this is Twitter's policy. This could also be a bug in how user-agents are handled.

Reply on news.ycombinator.com