this post was submitted on 13 Jul 2023
218 points (97.4% liked)

Reddit

13435 readers
3 users here now

founded 5 years ago
MODERATORS
 

It appears API rate limiting has effectively killed these alternatives. You essentially get nothing but “Too many requests” 429 errors.

Lemmy sadly does not have the active niche news and discussions I want. But now nothing can be read without going to Reddit. I hate Spez

you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 2 points 1 year ago (1 children)

Just wait, the devs will find a way.

[–] Cyyy 3 points 1 year ago* (last edited 1 year ago) (1 children)

difficult till almost impossible. i recently started coding my own client for reddit (i wanted a way to still get nsfw content when thirdparty clients go dead), and reddit is fucking annoying as hell. everything you do.. they smash issues towards you. every time shitty 429 errors (rate limiting) even if you are logged in. just using a useragent of a normal browser gets you ratelimited. so spoofing a normal browser don't works. sending a bit too much requests (like the Stealth app who is basically a parser for reddits website) gets your device ip banned. if you then open reddit in browser, they smash a error in your face that basically says "fuck you, we think you are a bot. gtfo.".

bypassing this rate limiting is almost impossible even if you try to spoof a browser. i tried the last few days and just gave up because its too annoying and buggy. the whole system of reddit is so annoying as a developer to work with.

it annoyed me so much, that i now think about making the app not for reddit but for lemmy. because reddit sucks. hard. fuck them.

[–] [email protected] 3 points 1 year ago (1 children)

I was trying to be positive,but after reading their announcement on github not so much anymore. Thank you for explaining in deep way is not possible to find a workaround.

[–] Cyyy 2 points 1 year ago (1 children)

the biggest issue is that they detect thirdparty clients coded as a website parser on their server and just block you. and bypassing this isn't really working well because of the rate limiting.

example: i just did send 3 requests where i first logged in, then asked for the recent posts of a sub.. and already after this 2 requests i got rate limited by error 429 and couldn't send any requests anymore.

so even just requesting the recent posts in a sub is an issue (with spoofed browser useragent). if you use a "legit" useragent it works better, but reddit exactly knows you're using a thirdparty client and can block or ban you whenever they feel like. so it's not really a good solution because every minute reddit could hit the killswitch. just not worth the time to develope a app if it gets killed off then anyway.

[–] [email protected] 3 points 1 year ago (1 children)
[–] Cyyy 1 points 1 year ago* (last edited 1 year ago)

based on the knowledge, i would say nah. but maybe there is somewhere on the internet a genius who can somehow gets it to work stable enough.. who knows.

i just checked the announcement of libreddit and it seems they used the same json endpoints i did for my project, so they probably encountered the same issues i did. and if they didn't found a good solution yet (even after working way more with the API and endpoints than me).. dunno.