HN user
It's a (now more than 20 years old) TV tracking website and community.
I've been using Claude 4.5 Opus (now 4.6) more and more these days modernizing and redesigning sections that haven't been touched for a decade or two. I don't trust LLMs much, but by breaking the work into small, self-contained tasks and testing constantly - I'm making surprisingly fast progress.
20 years and counting, working on https://next-episode.net (it's a TV/Movies tracking website and community).
I've dedicated this week to some maintenance tasks that are long overdue (mainly modernization of the code and the database), kinda delaying the inevitable (which is to work on harder tasks in my todo - like adding features to the mobile apps).
I can totally relate.
In addition to "no one to bounce ideas off" - no one to really share the pain and responsibility with when there are problems. It's all on you.
Thankfully, my users are very generous with their feedback, so when I do in fact do a "nice job" - I'm told so :)
For how long have you been building solo?
After 20 years, for me, I can tell you this - what helps me the most is exercising regularly (ideally on all weekdays) AND having kids.
Because of the kids, I rarely have the time to feel lonely and the regular physical exercises keep me in a (fairly) good mental state.
It felt the worst in my first years of solo dev, when all I did was work and hardly spent any time to socialize and take care of myself.
I'm sure it'll get better for you too as your project evolves and your life along with it.
This reminds me a of a very faithful (in browser) recreation of Windows XP I stumbled upon recently, may've even been on HN:
Good times.
I can tell you my reasoning.
While developing the dark mode theme for my mobile app, I experimented with pure black background first.
It looked off.
I couldn't figure out why it looked off (design wise, but also seemed weirdly hard to read) until I played with a bunch of other mobile apps with dark mode (which I used already and liked) and noticed their backgrounds are not black.
Then I just tried a couple of shades of "almost black" and the improvement was huge - easier on the eyes and more pleasurable to read.
I don't have a scientific explanation why this is the case, just I wanted to kinda directly answer the question with my anecdotal experience of why I do it.
Simply identifying them by IP so that each Google bot instance is a unique user ...
I rate limit because of aggressive bots (scrapers and ones trying random injection attacks etc).
And yes, it's returning 429 ...
Cool. Let me know if it turns out it's the same thing.
I can't find a problem with the 429 ...
I don't believe these numbers.
Here is a screenshot from my self hosted Matomo installation for my traffic today up until now:
https://i.imgur.com/WcfQKdT.png
Firefox blocks most analytics scripts by default, so these articles rely on very skewed data.
Any other devs with self hosted tracking solutions want to share how their browser stats look like?
9 and 12 boys.
They carry dumb phones for calls only (we gave them one when they started going on multi day trips from school or outside of it at age like 6/7).
They both have laptops and can game on and communicate with friends and classmates.
They have old iPhones they sometimes use at home (with no sim cards). Those don't leave the house and are very rarely used (mostly to chat in friends groups).
They don't watch youtube (except for music sometimes). They know what TikTok is (because all their friends watch it), but they don't (currently) have interest in it (mainly because we've had long talks about how terrible doom scrolling social sites and short videos is).
If at some point we give them smart phones with sim cards, we'll heavily restrict social media sites and apps, but I plan on keeping the dumb phones in use for as long as possible (hopefully until 15-16).
Not in the US, but yeah, I may be able to get a phone and an address through some 3rd party service(s), but I'm not sure it's worth the overhead, especially if it's just to pass the verification (and potentially users would not be able to really get in touch through them).
Yeah for now, I use "Advanced Access" to get the user's email address and pre-fill it so they only need to create a username.
Also these email addresses are automatically set verified (no need to go through the email verification process).
I guess I can make it work without the "Advanced Access", but it'll be a lot more cumbersome and I wonder if it's worth it.
Just a quick question on your traffic / conversion stats.
How come you have (almost) as many conversions as you have leads?
How do you count these?
I'm still experimenting different things with it so I can't tell you how exactly I'll end up using it, but yeah, ideally it should go beyond what you mentioned ...
I've made a screenshot of what a users sees once they start the trial:
https://i.imgur.com/Hc5taiX.png
The most important Premium features are listed there.
Here is a screenshot of the list of the prominent Premium features:
Please get in touch through the "contact us" link on the bottom of Next Episode so I can investigate what is happening.
There was no recent change affecting the movies list (other than some design tweaks which wouldn't limit the number of movies you see in your list).
Yeah, ads (for free users) and Premium membership which removes them and unlock features.
I run https://next-episode.net for about 18 years now.
It's a website + mobile apps that help users track what they're watching (TV shows and movies).
I wouldn't say trends over the years affected the business much, but I'm playing with the ChatGPT API currently to try and improve some aspects of the functionality and maybe add some new cool AI powered features - so there's that.
Thanks!
No cloud providers as far as I'm aware.
They were all from the same 4-5 ASN networks, all based in Russia.
Wasn't really worried about that.
I didn't do it as a super quick burst, but in a space of multiple hours.
First because the proxy servers were super slow and second - I couldn't automate it - their servers had some kind of bot detection which would catch me calling the URLs through script.
Instead, I installed a browser extension which would automatically reload a browser tab after specified timeout (I've set it to 10 sec or something) and I opened like 50 tabs of the honeypot URL and left it there to reload for hours ...
I tried but couldn't redirect back to my website as they stripped / rewrote all JS.
Yes, constructed the honeypot URL using the proxy site and called it (thousands of times) so I can get them to fetch it from my server through their IP so I can log it.
Same thing happened to me and my service (https://next-episode.net) almost 2 years ago.
I wrote a HN post about it as well: https://news.ycombinator.com/item?id=26105890, but to spare you all the irrelevant details and digging in the comments for updates - here is what worked for me - you can block all their IPs, even though they may have A LOT and can change them on each call:
1) I prepared a fake URL that no legitimate user will ever visit (like website_proxying_mine.com/search?search=proxy_mirroring_hacker_tag)
2) I loaded that URL like 30 thousand times
3) from my logs, I extracted all IPs that searched for "proxy_mirroring_hacker_tag" (which, from memory, was something like 4 or 5k unique IPs)
4) I blocked all of them
After doing the above, the offending domains were showing errors for 2-3 days and then they switched to something else and left me alone.
I still go back and check them every few months or so ...
P.S. My advice is to remove their URL from your post here. This will not help with search engines picking up their domain and ranking it with your content ...
I always thought it could look a lot better, modern.
I'd recommend next-episode.net as I use that for more than 15 years now, but then again - it's my service :)
It also has iOS and Android apps ...
Check it out and let me know !
Update 3: After about a day - they expanded their pool of IPs to more than 2k new ones.
I blocked what I could and then I blocked the whole country (Russia) behind a CloudFlare javasript challenge (so that any legit traffic that wanted to pass it - could).
Everything stayed blocked for about a day and now it seems they gave up:
- all domains that pulled dynamic content from my site now show some other content and do not try to scrape Next Episode
- all static domains that delivered the cached version of the Next Episode homepage now deliver some other website(s)
- CloudFlare shows no further traffic on the firewall rules I've set (so the bots from those IPs are, for now, gone)
As this post is relatively old now (more than 3 days) I doubt many people will go back and check for new developments, but I wanted to give this the proper closure and let you know of the apparent happy end.
Thanks for the suggestions.
I've added filing a DMCA removal request to Google to the list of things to do if this continues ...
The rest of what you mentioned we've discussed in prior comments and are indeed helpful in mitigating this.