The 1 year offline anniversary(-ish)

It’s been a while since you guys have had a proper update, hasn’t it?

In this post I want to talk about what you guys can expect, what not to expect and when to expect it. After that, also some rambling.

Important note: payments will be back from the start of January onwards

The comeback

So far I’ve been fairly vague about what is actually going on. I can however now finally say that we are back on track and regular progress is being made. The basis is there, the moderators are testing the new site and it’s already getting expanded upon. The reason it has taken so long I’ll get into after the specifics, but you don’t necessarily need to worry about that.

What to expect

Obviously video uploading, post controls (public/unlisted, only logged in users), filters, profiles, search and discoverability. These are all the things that I want to launch the new InflateVids with. It’s basically the minimal product that is what made the website so great for the community, in my opinion.

Most of these will be fairly familiar, there’s not much to do different there, but a major difference will be the filters. Posts will now only contain tags, categories are out the window. These tags will be pre-defined by us, the admins/moderators, to keep consistency and accuracy. 

Based on the tags on a post you’ll be able to then customize your filter. Don’t want any inflatable popping? You can enable that, but maybe you don’t like balloons being poke popped? I want you to be able to be more specific about your filters. The implementation isn’t final, so I can’t say much about it yet.

Public videos from the old website should mostly be recovered (thanks to a community member). If you had videos on the old site, you’ll be able to either transfer them to an account on the new site or download them in a zip archive once and they’ll be removed from our storage.

The video encoding, handling and hosting will now be fully done by a third party. This way I don’t have to touch any files ever, so it should hopefully be a bit more stable than all the encoding issues of the last website. And also many people uploading won’t bring the site down maybe, hopefully.

Also, yes, I’m using better hashing this time, Argon2. Yes I know the difference between encryption and hashing. Yes, I still need to implement salting before release.

What not to expect

You might notice a lot is missing from the previous site. I didn’t mention comments, likes/dislikes, direct messaging, notifications, playlists or even subscribing. A lot of these features I do want to implement again, but many are not strictly necessary for getting the core of InflateVids back online.

I don’t want to have the community wait any longer because I need to design some system to get comments going, especially when people seem to prefer to just share links and discuss elsewhere. All of these will probably come back in due time, over the following year. Except for direct messaging. Me and my homies hate unsecure DMs. Just use already existing apps, seriously.

When to expect

Currently, for all the lovely patrons, the deadline I’ve set myself is the 19th of February, 2025. One thing is for sure, the initial website will be for patrons only. This is to test the site, make sure it works, add stuff people still really miss, and to be honest get some credit going for the video hosting part. The site will be public once it’s nice and stable, and we as moderators are sure we can handle the wider audience.

The public release of InflateVids will follow between March and June of 2025.

Why it took so long

Oh boy, okay, this is where the rambling begins. If you just wanted to know when you can watch your favorite squeaky videos again, you got your answer. If you’re in for the long haul, strap in.

What timing!

It’s no real kept secret the hack happened during very unfortunate timing. My birthday is mid-december (Holiday bday peeps unite!), we booked a holiday around the time and work was… very stressful, to say the least. The website got defaced very early in the morning, around 3am, so it was only many hours later sleepy me looked at his phone and saw plenty of messages coming passed.

At this point, damage control. Take it all offline, disconnect any servers from the internet, see what was still recoverable. During this time I was definitely panicking, a lot. Besides the main issue of, you know, the site being offline and all the data being gone, there was a lot happening around it too. Many people took it upon themselves to start pointing at others (unnecessarily), some started deliberately causing more havoc, and I got caught in a lot of that crosswind.

The holiday

Some days after the hack, the already planned holiday started. In the days prior I did what I could so far and decided this was a time I’d just shut off. Be excommunicado for a bit, recover. Obviously I was still available for the moderators if anything really needed my attention, but at this point there was no take-backsies and what happened, had happened. This honestly helped a lot in calming me down, it was a strange time I’ll probably never forget. I felt very detached from everything. For some time, I was just a regular guy without all the responsibility (and with a lot of balloons and inflatables in his cottage). 

But sadly that didn’t last, and I needed to get done fixing the site. I put up a simple FAQ page, emailed all the people that had accounts, and started to think how to bring the site back up. I can’t just bring the old site back, and besides, I was already working on a completely new site. How hard could it be? Famous last words.

Version 8

That is, I think, the current iteration of the website. Turns out that having a website hacked that’s hosted/managed by a single person really does damage your self confidence. I kept ping ponging between different frameworks, languages, principles, and techniques. This version is final though, we’ll rewrite if needed in the future. The website will be fully written in Typescript and built with the SvelteKit framework. Authentication is all happening and being handled by us, since Clerk and Auth0 become real expensive real quick (and lack some flexibility). Have looked into hosting Keycloak, but that’s a bit overkill. Perhaps in the future.

Anyway, it took me a while and a lot of impromptu Telegram therapy to come to the conclusion that it’s never going to be perfect, there are going to be a lot of errors and mistakes, but I just need to deal with the fact that’s going to happen no matter what. Trust is to be earned, and a lot of people still seem to have that trust with me. Many in the community might not have that trust anymore, and I understand that. It’ll take some time to rebuild the trust and to get the website full steam ahead again.

Currently I am making a lot of progress on the website. Ask any of the moderators (they’re still the same ones) and they’ll tell you the same. Thanks to everyone who offered help, I need to get better at accepting it. Just please don’t get me to change frameworks or languages again, I just need to get shipping ;P.

See you soon!

- Kyan