New Kitten Release

• Implements cascading archives support

https://kitten.small-web.org/reference/#cascading-archives

Cascading archives¹ is useful if you have a static archive of the old version of your site and you don’t want to host it somewhere else and use the 404→307 (evergreen web) technique (https://kitten.small-web.org/reference/#evergreen-web-404-307) (the latter is useful if the old version of your site is a dynamic site and you cannot take a static archive of it).

If a URL cannot be found on your app, Kitten will trying it in the archive folders:

__archive__1
__archive__2
__archive__3

(In that order.)

So you can three older static versions of your site served without breaking older URLs unless they are shadowed by newer URLs in your site/app.

Enjoy!

¹ This is a feature that I originally implemented in Site.js (that’s going to be shut down tomorrow when Let’s Encrypt stops issuing certificates with OCSP-stapling – I don’t have the bandwidth to maintain two servers/frameworks; Kitten is Site.js’s successor). I’m planning on implementing this differently in Kitten going forward (so you can use the Settings interface to upload a zipped archive and it will serve it) but I need this for my own site for tomorrow’s shutdown so we have this simpler implementation in the meanwhile. Leaving things to the last minute? Who? Me? Never! ;)

Every once in a while, I read my old blog posts that link to other articles or blogs. 99% of those sites are now offline or show a domain-for-sale page. It kind of makes me sad. Most of my own websites are still online after 20+ years. I wonder why they gave up.

That being said, this instance will be online forever - I'll make sure of it. Even after my passing, I want it to live on. My websites have survived even when hosting companies went bankrupt, and so will this instance.

@timbray on #URI schemes of the #Fediverse (and #BlueSky).

If he writes good pieces like that, we rather should listen and probably implement his suggestions as they are great!

https://www.tbray.org/ongoing/When/202x/2025/04/16/Decentralized-Schemes

For the first issue (not being able to directly interact with a toot although web-logged-in), you can use browser add-ons like https://addons.mozilla.org/en-US/firefox/addon/mastodon4-redirect/ that redirect fediverse-links to your instance.

Google updated its URL structure documentation, favicon documentation and added a new Google Trends document https://www.seroundtable.com/google-url-structure-favicon-google-trends-docs-38307.html

Check out this #JavaScript #function for parsing #URLs into their components! Easily extract #protocol, #host, #path, #query #parameters, and more from any URL string. Perfect for handling URL #manipulation in web applications. #JavaScript #WebDev #URLParser

https://gist.github.com/AnandPilania/c4347005b220b348222b441287c05199

summarized here: https://www.eke.li/2023/02/domains-in-docs/

when writing #technicalDocumentation - don't use real #urls or #ipAddresses. you never know what will grab those and use them.

there are reserved #domains and #ipRanges for just this!

for #url: the top-level-domains
.test, .example, .invalid and .localhost and the domain example.com are reserved

for ip-adresses the blocks 192.0.2.0/24, 198.51.100.0/24 and 203.0.113.0/24 are reserved

they are always safe to use in #documentation

Bon ben j’ai décidé de ne plus partager les pouets qui utilisent des raccourcisseurs d’URLs.

Sur #mastodon une URL à un coût de 23 caractères quelque soit sa taille :

https://docs.joinmastodon.org/api/guidelines/#links

Il n’y a donc aucune raison de cacher des #URLs.

Bon, si le contenu est vraiment trop d’la balle et que ça utilise
https://frama.link/ alors j’y penserais peut-être à deux fois.

Merci, bisous, merci.