I was just handed this old but beautiful essay and I can't agree more:
"Cool URIs don't change"
I was just handed this old but beautiful essay and I can't agree more:
"Cool URIs don't change"
Falsch konfigurierte DNS-Einträge: URLs von Bose und Co mit Malware verseucht | heise online
https://heise.de/-10421430 #Cybercrime #Malware #DNS #URL #URLs
@mttaggart
You can also add alias/nickname for #bookmarks in Vivaldi .
You can even define a #keyboard (or #mouse) #shortcut or #menu item or toolbar #button to open, at once, one or several #url of your choice (via #CommandsChain feature)
Okay, Back of the napkin math:
- There are probably 100 million sites and 1.5 billion pages worth indexing in a #search engine
- It takes about 1TB to #index 30 million pages.
- We only care about text on a page.
I define a page as worth indexing if:
- It is not a FAANG site
- It has at least one referrer (no DD Web)
- It's active
So, this means we need 40TB of fast data to make a good index for the internet. That's not "runs locally" sized, but it is nonprofit sized.
My size assumptions are basically as follows:
- #URL
- #TFIDF information
- Text #Embeddings
- Snippet
We can store an index for 30kb. So, for 40TB we can store an full internet index. That's about $500 in storage.
Access time becomes a problem. TFIDF for the whole internet can easily fit in ram. Even with #quantized embeddings, you can only fit 2 million per GB in ram.
Assuming you had enough RAM it could be fast: TF-IDF to get 100 million candidated, #FAISS to sort those, load snippets dynamically, potentially modify rank by referers etc.
6 128 MG #Framework #desktops each with 5tb HDs (plus one raspberry pi to sort the final condidates from the six machines) is enough to replace #Google. That's about $15k.
In two to three years this will be doable on a single machine for around $3k.
By the end of the decade it should be able to be run as an app on a powerful desktop
Three years after that it can run on a #laptop.
Three years after that it can run on a #cellphone.
By #2040 it's a background process on your cellphone.
PHP may be the first language to correctly support RFC3986 and WHATWG URL at the same time if https://wiki.php.net/rfc/url_parsing_api passes.
Any URI user should read this RFC. It is a big step in the right direction. Good job by @kocsismate90 .
If the RFC is too long I created a polyfill for testing which works on PHP8.1+ https://github.com/bakame-php/aide-uri It will get tagged as stable once the RFC has passed.
Bookmarklets (and Custom URL Schemes) Are Criminally Underrated
https://silly.business/blog/bookmarklets-and-custom-url-schemes-are-criminally-underrated/
Today in BoneQuest History for January 30th 2014 "AUDITION FORM" https://bonequest.com/5481 #bonequest
#angry_panel #audition_form #comic_sans #deuce_pants #sfw #url
Today in BoneQuest History for January 25th 2019 "JOI" https://bonequest.com/7249 #bonequest
#deuce_spigot #drunk_deuce #jerking #joi #porn #single_panel #url
Recent work #productivity hack I’ve discovered. Phone calls.
I got into scheduling #Zoom meetings and it’s like… why? We don’t need to see each other.
#Phones. You can pace around. Not worry about background. No need to generate a #URL.
#Powerful and #disruptive.
#Help #Mastodon friends, we're having trouble with #Youtube #links not previewing properly, or previewing on some servers & not on others. Is there a preferred #URL style? Ex:
<youtu.be/IPSbNdBmWKE?si=Ur6DAZrQvApWIDwt> https://youtu.be/IPSbNdBmWKE?si=Ur6DAZrQvApWIDwt
OR <www.youtube.com/watch?v=IPSbNdBmWKE> https://www.youtube.com/watch?v=IPSbNdBmWKE
Hit to defeat a #paywalledlink
Copy the #URL & paste it in the window of the website below, then select which app you want to try to give you access to the paywalled url.
Hello blog readers and Fedi Friends!
Two months ago I managed to successfully federate this site, giving it #ActivityPub superpowers (if you are confused about what this means, I wrote about it for on my Ghost blog The Future is Federated).
This feat was the absolute highlight of my summer and I’d love to make it a habit to write more on here – inspired by Andy Piper‘s #100DaysToOffload (I couldn’t possibly write daily, but maybe weekly).
In case you missed it, on Friday I published this essay for The Future is Federated: “Beyond technical features: why we need to talk about the values of the Fediverse (part 1)” (source). It’s my love letter to the Fediverse, listing all the reasons why I truly cherish it.
What does this have to do with “carpe diem” (seizing the day)?
Well, earlier this month I learned an important lesson: if I ever have a project idea – especially about a domain name to register – I should not procrastinate but immediately take action.
In late August I started looking up available URLs for potential future Fediverse projects / self-hosted one-person instances. I thought: for my birthday, I will gift myself two domain names – elena.video and elena.social. They were both available and seemed perfect for the purposes of hosting my own Peertube server… and/or something else.
Well well well my birthday came and went and on September 18th I checked and saw that elena.video had been registered by someone in Russia on September 12th 2024!!!
At least I bought/registered elena.social.
Maybe one day in the not so distant future I will learn how to self-host a Fediverse project and use that as its URL / username. For now, elena.social is redirecting to the “Links” page of this site, listing my social media profiles. I just updated that page to reflect my views and some of the powerful points I shared in my Ghost blog post: “My favorite social networks are all part of the real Fediverse (Mastodon, Friendica, Pixelfed); I rarely use platforms owned by Meta & co. for reasons I outlined in this blog post (TL;DR: behavior modification and illusion of connection for purposes of surveillance capitalism).”
I’m excited but still, I can’t begin to tell you how disappointed I am (to this day!) for not registering elena.video when I saw it was available.
Well, you live and you learn. Carpe diem folks!
Cover photo by Anton Filatov on Unsplash
PS. I forgot to mention: use the back, forward, and refresh buttons on the browser. It should all work as expected. Because this is a *web* app, after all.
Of course it’s not just quantitative differences but qualitative ones too…
Since all our time isn’t spent trying to create planet-scale factory farms for human beings (which are hard to do and involve lots of complexity), we can concentrate on nice things like maintaining state in the URL… like Kitten Kawaii does.
URL-based state: https://ar.al/2024/08/19/kitten-kawaii-porting-a-react-library-and-next-js-web-site-to-kitten/#url-based-state
Full blog post: https://ar.al/2024/08/19/kitten-kawaii-porting-a-react-library-and-next-js-web-site-to-kitten/
View source: https://codeberg.org/aral/kitten-kawaii
Today in BoneQuest History for June 29th 2002 "HELP ME EVERYONE" https://bonequest.com/1368 #bonequest #atandt_bung_spigot #bathroom #bullying #bus #gay #judgement #sarcasm #url
Timothy McCurrach is presenting the talk "Navigating the maze of Django's URL routing: a deep dive" at DjangoCon Europe 2024
#DjangoConEurope2024 #DjangoConEurope #DjangoCon #Django #URL
The power of #URL parameters lets you unofficially turn off Google's #AI Overview.
#GoogleSearch’s “udm=14” trick lets you kill #AISearch for good
https://arstechnica.com/gadgets/2024/05/google-searchs-udm14-trick-lets-you-kill-ai-search-for-good/
#Mastodon:
Boosts, Shares, Reblogs, Retweets or Retoots?
The official version (with - 1 x “ - too much ):
““Boost” (as in “signal boost” or “rocket boost”) is a synonym of “reblog” or “retweet” in Mastodon.“
https://blog.joinmastodon.org/2018/08/mastodon-quick-start-guide/
PS - The interesting story why the #toot was retired:
https://mastodon.social/@Gargron/108352418240054955
Google Safe-Browsing: Echtzeitschutz durch serverseitigen URL-Check | heise online
https://heise.de/-9655635 #Webbrowser #Browser #Chrome #URL