Seven years ago everyone was up in arms about DoH and I wrote about how we implemented it in #Firefox (as I worked for Mozilla back then)
https://daniel.haxx.se/blog/2018/06/03/inside-firefoxs-doh-engine/
@bagder@mastodon.social I remember hearing about this because the UK ISP lobbying group nominated Mozilla as an Internet Villain when it rolled out: https://www.ispa.org.uk/ispa-announces-finalists-for-2019-internet-heroes-and-villains-trump-and-mozilla-lead-the-way-as-villain-nominees/
After I wrote (most of) the DoH implementation for #Firefox it was not too hard to subsequently add support for DoH in #curl.
If I’m going to be totally honest: implementing anything in #curl is about fourteen times easier and more fun than the thread- and object-spaghetti that is #Firefox code… But don’t tell anyone I said this.
@bagder@mastodon.social It’ll be even easier once you merge that LLM authored PR. The one refactoring everything to use a more modern and flexible component system.
At least three kinds of marshalling.
Cross platform of course.
Interfaces upon interfaces for that extra plattan-i-mattan rapid development cycle?eXtra Powerful Curl Object Model?
@bagder@mastodon.social … so we can soon expect new curl parameters “–run-with-gui” and “–enable-web-browser-mode“. Not much is missing for that, right?
@bagder@mastodon.social The thing that projects like Firefox (and Chromium) desperately need is a fresh rewrite from scratch…
@bagder@mastodon.social This will stay between you, me and your 34 thousand other followers
@sampe@piipitin.fi thanks!
@bagder@mastodon.social how do you feel about me browsing using lynx?
