joaomsa

Essential reading. More in-depth than an introduction, but without being overly impenetrable except to those dealing with the internals.

Thanks for tip, I've always wondered what was the root cause of those occasional stutters, specially since I often speed up videos

My pet peeve is with the still half finished look, it jumps out as really garish now to when you open the dev tools and see the clashing styles or the bookmarks/history dialog that hasn't been updated in ages.

Not to mention the white title bar in some dialogs even with dark mode enabled.

pledged

Might affect (dumb) uses that rely on the current increasing version number scheme in user agent strings and package/deployment scripts.

The Proposal for encoding the Phaistos Disc characters[1] lists the motivation behind including it (and rebutes some counter arguments). Principally, lots of publications discussing the Phaistos Disc would use it:

> Phaistos Disc characters, whether syllables, or letters, or board-game dingbats, have historical and cultural significance, as attested in the large number of publications dealing with it. As noted above, the Wikipedia has articles about the Phaistos Disc in ten languages. The 30+ documents listed in the bibliography in this proposal are by no means the only documents printed which deal with the Phaistos Disc characters. The English Wikipedia article gives Phaistos Disc characters inline in text as well as in tables (Figure 8). Other documents exist which present the Phaistos Disc characters either inline (Figure 8) or in tables (Figures 1, 2, 5, 6, 7). It is true that the image of the Phaistos Disc itself is easy to represent with simple drawing (see Figure 3) but the discussion of its characters inline in Latin text is not.

[1] https://www.unicode.org/L2/L2006/06095-n3066-phaistos.pdf

This looks really nice, do you have plans to implement text search of note contents? No matter how methodical I get with tags and hierarchies, nothing beats search for recall.

How to people usually get around file downloads and initial page loads with JWTs (besides signed urls), since you can't control and preempt these requests to also send the JWT?

Nice companion to: http://fetchrss.com/

My guess would be yes since there's this for Chrome: https://github.com/davidhampgonsalves/foxish

Since Firefox's WebExtension API is a close superset to Chrome's API, I would wager this could be easily ported now.

I did mention using bookmark keyword searches as hack to get sync, but I'm talking about proper support for syncing user registered OpenSearch[1] providers (and any user configured keyword shortcuts for them)

As of now you can add an OpenSearch[1] provider in Firefox using the javascript api `window.external.AddSearchProvider(engineURL)` where engineURL is the XML file description of the search engine like this one for IMDb search:

https://m.media-amazon.com/images/G/01/imdb/images/imdbsearc...

This is the method the Mycroft[2] project uses, which Firefox recommends. Chrome though, besides also supporting that API, looks for a search meta tag in a page's head after you visit it liking to the same OpenSearch provider xml

<link rel="search" type="application/opensearchdescription+xml" href="https://m.media-amazon.com/images/G/01/imdb/images/imdbsearc... title="IMDb">

In Chrome, after you visit their home page with that metadata it's registered as a provider so once you start typing Imdb in the omnibox, you see an indication to press 'Tab' in the right corner of the omnibox, and it switches to searching imdb instead. Even providing autocomplete if the search provider supports it.

Firefox could make registering these more discoverable from the UI (adding via a page action as an example at least?)

[1] http://www.opensearch.org/Home [2] http://mycroftproject.com/

While I've never experienced conflict issues with bookmark sync, one pet peeve of mine is how their sync implementation has never supported syncing search engines.

I mean proper OpenSearch support, different from using keyboard bookmarks with %s as a hack, since you actually get proper autocomplete.

In fact that whole situation there is just a mess, no proper way to use sites that implemented OpenSearch from the awesomebar like the Tab to search in chrome's omnibar.

My guess it's a convenience compromise. Nutrition labels list macro nutrient amounts in metric, but packaging still has the product amount more prominently in imperial measurements.

I wonder if Canonical will also release 17.10 as a Windows Store WSL app or stick to LTS releases and wait until 18.04 for an update.

Peter Adamson is pretty thorough, living up to the HOPWAG name. However, he's still consistently putting out episodes every Sunday, give it a few years and it'll get there for sure.