flas9sd

love that the author still uses the same dithered photos of Jajce, Bosnia - his hometown - as in Win9x days in the about page and website: https://www.irfanview.com/main_about.htm .. make me want to visit.

Edit: typo (thanks)

If you're currently into researching sqlite questions and it happens you prefer one sites explanations, it likely comes up in the first few pages of a result of new questions on sqlite. By looking into bookmarks and history the users preference is clear.

In general you're right as in: most often in the first few result pages, there is no hit of sites you prefer. I think an extension that lists the most popular pages that have user-agency to the top can already help without curated blocklists: rank wikipedia, stackoverflow sites, github and reddit to the top.

There was a Show HN for "hypersearch" in 2022 that did something similar but in a sidebar - https://github.com/abhinavsharma/hypersearch

I'd like to use the opposite approach: top-list search hit results on websites that I've bookmarked another subpath of or visited before. Difficulty being: when to stop requesting later result pages to get a hit. Both Firefox moz_places and Chromes Default/History sqlite databases have the info to populate such a list.

> It's about not using unaudited closed source code which cannot be proven to be secure or private [..]

Degoogling is not deblobbing and Lineage or /e/ use plenty of closed source software during runtime. The top parent and DivestOS author really is deblobbing* to some degree, but forks of LineageOS that introduce measures of "degoogling" hold onto vendor firmware blobs on androids /vendor partition for functionality. Those aren't known for connecting to the Google hivemind though.

* https://github.com/Divested-Mobile/DivestOS-Build/blob/de3ba...

My interpretation of the term degoogling fits the second part of your sentence, "getting away from online services": it is user agency in what network connections can occur, so either by default or optionally users can stop any signaling coming from the device they use. They don't have that freedom with the software the device came with.

do you still actively see this or is it a behaviour that occured a few months ago? I remember https://bugzilla.mozilla.org/show_bug.cgi?id=1840665

it's exactly the kind of device I'd base an argument on. The machinations of the mobile device and telecom industry are adversary to a majority of consumers. I don't want to buy a phone every 5 years to continue to make at best 32-64kbit/s bitrate calls to mom and pop.

The global menu top-bar and the HUD application search was what I missed most when I went with the flow and settled in GNOME >36ish.

Here some more screenshots for the uninitiated (with search used on GIMP in the first two screenshots): https://imgur.com/a/5XlgTO3

And the official docs: https://wiki.ubuntu.com/Unity/HUD

great idea to reduce scope and skip "the agent" part and let the users do it, can be added later. When you do, you probably want to introduce some means to query for and update IPs via api first, to have all-dynamic nodes covered. Currently users need at least 1 peer that is either static or has outside dyndns setup.

I do that for dedicated netflix profile (with company icon) having DRM enabled, very useful.

If Firefox did extend the Profile Manager to allow for this kind of personalisation and OS integration (profiles showing up as their own, customizeable launcher icon), that'd be another win.

thanks, today I learned.

Firefox has useful about:about pages.

With some revamp they could capitalize on some of them much more. After newtab, the last addition was "Firefox View" for recently closed tabs or synced tabs. I'd wish the same for history, bookmarks and downloads. Those very much show their origin as sqlite browser.

my impression of the mozilla firefox bugtracker too - you might be taken aback by the lifetime of some long standing bugs, but what is confirmed, stays open.

You'll collect low effort comments, but at least it avoids duplicates and the discussion is preserved in-thread.

That impression lead me to meaningfully contribute in the bugtracker.

VoWifi is using an ipsec tunnel for both SIP/RTP to the providers IMS stack. With its popularity I think it's likely network admins increasingly allow for it in networks you do not control.

because SoCs from 2018 (think SDM632) are easily able to run Android 13 and the production/marketing/buyer capacity so potent, the market and drawers at home are saturated. 4G will stay on the towers for another few years, so 30€ is the real low on the smartphone price spectrum - the device type is now ubiquitous.

the natural language to SQL space is interesting, seeing tools pop up. The on-premise property is key. It is fragmented and thus difficult to write integrations for, but text-input fields of SQL clients / notebooks / BI tools (think metabase) are looking to be NL-to-SQL equipped. At least it will be a race of vendors to provide this.

something I came to use later in my cli life - using `df .` as in most cases the working directory I'm in is the filesystem I want to get the info on

- hiding unused styles in Paragraph Style dropdown - https://bugs.documentfoundation.org/show_bug.cgi?id=152666

- highlighting for used Paragraph and Character styles - https://wiki.documentfoundation.org/File:Styles_highlighter_...

I'm using these alot, very welcome

having recently looked at a firefox performance regression for filing a bugreport, tooling that tracks performance (quite publicly) sees attention, easy upload to share tracing profiles also helps: https://firefox-source-docs.mozilla.org/testing/perfdocs/ind...

Their dedicated blog keeps you posted if firefox perf is your interest https://blog.mozilla.org/performance/

If you have a particular website you notice chromium being significantly faster with, for an easy report, there's https://webcompat.com/ - though bugzilla is better than it seems when coming from github issues

great for doing measurements in a cron to later plot them. Here someone measured TLS version latencies on gov.uk - https://nooshu.com/blog/2020/07/30/measuring-tls-13-ipv4-ipv... and has a good graph explaining the time_* variables

is this due to "bestmatch" suggestions being shown first? I turned off alot (and made the results rather long) to get more out of it. These are modified values from the defaults:

    browser.urlbar.maxRichResults   22 
    browser.urlbar.quicksuggest.migrationVersion     2 
    browser.urlbar.quicksuggest.scenario  history 
    browser.urlbar.showSearchSuggestionsFirst         false 
    browser.urlbar.suggest.bestmatch              false 
    browser.urlbar.suggest.remotetab              false 
    browser.urlbar.suggest.topsites          false 
    browser.urlbar.suggest.weather              false 
    browser.urlbar.tabToSearch.onboard.interactionsLeft 0 
    browser.urlbar.tipShownCount.searchTip_onboard 4
    browser.urlbar.speculativeConnect.enabled         false 
    browser.urlbar.shortcuts.quickactions  true 

the code and sql that determines the behaviour for this is at

https://searchfox.org/mozilla-central/source/browser/compone... https://searchfox.org/mozilla-central/source/browser/compone...

I get a lot out of this

true, but that'd be not part of the 1.3 release notes.

You still need scribus in the loop to get to cmyk, it's not a seamless workflow yet.

if you use Inkscape in print workflows, two thing stand out: rewritten pdf import and page margins and bleed.

To clean up path tracings from surplus edges, a changed "node deletion logic" is also welcome.

sounds lowkey, but making the dish with plenty of fresh vegetables as soup moves the needle (noodle, heh) back to the healthy end of the gradient?

I'm sorry for the packaging waste though.

I use the shortcuts and address bar extensively - they made me a quick bookmarker and more often I gauge what I have indexed on a topic prior to do a websearch. With "+" you can have more than one tag to narrow down a list if it has multiple tags. While just using history+fuzzysearch works for a lot of people, I'm browsing too much to have a small list when using the history on often used terms - thus bookmarks and tagging.

just gave it a test, the design additions are introduced gently. Defaults for listings are on the familiar view. Menu quick actions exist to go from compact to normal to more padding.

I had the 2-line message header view on my wishlist for a long time. Thunderbird did pretty great.

a statistic I'd be interested in: what percentage of searches can be answered computationally cheap. As in: Wikipedia title index, simple word lookup dictionaries. Indices that could complement a caching search-engine proxy to not hit its origin crawl repository.

A study[1] by wikipedia done with DDG notes it showed up in the top5 results and information module for ~13% of searches with a click-through rate for each at ~8% - so a total of ~16% click-through rate. Granted, that is not a number gained from title searches but the whole articles.

[1]: https://diff.wikimedia.org/2021/09/23/searching-for-wikipedi...

platforms create a two sided market, inflating their own value as middlemen and blackbox (if not publicly listed). What mastodon showed is: how high is the bill really? as in: given the opportunity, maybe users will foot the bill and having an incentive to optimize for cost.

My only worry is a tragedy of the commons scenario.

I've been looking at and trying the class of easy to host, api-fied db/table frontend tools for a while. Clicking around your demo the sought-after functionality is there and the roadmap looks promising, good job!

Some ui-breakpoints for a mobile view on tables are missing and not yet catalogued in the issue tracker. Use-case: string searching a table on the go.

what I look for in job runners is: can I sync any job definitions done in the UI to a git repository back and forth - and how is it handling secrets. Openjob.io looks like an early iteration.

Android APK security model is pretty apt for using mirrors - what you need is a trusted directory of known-good mapping of appid to signing developer key fingerprint to cover for trust on first use when installing - plus handling key revocation.

Mirrors publish the fingerprints, but it's unclear what verification they applied. Looking at history won't cover revocation, but it's already something. Certificate transparency logs could handle apks instead of domains too.

I understand the appeal of Aurora and similar - you let Google handle search, malware takedown and download traffic, but then you're also up to be cut off at no notice.

not much effort in listening to quality (web) radio stations. A playlist builts itself from shazaming or checking the playlist archive.