@parismarx hey, I listened to the System Crash episode from May 23rd yesterday (a little bit behind on my podcasts...), where at some point you briefly talk about search engines you have been trying out, including ecosia and qwant.

Not sure, what else you have tried, so I thought maybe I just throw in two that I have experiemented with and had great success. Perhaps you already know them. But those were not mentioned in the podcast. Until the european Open Search Index is ready, there unfortunately is no perfect solution, but meta-search engines have been the best compromise for me personally until then.

Obviously you still rely on big tech to some extent, but at least you diversify the reliance a little bit. Sometimes, this also takes the form of competitive compatibility, which is a notion I really like towards big tech.

#SearX/#SearXng is of course great and extremely customizable, open source and self hostable (for trying out, here is a German provider: gruble.de).

Personally, I have been using #MetaGer (metager.de), a German meta search engine, for quite a while now. I like the minimal and brutalist-esque UI of it. For me,  it just gets the job done for me, both in my personal life and in my academic life. A little disclaimer here: MetaGer is run by a German NGO, which I joined as a passive member a few months back, because I really like their mission. They also plan to create a new search engine, entirely based on the Open Search Index, called "nolm" (no LLM), which will also find its way as a search provider into MetaGer.

I do have some little gripes with MetaGer, which I remedied with my own flimsy solution stacked on top of it, that  provides  a similar feature like DDG's "Bangs".

Anyways, thanks for the great podcast and this episode in particular, which was very entertaining.

@kubikpixel besides using europe-hosted and ideally decentralized services, I also like to use specific search engines for specific tasks. Inspired by DDG's bangs, I created serçi to use bangs/keywords with my favorite search engine #MetaGer. you can #selfhost serçi on a tiny webhost or VPS. it's small, fast and minimal in terms of data usage! :)

@maexchen1 @pallenberg die kostenlose Suche von #MetaGer wurde nicht wegen usermangel geschlossen, sondern weil yahoo EU-Verträge aufkündigte.

Die Token-basierte Version von Metager funktioniert aber noch, wird beständig weiterentwickelt und hat erst kürzlich neue Suchmaschinen hinzugewonnen.

Looks like #MetaGer also implemented #Google queries via https://serper.dev/ now. It costs only 1 token per request, which is the same as #Brave and #Mojeek. Let's see how the search results turn out with this.

Two new changes in my Meta² Search Engine #serci:

1. Due to #MetaGer being forced to disable their advertisement financed gratis offering, the default search engine is now france-based #Qwant
2. Users are now able to choose an individual default search engine via a drop-down from all configured search engines. This will set a cookie in your browser.

So, if you (like me) want to continue using @metager@suma-ev.social via the paid tokens (or support them via a membership, which I would urge you to do), you can set it as your personal default search engine in serĉi again :)

Change-Logs for these two serĉi changes:
1. https://src.jayvii.de/pub/serci/commit/8634b8ee623f1b7abfc1815ea8aab53a643683af.html
2. https://src.jayvii.de/pub/serci/commit/9b9da13251d2e1fec60631846fc3d22c29c08df6.html

After a few performance improvements, I did some loading time tests for #serci, between my home network (wifi) and a random VPN where the service is running.

echo "html,redirect" | tee timestamps.csv
for i in $(seq 1 1 100); do
    HTML=$(curl -o /dev/null -s -w '%{time_total}' https://search.jayvii.de)
    REDI=$(curl -o /dev/null -s -w '%{time_total}' https://search.jayvii.de?q=test)
    echo "$HTML,$REDI" | tee --append timestamps.csv
done

Across 100 runs for loading the site's HTML (generated from pure #PHP) versus a redirect to a chosen service (here the default #MetaGer), I can measure on average 0.28s for loading the frontend (HTML) and 0.13s for processing input and issuing the redirect.

I am quite happy with this relatively low overhead, although performance may decrease a little if more services are added (currently: 47). At some point maybe an #sqlite database may be more efficient than my pre-constructed #json files which are loaded on-demand.

Lately, I've been using @MetaGer #MetaGer as my main #SearchEngine, but had been really missing bangs (known from #DuckDuckGo and other search engines).

So this morning I threw together a short #PHP site that I could use as search front, which handles bangs and redirects me to the correct search engine (is this meta-meta? ;D):
https://search.jayvii.de/

If you find bugs or want to contribute, feel free to check out my git repo and contribute:
https://src.jayvii.de/pub/traserci

Both #DuckDuckDuckGo as well as #Qwant seem to be full in on #AI powered summaries of search results. With every year, online search seems to slip more into #enshittification with #SEO madness and #AI generated content and searches.

Time to pick up other search engines like #MetaGer, #mojeek or #SearX