Text-based queries fail silently. An example from today where the user was lucky enough to notice some unexpected results but couldn't understand why they were returned. Situations like this makes it feel like we're building on sand... discuss.elastic.co/t/why-query-ds…

Looking forward to presenting at this next week

Hmmm. I assumed most queries that work in elasticsearch could be converted to mongodb. That may have been optimistic mongodb.com/docs/manual/re…

Look back on the news from 2022 with this visualisation tool I've been developing. #2022rewind news.inperspective.com

Recruiters are heavy users of Boolean search syntax. Here's a demo with LinkedIn data that shows how graphical interfaces can help youtu.be/56WSW_sORxo

New demo: taking patent search to the next level with graphical queries youtu.be/D5Tlhlq1mmM

Warning: your Lucene-based queries can easily be doing something you would not expect. youtu.be/y9xwzsKrTiA

Introducing qry.codes - draggable, merge-able search topics that can be combined with AndOr’s graphical Boolean search. Based on binary vectors.

Will be giving this talk on Thursday with a demo of new features.

Looking forward to talking about “fuzzy facets” using binary vector clustering at this week’s elasticon London. elastic.co/events/elastic…

My elasticon talk on qry.codes is up - Binary vectors and in-browser clustering

AndOr’s graphical query language now targets multiple databases: * Elastic/open search * Vespa * Redis * JavaScript db Making progress on the query compatibility matrix:

Presenting at #GenAISummit tomorrow - using AndorSearch to refine and cluster text, vector & geo queries, graphically. AndOr is built to support choice of back ends and this demo will use vespa.ai

Tag, drag, discover: How LLMs can create useful, instant filters from raw text youtu.be/mQPKLciQIHU

My talk on “emergent facets” at this week’s Search and AI Meetup youtu.be/U1RGTMesCq0

Honoured to join the lineup of search experts interviewed in this series. We talk myths, “OR is mORe,” and how vectors should be treated like plasticine, not porcelain. Thanks Itamar and Zevi! pulse.support/top-voices/mar…

Spotting significant terms was something I added to elasticsearch. Can we do it more cheaply in the browser with similar quality? It looks like the answer is "yes" youtu.be/Te_N8qveCKY