[02:08:38] <wm-bb>	 <fuzheado> Forwarded from fuzheado: FYI, new tool from Magnus:
[02:08:38] <wm-bb>	 <fuzheado> 
[02:08:39] <wm-bb>	 <fuzheado> https://meta.wikimedia.org/wiki/ToolFlow
[02:08:41] <wm-bb>	 <fuzheado> http://magnusmanske.de/wordpress/?p=710
[02:08:42] <wm-bb>	 <fuzheado> 
[02:08:44] <wm-bb>	 <fuzheado> Providing a data workflow tool for the average Wikimedian is pretty slick. Time will tell whether it will be reliable and supported, but we've been needing this sort of facility for a while (see previous discussions about Apache Airflow, Dagster, or other data orchestration tools)
[04:00:37] <wm-bb>	 <robertsky> i wonder which will buckle first, toolforge itself or toolflow. 😏
[14:33:22] <wm-bb>	 <harej> If it’s useful for me personally I’ll host my own copy
[14:35:50] <wm-bb>	 <harej> This seems like the open source version of Tableau
[14:40:30] <wm-bb>	 <fuzheado> I'd say it's more like Apache Airflow or Dagster (both already FLOSS) in novice mode. The nice thing is it seems to handle the authentication issues elegantly. (re @harej: This seems like the open source version of Tableau)
[14:41:57] <wm-bb>	 <fuzheado> Basically, if Listeria is Gen1, this is Gen2
[14:44:15] <wm-bb>	 <harej> Last month I announced something of a product roadmap https://harej.co/posts/2023/08/three-new-concepts-for-organizing-work-on-wikipedia-workspaces-buckets-sprints/ and this new tool plays VERY nicely into it, I think.
[14:44:41] <wm-bb>	 <harej> Magnus and I are aligned on the data silo busting mission
[14:46:04] <wm-bb>	 <fuzheado> This part? "My other goal for Workspaces is to provide a standard way to invoke bots to run on wiki pages. "
[15:00:57] <wm-bb>	 <fuzheado> Your daily reminder that yes, Petscan is still down:
[15:00:57] <wm-bb>	 <fuzheado> https://github.com/magnusmanske/petscan_rs/issues/141
[15:00:59] <wm-bb>	 <fuzheado> https://phabricator.wikimedia.org/T347311
[15:05:59] <wm-bb>	 <harej> More the idea of building datasets of page titles compiled from different sources (category tree, wikidata query, what links here) and then exposing that data over an API for bot developers to work with (re @fuzheado: This part? "My other goal for Workspaces is to provide a standard way to invoke bots to run on wiki pages. ")
[15:06:42] <wm-bb>	 <fuzheado> It sounds like you're ready to adapt these as generator nodes for Magnus's tool :) (re @harej: More the idea of building datasets of page titles compiled from different sources (category tree, wikidata query, what links her...)