Forest Gregg
Forest Gregg
if you don't set an explicit limit on container scaling, then [google defaults to 100](https://cloud.google.com/run/docs/configuring/max-instances#limits) google recently updated the [limits on container scaling](https://cloud.google.com/run/docs/configuring/max-instances#limits), such that if you set up datasette...
Right now, if click on the detail page for a row you get the info for the row and links to related tables:  It...
*Original title: Set content type for CSV so that browsers will attempt to download instead opening in the browser* Right now, if the user clicks on the CSV related to...
query result page is using 400mb of browser memory 40x size of html page and 400x size of csv data
[this page](https://labordata.bunkum.us/opdr-8335ea3?sql=with+most_recent_lu+as+%28%0D%0A++select%0D%0A++++*%0D%0A++from%0D%0A++++%28%0D%0A++++++select%0D%0A++++++++*%0D%0A++++++from%0D%0A++++++++lm_data%0D%0A++++++order+by%0D%0A++++++++f_num%2C%0D%0A++++++++receive_date+desc%0D%0A++++%29+t%0D%0A++group+by%0D%0A++++f_num%0D%0A%29%0D%0Aselect%0D%0A++aff_abbr+%7C%7C+coalesce%28%27+local+%27+%7C%7C+desig_num%2C+%27+%27+%7C%7C+unit_name%29+as+abbr_local_name%2C%0D%0A++coalesce%28%0D%0A++++regexp_match%28%27%28.*%3F%29%28%2C%3F+AFL-CIO%24%29%27%2C+union_name%29%2C%0D%0A++++regexp_match%28%27%28.*%3F%29%28+IND%24%29%27%2C+union_name%29%2C%0D%0A++++union_name%0D%0A++%29+%7C%7C+coalesce%28%27+local+%27+%7C%7C+desig_num%2C+%27+%27+%7C%7C+unit_name%29+as+full_local_name%2C%0D%0A++*%0D%0Afrom%0D%0A++most_recent_lu%0D%0Awhere+%28desig_num+IS+NOT+NULL+OR+unit_name+IS+NOT+NULL%29+AND+desig_name+%21%3D+%27HQ%27%0D%0Alimit%0D%0A++5000+offset+0) is using about 400 mb in firefox 97 on mac os x. if you download the html for the page, it's about 11mb and if you get the...
the project changed to an incompatible gpl license and it would be nice to have control over binary wheel releases. not actively developed so we would not be losing out...
The major barrier to efficient parallelization of blocking is the inter process communication of the records. Similarly, much of the potential benefit of parallelization of scoring is lost because of...
RIight now we create as many missing data indicators as there are variables, we should create as many data inidicators as there are unique fields.
the Exists variable should take the place of any simple MissingDataType dummy variable
because of issues related to invariance, it would be good to expose and document the following types * Data * TrainingData * VariableDefinitions