BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//WMTYLV
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-2023-WMTYLV@talks.osgeo.org
DTSTART;TZID=CET:20230628T160000
DTEND;TZID=CET:20230628T163000
DESCRIPTION:FieldMaps.io is a personal initiative originally created to dev
 elop offline interactive reference maps for humanitarian actors. However\,
  in short time\, it transitioned to helping develop common operational dat
 asets that form the foundation for humanitarian response planning. Over th
 e past 2 years\, enormous effort has gone into releasing a high-resolution
  composite dataset able to be updated daily from multiple sources. This ta
 lk will cover 3 aspects of the project.\n\nAlgorithm\n\nEdge-matching reso
 lves gaps and overlaps between hundreds of separate national data sources\
 , requiring an algorithm that can perform at global scale. The resulting m
 ethodology uses something akin to a euclidean allocation raster applied to
  vector space\, free of the compromises other approaches like generalizati
 on and snapping make. If you've ever been challenged by topology or data c
 leaning\, you might find some insights into solving your own problems with
  the ideas contained here.\n\nPipeline\n\nThe edge-matching algorithm invo
 lves multiple complex and computationally intensive steps. Although Geopan
 das and GDAL usually come to mind when building multi-step geoprocessing s
 cripts\, PostGIS ended up being the fastest and best scaling tool for tran
 sforming gigabytes of vector data. I'll challenge your assumptions of how 
 it can be used to create pipelines on both desktops and in the cloud\, and
  make a case for why you should include it in your next project.\n\nSource
 s\n\nA composite dataset is only as good as the foundations it builds upon
 \, and great care was taken in selecting which sources were used in this p
 roject. For international boundaries\, I'll go into detail about how I use
 d only public domain sources to create an ISO 3166 compliant dataset. At t
 he subnational level\, I'll highlight two projects that each curate update
 d administrative boundaries: one by the United Nations\, another by an aca
 demic institution.\n\nWhether you're a remote sensing specialist in search
  of the best topologically valid boundaries to run zonal statistics with\,
  a Python developer frustrated by your pipelines constantly running into m
 emory limits\, or just want to run this tool on your own boundaries\, I ho
 pe you come away from this talk with a valuable concept you can apply to y
 our own work.\n\nData: https://fieldmaps.io/data\n\nTool: https://github.c
 om/fieldmaps/edge-extender
DTSTAMP:20260422T232718Z
LOCATION:UBT C / N111 - Second Floor
SUMMARY:Creating Global Edge-Matched Subnational Boundaries - Maxym Malynow
 sky
URL:https://talks.osgeo.org/foss4g-2023/talk/WMTYLV/
END:VEVENT
END:VCALENDAR