← The land

Farm PAR-000010

Rietfontein

· show on the map

Portion
subdiv of 169
District
Prince Albert

This parcel has no matched survey polygon, so it cannot be drawn on the map. That is a gap in the cadastre match, not an absence from the record.

Why this farm is here

A piece of a farm already in this model. Not yet placed on the map.

What the record holds, in order

  1. undated boundary change subdivision: from Middelwater

Research notes

subdivision of Middelwater [TO VERIFY - probably wrong]. [29 Jul 2026, verification pass] The only support for this parcel is one unsourced legacy line ("20th c. - subdivided; a portion became Rietfontein", F01 prose) and PLK-000002 (no document). DOC-0039 (the 1883 verso) lists NO Rietfontein deduction from farm 169, and the modern cadastre has none. This looks like the two-Middelwaters trap run in reverse: the documented relation is Rietfontein (Zeekoegat district, PAR-000052) subdivided into a SECOND Middelwater - not our No.169 subdivided into a Rietfontein. Cadastre match stripped 29 Jul 2026: the record it wore (Ptn 38 of farm 56) belongs to Rietfontein No.56 south of Leeu-Gamka, unrelated to farm 169. Parcel kept pending

HE IS RIGHT, AND IT IS WORSE THAN A DUPLICATE. Both rows were still carrying the SAME cadastre key W052C061000000000056000380 and the SAME centroid -32.880993, 21.889268, which is Ptn 38 of RIETFONTEIN No. 56, south of Leeu-Gamka. The 29 Jul verification pass recorded that this match had been stripped; the note said so and the columns did not. So the two rows were not the same farm, they were two different farms wearing a THIRD farm's polygon, and the index showed both as 38/56. Key, centroid, area and match method cleared on both. Neither is pinned now, per 5: an unresolved place gets no pin. Whether these two rows should be one row is a separate question and is LEAD-0058.

[1 Aug 2026] NO-POLYGON SET, because clearing the key was not enough. Both rows have no farm number, so match_cadastre.py fell back to matching on NAME and put them straight back onto Ptn 38 of Rietfontein No. 56 near Leeu-Gamka - a third farm, and the same wrong pin they carried before. The sentinel is the only thing that stops a name match re-deriving it on the next run.

The research behind this page, kept separate from it: dated working notes, open questions, and what is not yet settled.

Open questions

What the record is missing

Found by breaks.py reading the model against itself, not written by hand, each one is a shape of missing record, not a fault in the data. The whole set is on the questions page.