I think we're working from different definitions of "representative" The sense I mean it in is, "the election centers that have reported results are demographically representative, and therefore predictive, of countrywide results".
That might be the case, but you didn't provide any evidence to suggest it is, so the more likely explanation is that you thought I meant something else by the word.
The results are from all voting centers. Once a voting center finishes counting, it sends it to the data center. So the data are sent more or less randomly to the data center. You are right that this does not mean that are truly representative. But for the sake of simplicity we can call it "representative".
The map on the Guardian page that this HN submission points to suggests otherwise re: it being a representative sample. The districts appear to be relatively large, and most districts are reporting 0% counted while others are at e.g. 50%. The districts with any votes reported at all are also strongly clustered.
That might be the case, but you didn't provide any evidence to suggest it is, so the more likely explanation is that you thought I meant something else by the word.