summaryrefslogtreecommitdiff
path: root/notes.txt
blob: e8262ac5c5aa4389a0e38f7add89800de3ccbc1a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31


notes
-----

Preference deduplication isn't worth it unless there exists some linear dedupe algorithm.

Multithreaded preference read-in isn't worth it unless the entire preference file is read in first,
which would increase memory usage to unacceptable levels. Might implement this anyway later as an
optional feature.



future direction
----------------

Util to list paper ids that fit specific criteria to doublecheck potential errors.

Test tiebreaker handling, specialise for federal/state elections.

More parsers/options to handle state elections, general STV data.

Multithreaded operation?

Make the goddamn results correct.
  Are they correct now? it's a bit murky with how the AEC records transfers in DOP logs.
  Does the AEC use truncated values or exact ratios for transfer values?
  Does the AEC check for candidates having quota when transfers are only partially done? (Tested. This probably isn't it.)
  Should partial votes be recovered when they total to a full vote from multiple bundles?