Say, why must HEAD go over the network, but GET needn't? Why can't HEAD just report from the cached copy according to the same rules determining its freshness as GETs?
When online with an already freshly cached copy, it would speed up the first step of: $ wget --content-disposition \ 'http://translatewiki.net/w/i.php?title=Special:Translate&task=export-as-po&group=core-1.14&language=zh-hant&limit=100' --2009-03-09 14:15:05-- http://translate... Resolving localhost... 127.0.0.1 Connecting to localhost|127.0.0.1|:8080... connected. Proxy request sent, awaiting response... 200 OK Length: unspecified [text/plain] --2009-03-09 14:15:08-- http://translate... Connecting to localhost|127.0.0.1|:8080... connected. Proxy request sent, awaiting response... 200 OK Length: unspecified [text/plain] Saving to: `zh-hant_core-1.14.po' 2009-03-09 14:15:08 (42.4 MB/s) - `zh-hant_core-1.14.po' saved [437934]
