james, yes, you have discovered that marcello cares more about fighting the indiscriminate downloading robots than taking care of the real humans... but something tells me angel michael is entertained by the notion that there's "indiscriminate scraping" of p.g. e-texts. but i suppose that, as the lead dog -- (a little sledder talk for newby) -- gutenberg.org gets lots of headwind. so... just choose one of the mirrors! i'd suggest that greg can point you to a certain mirror that's more in line with the assured-access that don remarked should be associated with an endeavor like this one founded by michael hart, an attitude i'd guess rubbed off on greg. because i'll personally attest that urllib is working just fine to grab files from it. but, if i might suggest an improvement, greg, could you authorize c.o.r.s. for it? (at least it appears to me it ain't working.) i'd need nothing more than "get" access, which i think poses absolutely no danger. but also, greg, just to avoid cors hassles, maybe i should re-request one of those accounts you so often promise to people. (but feel free to embargo any javascript i'd serve until you could verify its safety.) i'm about to (finally) make some big moves, and although i no longer feel any need to include project gutenberg, my hart is not so cold that i feel a need to exclude it either. so if you want to come along for the ride, just crawl in the back seat. and buckle up. -bowerbird