Hacker News Reader

Comment by manaskarekar | Hacker News Reader

You know, I do this a lot. It affects me in interviews as well. But the paranoid OCD geek in me wants to retain a local copy of all the information in case the internet ceases to be, and as much as I can in my brain.

I have considered locally copying wikipedia but given up after the effort that might be required. Anyone got any experience or tips?

This is also why I love man pages.

Edit: Just in case anyone is not aware check this out:http://en.wikipedia.org/wiki/Wikipedia:Database_download

I was asking if anyone knew of the easiest way and I see some options below. Thanks, I'll look into those.

View on HN

There is an iOS app called "Wiki Offline" which gives you a local copy of the English Wikipedia that has been compressed to around 3.5GB.

You can get a monthly XML extract for the English Wikipedia here:

http://dumps.wikimedia.org/enwiki/20120802/

Your choice whether to get the current snapshot or an archive of all edits from the beginning of time.

Amazing how 7zip can achieve 100-1 compression, so that all articles plus all edits are only 62 GB compressed.

He who controls wikipedia controls the future.

http://www.kiwix.org/index.php/Main_Page

Not sure why dumps don't occur more often though.

I also like this http://thewikireader.com/

That's awesome! I wonder if there are any options with a bigger screen, or if any of the ereaders can be made to store wikipedia locally.

> I have considered locally copying wikipedia but given up after the effort that might be required. Anyone got any experience or tips?

Try the wget command line tool; it's generally useful for site-mirroring. "wget --mirror" will recursively download an entire website; if it is impractical to download all of Wikipedia, you may test with the option "-l <recursion depth limit>" to get only a part; then "-p" will make it get all images and whatever needed to display the pages in question. And if you want to only get the text pages/reject the images, the "-A" and/or "-R" options may be useful. Consult the man page and experiment.

Please don't do this; Wikipedia provides dumps for exactly this purpose:

http://en.wikipedia.org/wiki/Wikipedia:Database_download#Eng...

(Further explanation regarding the use of robots, spiders and 'wget -m' here: http://en.wikipedia.org/wiki/Wikipedia:Database_download#Ple... )

Reply on news.ycombinator.com