Re^2: Is there a simple way to archive/download all of PerlMonks?

[marto]

"They already have an archived copy, after all."

They don't. I've checked several thread URLs, most didn't exist, those that did had snapshots that were years out of date. Besides, it doesn't match the criteria of the question, an 'offline' copy.

[LanX]

> thread URLs, most didn't exist, those that did had snapshots that were years out of date.

Did you check all 6+ domains? :)

I'm wondering how likely an old thread can be out of date, do you expect many monks updating what they wrote at 9/11?

Edit (answering myself)

Hmmm wait, besides editing we have indeed necroposts resurrecting old threads.

True, a backup service would need to check RAT or newest nodes regularly. (Or refrain to mirror only single posts)

And editing isn't recorded anywhere... :/

[marto]

I didn't check all of the domains, no. I think the only valid archive would be an up to date database extract of node content (and some of the other metadata), rather partial snapshots of page impressions from a moment in time.

Update: I seem to recall different domains having different robots.txt rules to impact indexing.

[LanX]

Otherwise I'd poll the XML per node.

If edits are/were reflected in the timestamps of the http headers, this could also be quite efficient in fetching updates.