2 Feb 2009 03:43
Fuck the cloud, but use it anyway
Ben Finney <ben <at> benfinney.id.au>
2009-02-02 02:43:09 GMT
2009-02-02 02:43:09 GMT
Howdy all, Thanks to Don Marti for the compare-and-contrast <URL:http://zgp.org/~dmarti/www/beware-the-cloud/> between Rick Moen's excellent WINOLJ screed and Jason Scott's more recent rant. There are a number of online services that take efforts to *not* keep data in a silo; e.g. feed readers that provide OPML documents of all my subscriptions, or bookmark sites that allow one's whole set to be downloaded in [some open format that I can't recall right now]. So, if I'm a clever smug bastard because I give preference to services that do allow constant access to my data in open formats, what's a sensible strategy for automatically backing all this online stuff up against the event of any one of those services suddenly not providing what I want? I could ‘curl http://feeds.example.com/bignose/opml/’ (or whatever the thing is I want to keep safe) on a daily basis; but then it occurs to me that I might not *notice* that one of those services is serving up garbage where my stuff used to be, until it's too late and my backup is overwritten with a perfect copy of garbage. That leads to “real” backup, of course, with all the administrative burden: volume management, rolling archives, indexes, blah blah. I know where to look for tools to do this for networked filesystems. I don't know where to look for backup tools for online services, since I'm a curmudgeon who only grudgingly got into this stuff so it's all a bit new for me.(Continue reading)
> there's a movement afoot to formally define what keeps a site from
> being a dangerous walled garden:
>
>
RSS Feed