Talk:NILFS

Comparison with ext3

Just a note, the comparison to ext3 is completely bogus - ext3 does not by design lose any partial writes (it can be configured to do so, but that is just an option), as all data can be forced through the journal, so this while comparison section needs to be redone, if this article shoulöd reflect reality in any meaningful way. The whole article reads more like fud/advertisement in favour of nilfs.— Preceding unsigned comment added by 91.89.172.190 (talk) 17:57, 11 March 2009 (UTC)[reply]

Another point of view: "all data can be forced through the journal". But it's not default option. It's usage have very negative impact on performance. Default journaling mode for ext3 is compromise between data safety and performance - it can provide good but limited data integrity. NILFS authors are assuming that their file system can provide much better data protection (I think we must wait a while to let NILFS mature to verify that), and in theory it's quite possible to achieve that without severe performance penalties.— Preceding unsigned comment added by Miloatwiki (talk • contribs) 22:01, 10 July 2009 (UTC)[reply]

Bad article

It looks like it was made by NTT or something. Seriously, "What is NILFS?"? I must merge that into description.

As well, the NILFS "to-do" list shouldn't be here. Maybe some highlights should go under a "Features" section, but that's about it. Nuke (talk) 02:51, 2 December 2012 (UTC)[reply]