My thought was not that one is inherently 'safe' on a journal that opts out of IP logging, but rather that getting the IPs via a file on a server that one controls is a far higher level of difficulty for the average user that I think most people involved in the technical aspect of Dreamwidth/IT understand. Heck, even having ever had control of a server bumps you in a stratosphere of technical competency that most people will never reach.

I know that it is really, really easy to get IPs with files, but your point involves a ramping up of my original argument that I think is unfair as well as greatly overestimating average technical knowledge. There's a difference between 'it could be done by someone with a fair degree of technical knowledge' and 'let's give out this information anyway.' There's always a way to get this information, but there's a difference between having to do something to get it and being handed it automatically. =/

Most people do not have the ability to get their own server, or even know how to upload a file onto it. However, everyone can see the IPs if they're auto-included in a comment. I'm glad that DW is having this discussion, but it makes me sad when this discussion is taken to 'well in X scenario' rather than really looking at the specifics of the question at hand.