>> I cringe at the buzzword-of-the-moment aspects >> of XML, but concede that it is at least worth considering as a format.
> Have you ever editted an XML config file? It sucks. Possibly not as bad > as an X config file, but definitely worse than .freenetrc. You'd usually edit the file with an app of some kind that runs it thru an XML parser first. Yes, this runs against the whole "edit with vi, and put a gui wrapper on it for those who don't care what goes inside" philosopy. I'm not a big XML advocate by any means, but I have to acknowledge the validity of what appears to be the core idea: it no longer makes sense to store data in a zillion different proprietary formats with a zillion different quirks and subtle bugs. Programs reading or writing to the file will almost always have to go through an XML parser, but so it goes. At least you can read it manually if you like, and edit it manually if you truly have to. Are we really better off making up our own format, and forcing *every* variant Freenet and/or client program to write their own parser to handle it? Maybe we are. Like I said, I'm not a big advocate. >> [regarding remote config] Oskar and Scott think I'm nuts, to put it >> mildly. :-) > Yeah, that's crazy. I just don't see why you want to dynamically configure > your node. Just write a new config file and restart. Yes, in many systems > that would be annoying (like Windows, for instance), but in this case it > seems more sensible to me to restart after changes. What kind of changes > would you want to make without restarting? And why do you want to do them > without access to the machine the node is running on? Avoiding a restart is not the issue I was looking into. By "dynamic", I meant that adjustments to resource usage would occur fairly often, and would usually be under programmatic rather than manual control. Avoiding restarts would be nice in that circumstance, but restarts per se are not the point of contention. I'm still pretty fuzzy on what I want, and the idea touches on issues that could start a lengthy exchange. Right now let's just focus on the 0.3 stuff that has a chance of being implemented. What we learn from that experience will help me hash out the details of what I'm trying to pursue. Who knows, it might even bring me to my senses. :-) --Will (not speaking/writing/mumbling for his employers) willdye at willdye.com _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
