Chad Carr wrote:
Very neat. Very complete. Do you have any test data (other than that in the comments) to fill this table? Especially GMPackageVariables and GMPackageVariableValues. This would be very helpful in what I am doing now to attempt to define a hierarchical config-db for the leaf box itself.
I'm really impressed by the .sql file so far, and I appreciate the fact that Greg has taken the time to review the developers documentation by David D that he cites in the file. I don't know how many people ever take on the project of cataloging every last detail and variable about an operating system in their lives. I'd just like to say that this critical foundation step could take you 6 months, and I would be just as happy as if you finished it this week. So please take your time, partly to have enough chances to sleep on it, partly so that folks don't consume all the time slices in their lives. Time-sharing is cool. No matter how much XML makes my skin creep, I'm ten times more curious about the amazing things that could have kept Greg interested for a whole year on this subject. So though I too feel strongly about simplcity, I think Greg needs a chance to show just how simple his complex system makes things. If we ignore some of the higher level Anaconda/Python aspects of Greg's idea for now, I'd like to have a chance to learn where Chad's and Greg's ideas intersect. I'm not sure they're all that differnt.
I also like how we both came up against firewalling and ipsec and put it
off till later! Those are definitely the two hardest problems! Microsoft has demonstrated how difficult it is to come up with
comprehensible GUIs for these two items (by failing to do so).
This is what got me to post, as I'm haven't done db mastering in 10 years, complex routing. I wrote a firewall for LRP a few years ago, rc.pf that used ipfwadm in the days before ipchains, but I never could get myself to write a more complicated firewall using ipchains that handled things like ipsec, dhcp, dmzs, proxy-arp, and the like because I just don't have access to complex networks. Charles came along with his versions, and with his far greater array of hardware and network topologies, he was able to grind his way to scripting the setups. So we can qualitatively reflect on that, but how do we quantize it? How do we (how does Greg) determine what all those values and procedures are to build proxy-arp and whatnot without either learning it all himself or sitting down with Charles and talking about it? We may need a live way of communicating, meeting once a week in a chat room or using mics or something. So in closing, I guess if I communicated more with others, I could have gotten more help on the parts I couldn't test. Cheers, Matt ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ leaf-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-devel
