MC> I looked at the version that Dale had put up, and I considered a lot of
 MC> the discussions I have seen and participated in, and then I decided to
 MC> make my own version of an xml fidonet nodelist.

 MC> I started with a DTD,

Why DTD and not a schema?

 MC> and then I made an XML file pointed at the DTD, and started entering
 MC> node information to walk the tree down to my address (1:211/107).  I
 MC> realize that it is incomplete, and there are probably things that
 MC> someone might want moved around, but I looked at this from a
 MC> perspective of experience I have had with walking xml trees using the
 MC> DOM, etc.  I think that the structure should be such that an
 MC> application can look at a node, check the attributes, and then decide
 MC> if it needs to step into the node.  That is why I use attributes, to
 MC> help with the processing decision of whether to look at the nodes data
 MC> or skip it.

OK. We've had some discussion about Attribute Vs Child.

 MC> The DTD includes a connection type of radio, but there is no sample of
 MC> this in my xml sample file. (use your imagination :}).

 MChttp://www.logartsol.com/ftn/default.htm,
 MC> which has links to both the sample xml and the dtd.  If you would
 MC> prefer to go right to the sample, the url is
 MChttp://www.logartsol.com/ftn/ftn_nodelist_sample1.xml.

 MC> Let me know what yous guys think :)

It is good to see some else interested! That is what I think.

With best regards, Dale Ross.  E-mail: Dale.Ross{at}p1.f1.n379.z1.fidonet.org

--- Fidolook Lite FTN stub