Need XML advice

I need some advice on this.  How would you format the XML?  I have an idea, but want a second opinion.  I'll post my idea tomorrow.  Also, should I write an DTD or a schema?  One or the other is required.  Which is more difficult and which is more extensible? Background: For an outside project I have been tasked with converting a number of human readable geophysics survey reports to XML to be imported into a modelling package.  The modelling package requires either a DTD or a Schema and another text file that translates the XML fields into the fields it requires.  Any fields tagged as a description or that it does not recognize is added to a description section for the model. Here is a snippet of my human readable report: === Project Name:  Southern Death Valley Gravity Survey Project Date:  June 1989 Data Collection Date:  December 1988 Data Collection Date:  March 1989 Data Collection Date:  May 1989 Surveyor:  lfs Surveyor:  rrm Surveyor:  ccg Surveyor:  lcm Surveyor:  tgp Description:  Data collected in the Southern Death Valley area trying to model the Garlock fault zone near the eastern terminus of the zone. Reference Benchmark:  Shoshone, California (36.2 N 116.4 W) Terrain Correction:  true Model Name:  SDV NS 1998fs Orientation:  NS015 Starting Point Lat:  36.68 N Starting Point Long:  116.36 W Data Point:  1, -12075, 9.8, 458 Data Point:  2, -11684, 9.5, 423 Data Point:  3, -10900, 8.6, 389 Data Point:  4, -9986, 6.3, 453 ... Data Point: n, 15698, 159.2, 1036 === The modeling package requires the orientation, direction, starting point and the data points (ID, location on a line in meters, gravity reading and magnetic reading).  The data points may or may not be in order of either ID or position on the line.  Additionally, the IDs are aplha numeric as opposed to straight numeric.  Also, the modelling application can accept a straight numeric additional ID for each point.

Here is my idea: <project name="Southern Death Valley Gravity Survey">   <description>     <date>June 1989</date>   <datacollection>       <date>December 1988</date>     <date>March 1989</date     <date>May 1989</date>     </datacollection>   <surveyors>     <surveyor>lfs</surveyor>     <surveyor>rrm</surveyor>     <surveyor>ccg</surveyor>     <surveyor>lcm</surveyor>     <surveyor>tgp</surveyor>   </surveyors>   <descriptiontext>Data collected in the Southern Death Valley area trying to model the Garlock fault zone near the eastern terminus of the zone.</descriptiontext>     <referencebenchmark>Shoshone, California (36.2 N 116.4 W)</referencebenchmark>   <terraincorrection>true</terraincorrection>   </description>   <data modelname="SDV NS 1998fs">     <orientation>015</orientation>   <direction>NS</direction>   <startingpoint>     <lat>36.68N</lat>     <long>116.36W</long>   </startingpoint>   <datapoints>     <point seqno="1" id="001">       <distance>-12075</distance>     <gravity>9.8</gravity>     <magnetic>458</magnetic>     </point>     <point seqno="2" id="002">       <distance>-11684</distance>     <gravity>9.5</gravity>     <magnetic>423</magnetic>     </point>     <point seqno="3" id="003">       <distance>-10900</distance>     <gravity>8.6</gravity>     <magnetic>389</magnetic>     </point>     <point seqno="4" id="004">       <distance>-9986</distance>     <gravity>6.3</gravity>     <magnetic>453</magnetic>     </point> ...     <point seqno="n" id="nnn">       <distance>15698</distance>     <gravity>159.2</gravity>     <magnetic>1036</magnetic>     </point>   </datapoints>   </data> </project> ----- Excess quoted text cut - see Original Post for more ----- ... ----- Excess quoted text cut - see Original Post for more -----