Hi, please find below a proposal for a XML-FAQ format. I hope it covers all suggestions Hans and Henning have made in the last discussions. More suggestions are welcome. notes: - the label= in <contextfaq> can be used for cross-references - the id= in <author> is a reference to the author database as requested by Henning - the modified= in <revision> should be used to highlight updated FAQs in a new ConTeXt-FAQ issue - there is only one key for sectioning, since IMO the "application" should know where this section (specified by id=) is in the document tree. This should allow easy reordering of the FAQ - and a question: what is the better solution for different types of parenthesis in TeX code; the 1st or the 2nd <texcode> block? Tschö Jens. <contextfaq label="jum-0001"> <head> <author id="jum"/> <revision published="2002-10-27" modified="2002-10-27"/> <keywords> <entry key="clipping">clipping</entry> <entry key="clip"> <tex>clip</tex> </entry> <entry key="externalfigure"> <tex>externalfigure</tex> </entry> <entry key="figure">figure</entry> </keywords> <section id="graphics-manipulation"/> </head> <body> <question> Is it possible to show only a part of an external figure? </question> <answer> <p> The command <tex>clip</tex> allows to show only a part of a figure. You can either split the figure in rows and columns in order to show one of the resulting segments or you can specify a viewport directly.</p> <p>The following code tiles the figure in 3 columns and 2 rows and shows the segment in row 1 and column 2:</p> <texcode> <macro>clip</macro> <sqbracket>nx=3,ny=2,x=2,y=1</sqbracket> <braces> <macro>externalfigure</macro> <sqbracket>yourfigure</sqbracket> </braces> </texcode> <p> In order to specify the visible part directly you can use:</p> <texcode> <macro>clip</macro> [hoffset=3cm,voffset=2cm,width=4cm,height=3cm] {<macro>externalfigure</macro>[yourfigure]} </texcode> </answer> </body> </contextfaq>