Adrian Aichner wrote: The core idea is that it will do a strict validation against the relevant DTD. None of Weblint's "I think thats a bad idea" stuff, but actual objective "This is correct (see section x in SGML spec); this is incorrect (see section y in SGML spec)". It currently has the added features of contructing an outline of the document based on Hn elements and of showing the logical parse tree; both as an aid to understanding what went wrong and how to fix it. Basically to visualize how an SGML processor sees the document. There are several extended features that are on the TODO list. ";imgonly" is one. A report in XML for easy parsing and possibly an XML-RPC/SOAP calling convention (which might make a nice interface that XEmacs could use, BTW). Options to canonicalize input, convert it to XML, or perform other transformations on it. All the hard work is being done in James Clark's SP (and possibly OpenSP and expat in the future); we're just builing a "friendly" shell on top of it. However, all that is waporware for the forseeable future. I intend to devote some time to it over the next few months, but 1) I'm not the most reliable of people in this department (Real Life has a nasty habit of interfering) and 2) Gerald Oskoboiny (the maintainer over at W3C; wrote the "Kinder, Gentler, HTML Validator" (which the W3C Validator stems from) if you recall) has a turnover time for patches that can only be described as glacial! Source is available from the W3C CVS server (links can be found from <URL:http://validator.w3.org/>) and I'm trying out chronicalling my development efforts using Advogato's Diary at <URL:http://advogato.org/person/link/>;. BTW, I'm mucking about with it because I use the Validator internally so my mods aren't always relevant to the W3C service. The net result is if anyone wants a feature, but don't want to implement it, I'd be happy to look into it, but I have no say in what the W3C does over at validator.w3.org. All I can do is supply patches, Gerald decides what goes up on w3.org.