Xmp specs for "description" field

Question

To whom it may concern (or to whoever may be able and willing to help;-):
In INDD I am able to enter xmp data into the description field as long as i want (which is also not cropped after saving), however the data is cropped to a certain number of cyphers/letters whenever in eported pdfs (no matter what export preset). I googled quite a while to turn up xmp specs regarding this, but could not turn up any. If anyone could tell my, why the cropping occurs or could turn up a workaround for it, i`d be grateful.
Kind regards and thanks ahead,
Aurel

Answer 1

Aurel and Carl,
I'm a little late here, but I want to thank Aurel for posing the questions, and Carl for explaining so much; I particularly appreciate his explaining the metadata consequences of Acrobat's "Save" vs. "Save As", and warning that changes in metadata embedding for CS5 applications [still] do not apply to Acrobat.
On a side issue, Aurel wrote:
> Also this pdf has all metadata visible upon vista explorer right click
> "file properties", which i cannot reproduce for my current files, but
> thats yet another problem. grrr. thanks anyways.
On my Vista64 machines, Explorer loses the PDF tab I see in on my 32-bit machine.  However, that tab come back when I run Vista64's 32-bit version of Explorer.  I use a one-line batch file with the line "%windir%\SysWOW64\explorer.exe /separate, /root, c:" as discussed in Adobe's Reader forum.
Like Aurel, I'm no programmer.  Still, I managed to make a custom XMP panel for IDCS4 for adding PRISM publication metadata to the journal articles I work on (using the generic models in the XMP SDK 4.1).  The PRISM metadata is fairly esoteric, but I've also been trying to convince the editors I work with that we should be embedding basic metadata, including Keywords, in PDFs to be made available electronically.  Thus I was more than a little surprised to read in one of Carl's post:
> One search engine for journal articles looks as PDF Doc Info
> metadata, most search engines avoid all forms of metadata.
That would seem to fly in the face of WWW folk wisdom (for example, at InDesign Secrets), so I'm must be missing something.
I see XMP SDK 5.1 has been released and  I expect to upgrade my CS4 suite one of these days.  However, it's tempting to wait until Acrobat's handling of XMP metadata catches up.  As things stand now, I'll probably continue to send editors screen grabs to proof custom metadata.  Otherwise I'd have to  port my modest custom panel to XMP 1 for Acrobat, convince the editors to buy the full version, and then show them how to install the panel -- effort that I assume the next Acrobat will make redundant.
Thanks again,
David