metatags (was in - Re: [Typo3-dev] You are the Programmer)

dan frost dan at danfrost.co.uk
Sat May 22 16:59:04 CEST 2004


Dear all,

I thought you'd all like to know that we (3ev and I) have just finished 
a first version of a fully-functional metadata extension. Current 
version has the UK's government list imported from CSV (but could easily 
be from XML), "mount points" in the metadata category tree, keyword to 
category mapping, scanning of uploaded documents, Bayesian learning 
logic to improve suggested categories each time, possibility of 
publishing all your categorised content via SOAP, tags in the header and 
so on and so on...

This is still in development, but if you're interested then please 
contact me.

Regards,
dan


Martin T. Kutschker wrote:
 > Patrick Gaumond wrote:
 >
 >>
 >>> A- In general we could setup a few info "system-wide" (content-type,
 >>> shema, others like publisher or rights) values that should be
 >>> overwritable on a page basis.
 >>
 >>
 >> Extended keywords (n_extkeywords) does something similar I think. From
 >> the description: "It also adds the possibility to have the keywords
 >> accumilated over the rootline enabling one to set site-wide keywords
 >> along with branch-wide and page specific keywords."
 >>
 >> It doesn't replace the standard value, but appends to it, if you
 >> configure it that way.
 >
 >
 > Yes. Something like that was already on my mind.
 >
 >>> B- Language should lookup in Typo3 so we don't need to setup manually
 >>> the user selection in case of site with multiple language. (I don't 
know
 >>> if possible in TYPO3)
 >>
 >>
 >> Definitely the way to go. TYPO3 knows the language of a page from the
 >> "Website Language" record, which we might need to enhance, as we need
 >> the ISO language shorthand for DC, and the language record in TYPO3
 >> only knows a rather descriptive title.
 >
 >
 > See the "Keyword wizard" extension. Keywords are not translated during
 > rendering, but the UI shows only the keywords of the pages' language
 > (restricted to the domain record as well).
 >
 > Is this what you had in mind?
 >
 > Additionaly/Alternatvely I can imagine a set of id-based keywords (ie
 > records with a uid)) with translations in the db. With this setup
 > id-based all pages (incl. alt.lang. pages) will get the keywords from
 > the default lang page. The actual (translate) keywords come from a table.
 >
 >>> C- Author, Creation date and Modification date should come from TYPO3
 >>
 >>
 >> True. Even other data might be gathered automatically (e.g. if you
 >> already supply a description for the 'normal' meta description tag,
 >> this could be reused). Further data may be fetched from the
 >> corresponding fields for pages of type advanced.
 >>
 >> After all, we should probably talk to Kapser and discuss the option of
 >> enhancing the 'Meta tags, extended' extension directly (maybe taking
 >> over maintenance). It does a basic set of things we need, and would
 >> 'only' need to be enhanced with the automatic gathering of data where
 >> applicable and the option of providing page-specific entries.
 >
 >
 > Yeah, but make all this configurable (or at least overridable).
 >
 >> Maybe we should write a 'formal' project description to base further
 >> discussion on?
 >
 >
 > Why not? With specs the code should be better from the start.
 >
 > Masi
 >




More information about the TYPO3-dev mailing list