NYCPHP Meetup

[inforequest]

csnyder chsnyder-at-gmail.com |nyphp dev/internal group use| wrote:

>Just to add some fuel to this fire:
>
><META> tags are toooo 90s, but if the client likes to see them in the
>code they aren't going to hurt anything (subject to John Andrew's
>observations, of course). They will only ever be used by internal
>search engines, because no one else is gonna be looking for them.
>
>The most robust and extensible way to do the same thing is to use
>embedded RDF with Dublin Core extensions, which is what Sir TBL was
>pimping in that 2001 article that Peter sent. At least if external
>search engines are looking for Creative Commons or Friend-of-a-Friend
>metadata, they will find and parse your RDF.
>
>Like many great solutions, RDF will probably never catch on because
>people just can't bear to think in tuples. Hence the Web 2.0 style,
>which is preferred by the in-crowd right now:
>
><link rel="robots" href="/robots.txt" />
><link rel="category" href="/path/to/category.html" />
><link rel="author" href="/path/to/author.html" />
>... etc.
>
>Until one or more of the major search engines puts up a sign saying
>"Semantic Web Now Open" with a list of names/relations to use on your
>content, it's all a bit of a crapshoot. If you have some internal
>reason to put metadata on your pages, then decide on a namespace and
>go for it. Otherwise, I'm not sure why you'd bother.
>
That's the point exactly... as php coder/developer, you're not sure why 
anyone would bother. But is that your call?

Meta data is "about" the content, and as such it may have a value of 
"empty" but it has to be able to accept a value if one is desired.If one 
is desired, it is likely to be contextual... so the coder needs to be 
prepared for that. Sure, it's not worth it... until one day someone 
needs it and it's a tear-out job just to make it possible.

<meta name="robots" content="noindex, nocache, follow"> is an important 
meta tag today for all the major search engines. It is 
context-dependent, where in this particular case the context is the page 
as the page's purpose is defined within the site. It is also temporal... 
it may need to be something some times  and different other times.  It 
may also be visitor-dependent, as itis in the case of IP-based content 
delivery. So what is the hook for this dynamic meta data? The coder 
usually won't know... the site achitect would, as would the strategist 
working at the SEO level.

In most cases I see (every case aside from my own sites?) there is no 
hook at all.

Now after we talk about the dynamic nature of meta tags, we can talk 
about the dynamic nature of anchor text. and then CSS positioning 
outside of content. And styles. And on and on.

I see so much talk about "dynamic sites" and yet so little recognition 
of dynamic "sites".... just dynamic pages. I could write a book about 
this... oh, wait a sec....

-=john andrews
http://www.seo-fun.com









-- 
-------------------------------------------------------------
http://www.seo-fun.com

"writing is often like driving a truck at night without headlights, losing your way along the road and spending a decade in a ditch"  -- Gay Talese