loosely coupled semantics

[Jeff Stai]

I was reading this page about microformats:

http://en.wikipedia.org/wiki/Microformats

The explanation about microformats and loosely-coupled semantics was easy for me to follow, as someone who understands and sometimes codes HTML.

I can see from this explanation that it CAN be friendly to the person who codes HTML. In the example, you can take the body of code in an "hCard" format and drop it into a printable signature block without changing it.

But how important is this in a world where almost nobody actually codes HTML?

I can see one thing: having the semantics be something nearly directly printable as HTML means that code I embed in my web page or blog post is visible and I can instantly verify it is correct. If instead it is hidden I might notice my embedded code is for the Viognier when it should be Syrah...

But what if some of the information is presented graphically? My wine web pages are an example of this problem, for example:

http://www.twistedoak.com/tempranillo.html

In my case I would prefer to have the code hidden.

Question that I don't know the answer to: If we hide XML on a page, will google index it?

- j

--
Jeff Stai j...@twistedoak.com
Twisted Oak Winery http://www.twistedoak.com/
Winery Blog http://www.elbloggotorcido.com/

[Jason Coleman]

Jeff, I'm a big fan of your wine geek sheets (see the link in Jeff's email). I'm glad they've made their way into the discussion here.

 

As for what to do about information that is show graphically: it is possible to embed the same information in html and hide is using CSS styles. So if you had this markup above or below your image:

 

<div class="geek_sheet">

  <span class="residual_sugar">0.01%</span>

  <span class="titratable_acid">0.63</span>

  <span class="volatile_acid">0.071</span>

  <span class="ph">3.48</span>

</div>

 

... you could make it invisible on the page using the following CSS code:

 

div.geek_sheet { display: none; }

 

There is a problem with this though. I'm pretty sure that Google and other modern search engines are programmed to ignore hidden text. The solution would be to make sure that all of this juicy information is inside the images alt tag if it's important to indexing. Google and other search engines actually put a precedence on keywords found in alt tags.

RE: human-readable vs. machine-readable.

 

I need to do a little more soul searching on this one. Everyone is making good points. You don't need the microformat to be human readable if you have proper tools to get the data out of the writer's head and into the format. So what are we gaining by having the embeded data human readable?

 

Folks should also remember that XML itself (which is relatively falling on the machine-readable first side of this discussion so far) was actually developed to be more human-readable than other data formats. 

 

I think Jeff is hitting on something when he talks about being able to visually verify the correctness or completeness of the format by looking at the code. By human-readable, we're not just talking about the end users. Programmers are humans too. Most of them.

 

One advantage of the microformat scheme in particular, of embedding meaning into the class attributes, is that this kind of activity is already very familiar to web developers and designers. Those same class attributes can be used to style the data. And the data is built out of valid html elements, which makes it easy for coders to move the stuff around.

 

- Jason Coleman

[Werner F. Bruhin]

Hi Jeff,

Jeff Stai wrote:
> ...
> But what if some of the information is presented graphically? My wine web pages are an example of this problem, for example:
>
> http://www.twistedoak.com/tempranillo.html
>

Were I would see that vinoXML would bring something to the above site is
on this type of page:
http://www.twistedoak.com/geeks.html

When one clicks on a particular wine one gets a PDF like this:
http://www.twistedoak.com/pubs/2003_murgatroyd_geek.pdf

Very nice and comprehensive information but now how is someone getting
this into his cellar book? Retype or copy/paste are the only solution.
If you would use vinoXML you would keep the PDF, as many people are used
to and like this format, but you would provide a second link which
allows your customer to download a vinoXML instance.

Have a look at the very simple example we put on the site:
http://www.vinoxml.org/html/how_to.html

> In my case I would prefer to have the code hidden.
>

Absolutely, XML is only for the techy geek and should never ever be
shown to a user. It is just a nice why to store data in a format which
many different tools can read and transform into a form which is
consumable for a user.

> Question that I don't know the answer to: If we hide XML on a page, will google index it?
>

I can't not find a definite answer, but as Google uses XML for some of
their own stuff (see e.g.
https://www.google.com/webmasters/tools/docs/en/protocol.html) I would
think they will - probably more likely then them indexing e.g. a PDF file.

Werner

--
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html>
<head>
<meta http-equiv="CONTENT-TYPE"
content="text/html; charset=iso-8859-1">
<title></title>
</head>
<body dir="ltr" lang="en-US">
Cordialement,
<br>
<b>Werner F. Bruhin<br>
</b><br>
Le Livre de Cave<br>
<a
href="http://www.thewinecellarbook.com/fr">www.thewinecellarbook.com\fr</a><br>

<p><br>
</p>
</body>
</html>