How YOU Can Make the Web More Structured - Page 2
Tags in Blog Posts
The concept of tagging, which was popularized by services like del.icio.us and Flickr, is now
commonly understood and is ubiquitous. The idea of humans tagging content to categorize it and later to find it is a simple, yet
important bit of the web infrastructure. Most major blogging platforms support tags. The tags are standardized based on the
rel-tag microformat. You can see the implementation on ReadWriteWeb - each
post is tagged with a set of tags.
For example, one of our recent posts contains this tag:
<a href="http://www.readwriteweb.com/tag/twitter" rel="tag">twitter</a>
The tag has several benefits:
- Readers can instantly click to find other posts with this tag
- Search engines can better classify the content
- Semantic tools can offer additional services such as finding related content, pictures, and video
Tags are similar in principle to keywords, but provide more flexibility because they are inside the post and can have richer content. In principle, it could be possible to add more information into the keywords meta tag in the head of the document but it has existed in its current form for several decades and is thus probably not likely to change. In any case, all modern blogging platforms make it trivial to tag content, so there should be no excuses.
Standardizing Blog Templates Across Platforms
In the nineties people created web sites. These days only companies have web sites, individuals
have blogs and social network profiles. There is a great opportunity to standardize and structure the information because blogs
and profiles are based on templates. Consider a common structure for each blog. One or a few sidebars and the central area for
the content. In the content area, on a post page there is a post body, date, author and tags - a minimum set of elements.
Why not standardize on a few things here?
- <div class="post"> - a container for the post body
- <div class="sidebar"> - a container for the sidebar
- <div class="author"> - a container for the author
- <div class="date"> - a container for the date
- <div class="tags"> - a container for the tags
- <div class="comments"> - a container for the comments
Platforms already do have very similar things in place and standardizing between them is rather simple. In no way would this be a competitive advantage or disadvantage to them, but it would be a big help towards making the web more structured. Extending on these basics, it would also be helpful if widgets were wrapped into standard enclosures. A simple widget tag can go a long way toward distinguishing widgets from the other content in the sidebar.
If blogging platforms standardized on these basic conventions, likely major newspapers would follow as well.
The situation with social network profiles is different, as the information contained in them is not public. In addition, there is a competitive advantage to Facebook in having its own proprietary structure. However, entities like the DataPortability group have been created precisely to deal with this problem and Facebook just joined. So we may yet seem some progress on that front.
Comments
Subscribe to comments for this post OR Subscribe to comments for all ReadWriteWeb posts
Microformats rock! It's one of those things that you wonder why nobody thought of them sooner. I'm excited to see how the new API in FF3 will allow better exposure of microformats in the future.
A fundamental microformat is still missing: one to represent measurable physical quantities (i.e. Any Data Value!) into semantic HTML.
This would allow browsers/plugin and js scripts to search and collect data from various URLs (in any unit-of-measure) and proceed straight into processing and mashup tasks; no more scraping, nor Readme.txt to describe how to interpret the downloaded .csv file.
Some preliminary notes are on the microformats wiki, but nobody seems interested enough to get the discussion going and the 'hmeasure' thing out of the Draft-state limbo.
Regarding the standardising of blogging templates, there has been some excellent research/analysis done on that front to see if there are any significant numbers of common objects in HTML.
After a bunch of folk provided some excellent data, Google thought they'd step in and provide a small billion document set for comparison.
You can find their work and links to others on at Google's Web Authoring Statistics.
Bravo - we'll get to that semantic web...even if we wear are fingers down with the extra typing!
For the structure for blog posts, I recommend following the microformat hAtom:
http://microformats.org/wiki/hatom
Excellent post. Excellent subject. But I do not see any mention of the "anti-structure" forces at work.
I was a Joost pre-beta tester ONLY because of their ( now retracted ) promise of annotated television. Some secret meeting occurred at Joost and the decision was made to abandon RDF and/or Microformat markup for the videos. No explanation. Nothing.
Also, I want to bring to your attention the recent killing off the WC3's plan to have plain mark-up in HTML 5 for videos by Adobe:
http://www.digitalcitizen.info/2007/12/30/ogg-theoravorbis-as-default-for-video-scuttled-in-html5-spec-who-benefits/
Ogg Theora isn't that great of a file format, but it shows that our altruistic attempts at making a "Read Write Web" will be stopped by very greedy, close minded people.
rather than using a class of post for the post container. Wouldn't it make more sense to use a class of hEntry and go with the hAtom microformat?
Microformats, standarized blogging templates, and even HTML meta tags are all great ways of exposing additional "structured" information in web pages.
Additionally, services like Orchestr8's AlchemyGrid ( http://grid.orch8.net/extractions/grab ) make it easy to 'apply semantic structure' to existing websites, exposing their data to semantic webapps, mashups, and other stuff.
Really useful article. Web structuring is more important as it may help site to get good rank in search engine.
Looks like this page is missing the keyword, date, and location info ;)
I agree; this is a great aritcle. I think that having a more structured web will not just help search, but will help usability as more and more people are getting away from the alrge-screened terminals for accessing the web. Having the abiltiy to search distinctly, and then manage what you've found seems to be a ripe recipie for making mobile and vertical applications of the web grow healthier.
Of course, when semantics are entered into the equasion, advertising can become much more targeted, and trends are much easier to notice across users. Privacy folks could have a ball with this kind of info. And that's the cost of being more organized, more people can find ya.
Sounds great, but be carefull how much weight you give this meta data. If maybe there was someway the page author (or blog template for that matter) can have a very small blue square in the upper right most part of the page that when clicked allowed the user to thumb up or down the authored meta data as well as supplement.
This can allow you to give more weight to top down meta based on qty of thumbs while allowing a community to validate it from the bottom up. Best of both worlds?
@LucaPost:
Agree on your point about measures, discussion does seem to have stagnated on this. I've looked into the hRecipe spec and that touches on the need to mark up quantities. I see some issues with the abbr pattern in that parsers would have to know the spelling and misseplling of measure units in order to convert them.
Also there are a wide range of units out there and marking all of them up in a standard way is a challenge and I think this is why not much is going on here. I don't think it's apathy from the very small microformat development community, more a case of the difficulties involved with quanity units.
Please feel free to join in on the mailing lists.
Cheers
Lee
yep, just few of these steps have a great impact on your SERP's.
In terms of location (geotagging) metadata elements or microformat tags are all very well and good--if you're a machine. They aren't generally visible to humans who may otherwise miss the fact that location-specific information is associated with a particular blog post/ photo/ feed etc. That's why I propose a web "standard" Geotag Icon to add visual identification to geotagged content:
http://www.bioneural.net/2008/02/21/a-web-standard-icon-for-geotagging/
If more people see other people tagging their content ("What's that icon?"), more people will start tagging their own content.
Wordpress.com’s semantic tools such as categories, tags, urls for individual posts, author’s name generated automatically to each post, dates per post, seem to mimic the function of metatags and are Search Engine friendly. As well, when my del.icio.us tags and wordpress.com’s tags and categories are syncronized, I think this performs a similar role to metatagging. To make it even more elegant, del.icio.us offers suggestions for popular tags used by other del.icio.us users on posts and sites that have already been entered into their database. For example, del.cio.us suggests these tags for Alex Iskold’s useful post on structuring the Internet through metatagging: Blog, blogging, code, CSS, Design, development, findability, folksonomy, howto, HTML, marketing, metadata, readwriteweb, semantic, semantic_web, semantics, semanticweb, tag, tags, tips, trends, visualization, web, web3.0, webdesign, XHTML, markup, Internet, microformats.
After reviewing the ReadWriteWeb article on structuring the Internet, I looked up the New York Times metatags offered as a best practice model by ReadWriteWeb and attempted to adapt them to my own Speechless blog. Wordpress quickly eliminated my outlaw codes leaving no trace.
On my wishlist for metatags or semantic web structure would be basic webliographic information for serious content users who want to attribute ideas and content.
I read your informative, up-to-date, well-researched blog regularly, often experimenting with ideas, technology and tools that you recommend.
Iskold, Alex. 2008. “How YOU Can Make the Web More Structured.” >> ReadWriteWeb. Uploaded. January 30, 2008 10:48 PM. Accessed February 2008.
Really useful, thanks!
Great post...excellent info..I learn something new today!
thanks for your subject. it is very important for internet users.i will write your site .. please write
me back. thank you