Categories
Uncategorized

Latest Blogware developments

Major improvements to Blogware’s “Help” files

Photo: The Beatles, in a promo photo for the movie 'Help!'

The Help files for Blogware are improving in leaps and bounds as I write more every day. Check them out!

  • If you’re a Blogware user, you’ll find more useful information to help you get the most out of Blogware (and more is on the way!).
  • If you’re not a Blogware user, but interested in becoming either a Blogware reseller or user, they’ll give you an idea of what Blogware is like.
  • If you have any suggestions about the Help file, whether it’s to correct an error, add something that’s missing, or clear up something that’s unclear or confusing, let me know in the comments for this entry!

I make additions to Blogware’s Help files every day, and my plan is to make it the most complete and useful help file for any blogging software out there.

If you thought you heard the sound of a gauntlet being thrown, that’s because I just threw one.

Blogware supports book, music and movie reviews!

“Big fat hairy deal,” you might say. “There’s no difference between a blog entry about what I did last night and one with my review of Cory Doctorow’s novel, Down and Out in the Magic Kingdom.”

You would be correct — and therein lies the problem. We humans can tell the difference between blog entries that are reviews and blog entries that aren’t by context; our fancy-pants brains are pretty good at that sort of thing. But our dumb computers aren’t — they’re only good as following incredibly simple instructions with incredible speed. As far as a computer is concerned, there is no difference between a blog entry about what you did last night and what you think of Cory’s book. Or the Yeah Yeah Yeahs’ album. Or that masterpiece of cinema-guano, Freddy vs. Jason. To a computer, it’s all text (actually, it’s all numbers, but let’s not get too nit-picky).

The end result of this ignorance of your computer’s part is that you end up having to searching for things on the Internet in a roundabout way. When you Google for reviews of a certain book, you’re not actually asking for reviews of that book, you’re asking for documents that contain words that you think are likely to be in reviews of that book. The outcome varies; sometimes you get lucky and get what you want, sometimes you have to sift through the results.

Since computers are terrible at classifying information by looking at its context, methods of marking data with metadata — additional information about that data — were developed. Using metadata, it’s possible to mark a blog entry with a little note for a computer’s benefit that says “this is an album review for album X by artist Y, and my opinion of this album is Z“. Once data’s been marked this way, it’s possible to build programs that can better answer questions like “What do people think of album X?”

Blogware’s reviews are the first step in this direction. Let’s consider one kind of review that Blogware provides: book reviews. In addition to the standard fields for a blog entry (such as “title” and “body”), Blogware also provides extra fields for you to add metadata about your book review. Here’s a screenshot from the entry page for a Blogware book review showing these extra fields:

Screen shot: 'Book Review' section of Blogware's 'Post Book Review' page.

Readers of the review see the usual blog entry accompanied by this little summary box that displays the extra information entered in those extra fields as show below:

Screen shot: 'Book Review' summary table.

Having a summary is like this is a convenience for human readers, but the same information is also put in a form that computers can read. This opens up all kinds of possibilities. It’s now possible to ask a computer to find reviews of a specific book. Or find what people consider to be an author’s best book. The attention span-challenged can find books under 100 pages and David Foster Wallace fans can find other authors afflicted with logorrhea. Using this metadata, it’s even possible to get fancy and find out a hot new author or when an old author “jumped the shark” by correlating reviews with publishing dates. Once a computer is given some kind of understanding of the data, all sorts of useful (and useless, but fun) data-crunching is possible.

If you’d like to read more about entering reviews, check out these Blogware Help pages:

Some geeky detail

Blogware’s review metadata is included in the RSS (2.0) feed. It uses Alf Eaton’s RVW (RSS Review) Module for RSS 2.0. If you want to see RVW in action, take a look at Boss Ross’ review of the Danny Michel album — first the human-readable version, and then the snippet of the RSS feed shown below:

<rvw:item>

      <rvw:link>

      

      </rvw:link>

      <dc:identifier>UPC:B0000BZNJB</dc:identifier>

      <dc:title>Tales from the Invisible Man</dc:title>

      <dc:creator>Danny Michel</dc:creator>

      <dc:date>2003-08-19</dc:date>

      <ent:cloud ent:href="">

          <ent:topic ent:id="Alternative Rock"

          ent:classification="music" ent:href="">

              Alternative Rock

          </ent:topic>

      </ent:cloud>

      <rvw:rating>

          <rvw:minimum>1</rvw:minimum>

          <rvw:maximum>10</rvw:maximum>

          <rvw:value>9</rvw:value>

      </rvw:rating>

  </rvw:item>

  

I’ll cover this in greater detail in a more suitable blog for this topc: The Farm.

As with the help files, questions and comments are welcome.

10 replies on “Latest Blogware developments”

Interesting feature. I like it. Is the appearance of the review information user-modifiable using CSS?

[geeky comment below – non-geeks can safely disregard]

I’ve noticed an issue with the partial-text RSS feed for your blog: It seems to truncate after a set number of characters, even though it’s HTML and not plain text. This is an issue since the HTML can (and regularly does) get truncated in the middle of a tag, or some other unfortunate place. I noticed it when my aggregator choked on your feed a couple times when the source got truncated during an open tag.

I don’t know Ruby, or I would try to write a patch, so I’ll outline my suggested fix. If the text that is getting truncated is HTML, have an HTML parser keep track of the current tag level (not in tag, in some nested tag, etc), and save the index whenever it enters a tag. Then, if you need to truncate, if it’s in the middle of a tag use the “checkpoint” index saved earlier.

Oh, also, the links to images in your entries are relative links, not absolute, and (most) aggregators don’t follow relative links. Result: no images in aggregator. Maybe Blogware could automagically replace relative links with absolute when it’s generating the RSS feed?

Ugh – Dublin Core namespace items being used where intrinsic RSS 2 tags would do?

Isn’t that what El Davissimo called “funky”?

I’m happy to see the Dublin Core being used. As an extra step I’d recommend offering a lookup service so that you don’t have to type in the full details for the book.

See: http://emmajane.net/read/ (the side bar)

or http://www.bookcrossing.com/ (adding a new book includes a search feature for previously entered books)

PS How do I retrieve my username/password?

Hi, Emma!

Perhaps there’s some way to harness the Amazon API in a way similar to the way that CD-playing software accesses the Gracenote CDDB. It would be convenient for users and reduce the number of errors (the only thing worse than no metadata is bad metadata).

If you’ve forgotten your username/password, visit http://www.blogware.com/users/index.cgi and enter either your username or email address; your username/password combo will be emailed to you.

Just out of curiosity (I know, old post, but I just came across it) any plans for Blogware to support say, like Software reviews, as well as the music, books, and movies?

Leave a Reply