Comments in the code below should give hints as to what the parts are doing. I relied on the Calibre manual news page, its API documentation, and hunting around the Recipes forum at Mobileread to figure this out.

from calibre.web.feeds.news import BasicNewsRecipe

class Savage_Minds(BasicNewsRecipe):
    title = u'Savage Minds'
    description = 'Notes and Queries in Anthropology - A Group Blog'
    oldest_article = 7
    max_articles_per_feed = 100

    #"cover_url" specifies a special cover image to use so it's pretty.
    cover_url = 'http://savageminds.org/wp-content/themes/SM2009Test/images/sidebar/sidebox.jpg'

    #"use_embedded_content" forces Calibre to ignore any content in the RSS
    #feed and to go get content from the articles linked in the feed.
    use_embedded_content = False

    #The next two items keep Calibre from cleaning the HTML or using
    #the blog's own stylesheets. Seems to work for me.
    auto_cleanup = False
    no_stylesheets = True

    feeds = [(u'Savage Minds Entries', u'http://savageminds.org/feed/')]

    """
    Next, "keep_only_tags" makes sure we're just grabbing the part of the
    blog page that contains the post and its comments.
    "remote_tags" cleans out some superfluous HTML that makes no sense in
    a kindle-ready ebook. This is all operating on the CSS in the website.
    I used the Firebug extension in Firefox to determine the appropriate
    element classes and ids to eliminate.
    """
    keep_only_tags    = [dict(name='div', attrs={'id':'content'})]
    remove_tags = [dict(name='div', attrs={'class':'meta clear'}),
        dict(name='div', attrs={'class':'snap_nopreview sharing robots-nocontent'}),
        dict(name='div', attrs={'id':'respond'}),
        dict(name='div', attrs={'class':'c-grav'}),
        dict(name='span', attrs={'class':'c-permalink'})
        ]

One thing Calibre can do is “fetch news”—download content from blogs and other sources and format it for your ereader. You can do this manually, or Calibre can automatically fetch it on a schedule and even automatically email it to your device. So I set it up to get a couple of blogs and news sites every day and email it to my kindle’s address so I’d have news to peruse each morning.

But there was one “first world problem.” Sure, if you use a USB cable to connect your device and then put the news on it that way, the items show their date, which is convenient for dated content. But if you have the news items automatically emailed to your Kindle, which is more convenient, the items lose the date and simply say “calibre,” which is their “author.” It’s a problem of metadata.

Then I discovered “metadata plugboards” in Calibre and figured out how to fix it so any news emailed to my Kindle shows its date, while everything else emailed shows its author. Here’s what I did:

1. click Preferences
2. click Metadata plugboards under import/export
3. next to “Add new plugboard”
1. select “any format” under Format
2. select “email” under Device
3. copy the following code and paste it into the “source template” box:

   {tags:'contains($, 'News', field('date'), field('authors'))'}

4. select “authors” under Destination field
4. click Save plugboard
5. click Close on the Preferences window

For example:

I was thinking recently about serial fiction. The BBC has done an updated adaptation of Arthur Conan Doyle’s Sherlock Holmes, with cell phones and twitter. The creators’ contention is that Sherlock isn’t defined by 19th Century ephemera like frock coats and gas lamps, but by things like the unique characterization of a detective who solves mysteries by deductive reasoning, and explains his methods and reasoning.

I also learned that House, MD is loosely based on Holmes. Certainly Holmes was a popular precursor to many other popular and recurring characters in detective fiction like Agatha Christie’s Poirot.

I found myself thinking of novels and serial fiction as such modern phenomena: television shows, characters like Poirot, science fiction “universes,” and all these things leading to the peculiar phenomenon of “fans” who follow specific characters or universes. But on reflection, people have always liked to revisit favorite characters and worlds. From Greek gods and heroes to Native American legends and stories, “experiencing a very real social connection to imaginary beings and telling story after story about them” seems, on reflection, to be a human universal.

But where does the Western mass media version originate? The creation of so much serial fiction that we can view it as a phenomenon — and product! — unto itself, with fan clubs (Hercules spawned roman cults, so I guess that’s not necessarily new…) — where does it come from?

Let’s work it backwards: authors can make a living writing serials and reusing characters and worlds because owners of presses (magazines, newspapers, TV networks) will happily pay them. There is an audience. Where does the audience come from? Increased literacy and spending power, perhaps. Maybe also Marxist alienation! But let’s get more material. Where did magazines et al. come from in the first place? Because Wikipedia suggests that serial fiction has been around almost as long as magazines.

It looks like magazines appeared out of the newspaper and gazette subscriptions in the 17th and 18th centuries. But where did they come from? It turns out the earliest regularly-published, standard-format publications were handwritten newsletters in Italy and Germany in the 16th century. Merchants and politicos were willing to pay for information to get an edge. And, I guess it’s not surprising that enterprising publishers inventing subscription busines models and the like saw a benefit to adding entertainment and art into the mix. The first magazines begain in the mid 1700s, and the first serial fiction was being published soon after.

Oh, and those early handwritten newsletters were handwritten because printing presses were too slow at the time! Oh, hello, Internet!

anyway..

One problem is the disconnect between food science and reality.

For example, at this enticing recipe for broccoli and gruyere gratin at noble pig, a commenter references a similar recipe with “half and half.”

There’s not necessarily anything wrong with this, but, as a geek, I can’t help thinking in terms of longevity and translatability of information. If I’m not living in 20th/21st century Earth, what does half & half mean to me? It’s a chemically unique solution to a very specific set of circumstances.

OK, arguably, “chicken” is a very unique solution to a specific set of circumstances, and there are a lot of different kinds of chicken (let alone chicken parts!), but, if I say “chicken” you have a sense of what I mean that relates to an animal, a growth pattern, a biological reality. If I say “half & half” you only know about a packaging representation in your grocery store. How do you make a chicken? How do you make half & half? Which one is easier to grasp?

I think it’s worth thinking about. I’d rather a recipe that references milk & cheese in various quantities than half & half.

WordPress provides a rudimentary feature to turn individual categories into tags. Entropy and mess in tags is ok, as they provide emergent information, “wisdom of the clouds.” But categories are not tags, and I have plans for them, so I want to pare them down, and not post-by-post.

Thank goodness for Batch Categories, a miraculous plugin that lets you filter by categories, tags or even keywords, and update, remove, or replace categories and tags across whole groups of posts. I’m not done yet, but it’s making this mountain climbable.