08/23/10

My Anthologizd UX (Part 2)

In Part 1, I described my first experiences installing and using Anthologize.

Importing Content
The image problem I encountered in Part 1 seemed like a weird thing to have happen, because I’d heard that Anthologize would allow me to gather content from other blogs and turn it into a (e)book.  If remote images caused Anthologize to fail this function would be of limited use.

I really enjoy Nicholas Nova’s blog “Pasta and Vinegar” so I decided to use it as a test.  I entered the RSS feed into Anthologize and it brought back a list of all of Novas’ posts.  And I mean all of them.  Checked. Ready for importing.   I really didn’t want all of his posts, just a few.  A desirable feature here would be a checkbox that flipped the import status of all the posts so that you could select just a few.

After clicking “Import” I was taken back to the main My Projects screen.  I was a little befuddled about where all the imported posts had gone.  I clicked on the last project I had opened and realized that the things I imported were now included in the list of posts (including my local blog posts).  At first I’d hoped that the filters could also be applied to the imported posts (this would make selecting from a large number of imported posts easier).  Unfortunately, the tags and categories are based on the ones from my blog, not the ones imported via the Pasta and Vinegar RSS.   I can’t say this is an Anthologize problem, it looks like the Feedburner feed from P&V doesn’t include any of that information (a quick peek shows that my WP feed does – will have to try this again with a different blog to see if Anthologize recognizes it).

I was able to add the imported posts to a new project and it exported without any troubles – no image errors like I’d seen on my blog.  Although Novas’ posts do include embedded images from Flickr they use a standard img src tag, not the Flickr wrapper .  So maybe my theory above about why my images didn’t work isn’t correct, the extra code for remote embedding maybe causing the problem.  This is causing me to rethink how I post Flickr photos to my blog and consider what some of the longer-term preservation impacts may be.

To test the import feature little further, I went back and imported the feed from Sowing Culture,  IMLS DCC’s blog which does include categories in the feed but they were not recognized by Anthologize.   Even though the image references did point back to Sowing Culture,  the PDF export went off with out a hitch.   On examining the PDF,  I found that the pages were breaking in the wrong place (after the post title, which left the titles dangling on the preceding page). I went back and looked at the other PDFs and found they exhibit the same behavior.   Finding page breaks in a straight run of HTML is a tough nut,  but since these are breaks across different Anthologize library items, I would think it would be possible.  I would only expect the posts to run as a complete text if I’d “Appended” one post to another.

Fortunately I’m not just stuck with the default PDF output,  Anthologize also allows you to export your project in ePub, TEI (with HTML) and RTF.  In Part 3, I’ll take a closer look at each of these to see what’s going on under the hood.

08/21/10

My Anthologize UX (Part I)

Since I wrote last about Anthologize I thought I’d do some reflection on my installation & initial use of the tool. Rather than be a grumbly mumbly, I really need to start making some contributions to these kinds of project.  Plus,  I’ll be meeting with a colleague on Monday who is also interested in Anthologize.  I thought it would be helpful for me to have run through the installation first.

What follows is a kind of “think-out-loud” exercise about my first experience with Anthologize.  I intentionally did not read any of the documentation before trying Anthologize out – I wanted to see how far I could get just using it just based on the interface.

Installation

Although I already downloaded the Anthologize plug-in, there had been a lot of traffic on the development list.  The dev team had already released an update to fix some of the initial bugs.
Anthologize installs like any other WordPress plugin, so I added it to good ole Inherent Vice here.   I was a little wary about this, because I wasn’t quite sure how Anthologize would affect my regular blogging and plugins, etc.  But since it was just a plugin, it seemed worth trying it (I’d backed up my database earlier in the evening before upgrading my WordPress installation to the required 3.0 version).  Once activated, Anthologize added a new panel to my administrative dashboard.

Creating a Project

Under My Projects I created a new project, adding a title and my name as an author.  I thought I’d use Anthologize to gather all the posts I’d made about my Collections Dashboard research.

After adding a project, I was presented with the list of “My Projects” which now included the Collection Dashboard Diary that I’d just created.  It wasn’t quite clear what to do next, but as I hovered over the title of my project a sub-menu appeared. Project Details took me back to the screen I’d just completed with title and author information, nothing new here but nice to know you could change these after creating the project.  I went back to My Projects and hovered over the project title again, selecting “Manage Parts.”

Manage Parts gave me options for Items and Parts.  Items included a filter that would change which posts from my blog were displayed and a list of titles of previous posts.  I assumed I would be able to click on the posts, but nothing happened.  Apparently you need to first create a “Part.” After creating a part it was possible to drag posts from the Items column to the Parts column.  It might be helpful to put something here that suggests dragging. This was a familiar pattern to me, but I don’t know if the interface makes it obvious that dragging is what was required.

Once  posts were added to the Items menu, an additional set of links appeared. I clicked on the Edit link and was taken to the Anthologize edit screen.  Here I was able to make changes to the original post – or so I hoped.  I was a little unsure whether Anthologize would change the original post on the blog,  or whether I was editing a copy of the post.  The title of the WordPress page was “Editing Anthologize Library Item” -  I took this as assurance that I wasn’t editing the original post.  I added some text and saved it,  but I went back to make sure nothing changed in the original post.   Nope, Anthologize had made a copy of the post for me to edit.

The other option was to “Append” items.  I presume that this would combine two blog posts into one Anthologize item – but I didn’t have anything I wanted to combine yet.

Exporting a Project

The only thing left to do was Export my project.   Anthologize walks you though a several step process that lets you select a Creative Commons license (a very nice touch to have this built in) and add a dedication and acknowledgements section.  Lastly you can choose from several different publication formats, page size (8.5×11 or A4), font size and type.

Here’s where I ran into my first problem.  On clicking export I received the following error:

TCPDF ERROR: [Image] Unable to get image: /tmp/img_sm3Fuu

I went back and looked at the posts.  None seemed to have an embedded image with a file path like that.  I did have a Flickr badge in the middle of one post, so I removed it, thinking it could be causing a problem.  No luck.  I looked in the user forum but didn’t see an obvious solution.  There was some conversation on the development group about a similiar problem, but it didn’t seem to have anything to do with the posts I was trying to publish.

I then re-exported the project as a TEI (with HTML) file which went without a hitch.  On reviewing the TEI,  I realized that the posts I was trying to publish had been written on a related blog for the IMLS DCC project and the embedded images actually lived on that blog. The error that had been reported in the dev forums had to do with remote images. Once I’d removed all of the image sources,  the PDF was generated without a problem.

I created a new project and included several posts that I know used local images – these also exported without a problem.   Since I often use the Blog This feature for Flickr to publish posts to Inherent Vice, not being able to include remote images could be a problem for me.  I presume that images located remotely on Flickr will cause the PDF export to break.

This was my first impression of the Unable to Get Image problem.  In Part 2, I’ll explore what’s going on here a little more closely, import some content from another blog.  In Part 3 I’ll take a closer look at the default TEI markup.