Måns,
All your suggestions are excellent if I only want to view the PPT
content within a TW context. However, what I am trying to do is create
a turnkey method of converting PPT content into TW-editable content.
Why? This goes hand-in-hand with my post on Scientific Research
(
http://groups.google.com/group/tiddlywiki/t/7ea2c7a6f43e58b0?hl=en).
I haven't finalized permission from my professor but I believe my TW
research is going to focus on the benefits of students and faculty
using TW instead of PowerPoint as a delivery mechanism for lecture
content. I hope to be able to conduct a trial with several different
professors in different academic disciplines where they will present
one lecture as they normally do and another lecture through a TW.
After the students are tested on the material from both lectures I
will survey to determine if retention was improved by the use of TW as
an editable resource.
The easier I can make it on myself to migrate content to TWs, the
broader my trial and survey will be. The window of time is just a few
weeks and I need to reduce the workload as much as possible.
Additionally, at the conclusion of my research, if my findings are as
positive as I believe they will be, it is likely some of those
professors might consider adopting this strategy more widely. For
*their* sake I need to find or craft an automated solution.
In tinkering with PowerPoint 2007 with a 26-slide sample PPT I have
determined that the PDF and MHT formats (Save As...) accurately
reproduce the look of each slide. The MHTs render poorly in Firefox
(my preferred browser) and render nicely in IE *if* I place the MHT
file in the root directory or a directory with no spaces or special
characters in the path. The Publish --> Create Handouts in Microsoft
Word option did a tolerable job of creating (too) small images of each
slide and adding speaker notes beside or below each image. If I chose
"Outline only" I get all the slide text but no images of the slides or
embedded images, no speaker notes, and no formatting or layout of the
slide text in a way that could be easily combined with the other
Publish method (small images and speaker notes). There is even a
Publish --> Publish Slides options which creates a PPTX for each slide
in the deck. I haven't experimented with these files yet but I
wouldn't be opposed to constructing a batch process against all the
PPTXs in a directory.
In tinkering with OpenOffice.org's Impress application, the same
sample PPT imports flawlessly. Export... --> File format: HTML
Document yields the same results as File --> Preview in Web Browser:
it generates an HTML file with an embedded image for each slide and a
second HTML file with the slide text for each slide, excluding the
slide titles. These render well enough, except:
* The text HTMLs don't have the slide titles. This is where I would
cut/paste the text from so the missing titles is a problem. However,
if my solution requires grepping and schlepping content amongst HTML
files, then grabbing the slide titles from the <title></title> tags
will not be an issue.
* The text HTMLs have a bug. The first paragraph of slide text is
pushed into the highest <h> tag on the page. For example, a page with
only several bullets (<ul><li></li></ul>) has the first bullet of
content pushed into an <h2></h2> or <h3></h3>. Again, if grepping and
schlepping must occur, cleaning up the content will just be another
step in the laborious process.
* The image HTMLs have a navigation bar at the top of the page for
First page, Back, Continue, Last page, Overview, and Text. The text
HTMLs don't have the same icon bar. They just have hyperlinks. Boring
and inconsistent. Possibly not relevant if I don't use the HTMLs
unmodified.
* There is no option for exporting just the embedded images within
slides, not the entire slide as an image.
I suspect my solution might involve each tiddler containing an image
of the original slide and the editable text. This will provide the
student a reference point for the original content after they have
modified the text to suit themselves.
Other possibilities
I uploaded the PPT to Google Docs. It looks fine there but the only
"Download presentation as" options are PDF, PPT, TXT. The TXT is
useless.
Moving beyond the conventional to the extreme, I have access to many
different applications, e.g. SnagIt, FAR (Find and Replace), numerous
editors for any imaginable content type, that might contribute to a
multi-step solution. I hope to avoid such complexity but here is my
plan-of-action as of this moment. I'm mostly thinking out loud here,
subject to change:
With an instructor-provided PPT I will craft a process that:
1. Imports the PPT into Impress. When Impress opens a PPT it converts
it automatically so there is actually no work for me to do here except
pass the filename and -o parameter to the Impress executable at the
command line.
2. Exports all slides to HTML specifying a destination directory and
pre-defined design.
- Future refinement: Create destination directory if it doesn't
already exist.
- Future refinement: Warn if destination directory already contains an
export.
- Future refinement: I haven't found a command line option to run the
Export process automatically so I will need to create a macro or
explore the API Project (
http://api.openoffice.org/). The solution
will need to also allow for the selection of one of several pre-
defined designs. I expect every instructor will want their tiddlers to
look "just so" so I'll likely have maintain several designs (CSS,
icons, etc.) choices. My ultimate goal is to not have to interact with
the Impress UI at all.
- Future refinement: Embed the instructor's name, course name, course
ID, and date/timestamp of the original PPT into the content (for
versioning). At the least this information should be embedded in the
tiddlers as non-editable. Integrating it into each image would be
better. Both would be best.
3. Identify the TW that will receive the content. For now that will
have to be a single TW per PPT.
- Future refinement: Option to download the latest/greatest TW from
http://tiddlywiki.com.
- Future refinement: allow multiple PPTs to import into a single TW so
the all the course content can reside in one TW.
4. Create one tiddler per slide, embedding the slide image and
wikifying the slide text. I suppose it is possible for a TW plugin to
be able to digest the HTML files (drag & drop) but I do not have the
time to develop that solution right now. I will have to do some
grepping to push content around and into tiddlers. I've done mass HTML
updates before so the learning curve will be shorter. In my
"automation" process this is the step that will bog me down the most
initially but yield the greatest ROI in time savings once the solution
works.
5. Close, test, and distribute the TW.
An alternative to steps 2-4 could involve having PowerPoint publish
PPTX files for each slide and me manipulate the content from that
source instead of the Impress HTML files.
Måns, I've gone on and on here not to weary you but to have a single
place where my brain dump can reside.
Any and all suggestions, corrections, enhancements, or any other type
of adjustments are welcome from all comers!
Thank you!
Craig
craig.p...@gmail.com