Category Archives: Uncategorized

Fixing Orange Canvas on Windows 7+

Several users have reported that Orange Canvas has ceased to function properly on their version of Windows (7 or later). The symptom can hardly be overlooked: nothing happens when trying to launch the application.

In at least some of these cases, the problem can be fixed by taking the following steps:

  1. Locate the directory where Orange Canvas has been installed. This will usually be something like C:\Python27\Lib\site-packages\Orange.
  2. From this directory, further navigate to the OrangeCanvas\application directory and open the file named (not .pyc) in a raw text editor like Notepad (or better yet, your favorite code editor).
  3. In this file, locate the function Restore() and, whithin this function (around line 723 in the current version of it), the following 3 lines:
            if not os.path.exists(self.last_scheme_dir):
                # if directory no longer exists reset the saved location.
                self.last_scheme_dir = default_dir
  4. Replace them (exactly) with the following 5 lines (including leading spaces):
                if not os.path.exists(self.last_scheme_dir):
                # if directory no longer exists reset the saved location.
                    self.last_scheme_dir = default_dir
            except UnicodeEncodeError: 
                self.last_scheme_dir = default_dir

Getting started

I’ve been thinking about starting an academic blog long enough, it’s time for me to get started.

Why would I do such a thing? In part because I’ve come across such inspiring blogs as Ted Underwood’s, Matthew Jockers’, and Jeff Atwood’s, to cite just a few, which prove that if a blog is lame, it is not necessarily due to the nature of the media (awkward as it may seem, this is actually intended to be a praise to these authors).

Also, I’m the author of an open source text analysis program called Orange Textable, whose most salient feature is its adoption of a visual programming interface. This design aims to provide non-technical users—in particular typical humanists—with (part of) the computing power normally reserved to programmers, albeit (almost) without coding, and of course within the restricted domain of text analysis.

Being the author of a software tool doesn’t make a compelling case for starting a blog per se, but the nature of Orange Textable‘s user interface is such that it can be more easily understood, I believe, through the examination of a number of diverse use cases. And since I’m frequently exploring such use cases in the course of the software’s development, I figured it might be useful to document them in passing on a blog.

I’d also like to use this channel as a means to share my views on some more general issues about Humanities computing and text analysis in particular. I hope that teaching in this field for a couple of years has given me a notion of ways in which beginners could avoid getting right from the start habits that they’ll wish to get rid of when they get more experienced. Best practices, if you will.

And who knows, maybe this turns out to be a good way to draft a book on visual programming for text analysis…