What types of files can Flesh process?
Plain Text files (.txt)
Microsoft Word documents (.doc)
Rich Text files (.rtf)
Adobe Acrobat files (.pdf)
Some Adobe Acrobat files are protected or have the data stored in difficult to access ways, and thus cannot be processed. However, most should work properly.
HTML files (Mac OS X only)
Mac OS X users can also process HTML files. HTML processing support Windows and Linux users will be added in a future release.
How do I process a file?
Launch Flesh the way you normally open an application; usually this means double-clicking on the Flesh icon.
Next, click the 'Choose...' button, and select a file of the types listed above. Alternatively, simply drag a file on top of the open Flesh window.
Finally, click Process.
How does Flesh work?
Flesh works by analyzing the text in your documents, looking for words and sentences. When Flesh finds what it believes to be a word, it then calculates the numbers of syllables in that word. Flesh then uses the total words, syllables and sentences in your document to determine how difficult your document is to understand.
This sounds relatively simple, but it gets tricky fast. It is easy for a human to look at a document and tell what is a word, how many syllables it has, and where sentences start and stop. We know that the title of a document, or a section heading isn't part of the next sentence in that document, that when we see 'Ms. Smith' the '.' after 'Ms' does not indicate the end of a sentence, and that 'VII' is the number 7 and not a word that sounds like 'v-aye'. Getting a computer program to make these and may other determinations quickly and accurately takes a bit more work.
If you really want to dig deep into how Flesh works, feel free to go browse through the code. Check the 'CVS' link at http://sourceforge.net/projects/flesh/.Also, check out the other projects that helped make some of the improvements in Flesh 2.0 possible: Jakarta POI (used by Flesh to read Microsoft Word files on the Windows and Linux versions) and PDF Box (used by Flesh to read PDF files on all versions).
What do the results mean?
Flesch-Kincaid Grade Level
The Flesch-Kincaid Grade Level is an index that gives the years of education required to comprehend a document. For example, a document with a Flesch-Kincaid Grade Level score of 10 would require a reader have about 10 years (or a 10th grade level) of education to comprehend the document. The Flesch-Kincaid Grade Level is calculated using the number of words, sentences and syllables in a document, using the following equation:
(0.39 * Average Sentence Length) + (11.8 * Average Syllables per Word) - 15.59
Flesh Reading Ease Score
The Flesch Reading Ease Score indicates on a scale of 0 to 100 the difficulty of comprehending a document. A score of 0 indicates an very complex document, while a score of 100 indicates a very simple document. A Flesch Reading Ease score in the range of 40-50 would correspond to a relatively complex document that might score a 12 as its Flesch-Kincaid Grade Level. The Flesch Reading Ease Score can be calculated by using the following equation:
206.835 - (1.015 * Average Sentence Length) - (84.6 * Average Syllables per Word)
Like all readability indices, the Flesch-Kincaid Grade Level and Flesch Reading Ease Score provide estimations, and are only meant to be used as such.
What are the options in Flesh?
End sentences with semi-colons (enabled by default)
Normally, Flesh records a sentence when it finds a period, exclamation mark question mark or semi-colon. If you would prefer, Flesh can skip right over semi-colons and only record sentences as ending with periods, exclamation marks and question marks. With this option disabled, groupings of words with semi-colons would then be counted as one long sentence instead of two.
End sentences with colons (disabled by default)
Flesh can also consider colons to end setences (just like periods, exclimation marks and question marks). Normally colons do not end sentences, so this is disabled by default.
Count titles and section headings as parts of sentences (disabled by default)
Normally, when Flesh sees the title of a document, a list of section headings (like a table of contents) it ignores them, as this can make your document appear more difficult than it actually is to comprehend. Leaving this option disabled will nearly always give you more accurate results.
When will Flesh be updated?
I wish Flesh would (insert your idea here) . . .
Feedback is great, so if you have any ideas for future development, or would like to contribute to the project, feel free to email them to firstname.lastname@example.org.
- Back to Flesh home page.
- All platforms (4.6 MB)
All versions require Java 1.5+.