Skip to content

Latest commit

 

History

History
17 lines (17 loc) · 2.31 KB

README.md

File metadata and controls

17 lines (17 loc) · 2.31 KB

Linguistics-of-Science

Unspoken requirements for communicating in the sciences. This list of criteria has been compiled over a number of years by Onri Jay Benally on a private cloud database. The inspiration for this public repository comes from the use of a platform called Google Keep (https://keep.google.com), where ideas and thoughts are recorded.


It is similar to the use of Google Docs or Google Sheets in that you can find it on the upper right corner of the Google homepage.


• All about the use of independent sentences (phrases), analogies, adjectives, verbs that provide the adjectival functionality, and even metaphors.
• No fear of repeating the explanation in more ways than one.
• In math-related work, preferring the use of "out of" instead of "over" when talking about fractions or numerator(s)/denominator(s). Usage of "something" out of one unit of some denominator or group of denominators.
• Provision of immediate examples.
• Attributing descriptors or adjectives as pre-terms before the actual use of a word. Ex: excitatory stimulation with high KCl delivery. Another ex: inhibitory stimulation with TTX delivery.
• A new discovery tends to show its face beginning with the phrase "unless...".
• "Much less information is, however, available on..." can be used to express acknowledgement of limited information about a topic or sub-tropic. Other variations include "a limited amount of (...) solutions has been demonstrated" and "the lowest reported (...) for (...) has been demonstrated".
• When displaying images of objects, be sure to draw a scale bar using a reference of known distances near the object(s) being shown.
• Providing clarity by stating why something is important: "this" = advantageous vs. "that" = disadvangeous.
• In the quest to understand (...), what is the upper limit of (...) & what is its lower limit? Bandwidth can be used here, not neccessrily in the context of electronics, but it can be abstract.