Skip to content

Latest commit

 

History

History
38 lines (24 loc) · 1.27 KB

README.md

File metadata and controls

38 lines (24 loc) · 1.27 KB

BIP-😸

BIP😸 is an informational BIP that outlines a general guideline for encoding Bitcoin addresses as strings of emoji. It includes example code for working with emoji addresses.

OUTLINE OF WORK...

  1. Grab all images from here: https://unicode.org/emoji/charts/full-emoji-list.html

  2. Parse the table using this: https://stackoverflow.com/questions/6325216/parse-html-table-to-python-list

  3. Save each image as a PNG with a name like: cldr_short_name-platform.png

Note:

00:47 < Terminus> faceface: echo
"R0lGODlhEAAMAKIFAF5LAP/zxAAAANyuAP/gaP///wAAAAAAACH5BAEAAAUALAAAAAAQAAwAAAMsWLPcrSHKOVcgOOtsiRCbBxLdF5rkcIVhx27uy6kagNlafOM5LXMKh3BRSAAAOw=="
| base64 -d > ~/foo.gif 00:47 < Terminus> it's a gif btw so the mimetype is wrong.

  1. Apply random transformations using Keras: https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html

  2. Try to classify cldr regardless of platform...

Note: Find the list of cldr that are most accurately classified across platforms (e.g. those that are most similar / least confused across platforms).

Note: Classification or OCR?

  1. Create a list of 58 robustly classified emoji across platforms.

  2. Map them to base 58 characters... Somehow.

  3. Create libraries to work with this encoding.