gparuthi

Link for xml containing country codes and regions

http://opencirce.org/data/country-DAC-UN.xml

gparuthi

{
  "countries": {
    "country": [
      {
        "ISO3166": "AF",
        "name": "Afghanistan",
        "UNRegionCode": "034",
        "UNRegionName": "Southern Asia",
        "DACRegionCode": "689",
        "DACRegionName": "South & Central Asia"

gparuthi

##Install Python

http://www.python.org/getit/

##Install sqlite3

https://sqlite.org/download.html

gparuthi

[2014/2/21] Works with MOTOX:
Name: cc
APN: ccdata
Proxy: [Not set]
Port: [Not set]
Username: [Not set]
Password: [Not set]
Server: http://mmsc.mobile.att.net
MMSC: http://mmsc.mobile.att.net
MMS proxy: 66.209.11.32

gparuthi

##Peer to Peer everything I was thinking about a peer to peer social network earlier. Well, I was actually thinking about peer to peer everything where agents work on your designated machines and work on your behalf to provide- home automation, quantified self, etc. Goal is to ensure privacy, security, etc. You could even sell your data directly instead of letting facebook or google sell your data.

I came across Huggin. Which uses agents to act on your behalf.

##Social network

The first step would be to allow users to use a storage of their choice (local, cloud-storage, etc).

I was digging into this and realized that this would be similar to the Higgins Project. I was wondering if there is progress on this.

gparuthi

# Publish progress to redis 
# publish the status of a long running parallel from ipy cluster process to a redis server
# a webservice can then read this and feed it to a javascript visualization

def pub_to_redis(pres, title):
    import Redis
    r = redis.StrictRedis(host='localhost', port=6379, db=0)
    n = len(pres)
    p = ProgressBar.ProgressBar(n)
    while True:

gparuthi

in upload handler
in file close
..
----------------------------------------------------------------------
Ran 2 tests in 0.021s

OK

gparuthi

from IPython.html import widgets # Loads the Widget framework.
from IPython.core.magics.namespace import NamespaceMagics

def _fill():
    v = _var_name.value
    try:
        ve = eval(v)
        vtype = type(ve).__name__
        _modal_body_label.value = str(ve)[:20]
        if vtype == 'DataFrame':

gparuthi

# This code publishes HITS in same batch on Amazon Mechanical Turk
import mturk  # check out https://github.com/ctrlcctrlv/mturk-python
conf = {
"use_sandbox" : True,
"stdout_log" : False,
"verify_mturk_ssl" : True,
"aws_key" : "your_key",
"aws_secret_key" : "your_secret"
}

gparuthi

Today, I was imagining Ipython notebooks in a different way.
I was thinking about a non linear way of writing code. The current linear way means we write a code cell and then execute and then write another. If I would like to connect the output of one cell to another, I have to do it manually or create a function.
I wish we could set a cell of code as a code block, that can be easily be found by other notebooks. 
So I dont have to copy paste the cell or create a module for the cell.
I wonder if we could also share code cell in an online repository. One starting point would be to make cells out of stackoverflow code blocks.

Stackoverflow search and import inside the notebooks! Thats a good idea.