(Replying to PARENT post)

I had three fun ones.

I wanted to learn powershell, and I was interested in the history of worms on the internet, so I made a script to check https://isc.sans.edu/ and send me an email if the status was not green. Turns out, powershell has an all-in-one function to send an SMTP email, which I thought was hilarious. I used windows task scheduler to run it every hour.

Second, I was struggling with my Cable internet connection, and it was incredibly difficult to nail down what was causing my slow speeds. To help with the diagnostics, I wrote a little python script to download the files that Netflix uses for speedtests (check Fast.com) and logged the average of 10 runs to a file. It similarly used task scheduler to run every 10 minutes and as a result I now have a big old dataset of SpeedTests and unix timestamps which I graphed while playing around with Pillow

I wrote a python bot to Read project Gutenburg books to twitter. This was to participate in a contest that involved seeing which "team" could get a hashtag into more tweets. A twitter bot takes about five minutes to make, as I found a brilliant little Python-Twitter library. Unfortunately, my teammates got cold feet and convinced me to de-rate the bot's posting frequency, meaning we barely lost to a group of 60 or so sorority girls doing it the "correct" way. However, as we did it sort of below-radar, the loophole of automating it is still open and valid.

Currently I'm working on automating the process of creating photoreal scenery for Microsoft Flight Simulator. The FSX scene has built tools to turn open source land-use data into ground scenery like houses and forests, and Microsoft included the SDK and Documentation as part of the game[1], but since high resolution photo-real satellite photography is an incredibly expensive asset, nobody has automated that part of the process. I most likely will not be successful, as I have no source for the photo-real imagery.

Personally, I find that absurd, as my tax dollars as an American likely pay for incredibly high resolution (higher than 1 foot per pixel) satellite imagery of the entire world, and I'm not sure how the NSA providing that to citizens would hurt national security. The Stallman in me feels that I've already paid for that data, and it should be public domain. Hell, even a year old would work fine. The biggest problem would be Google and Microsoft lobbying to prevent the devaluation of their expensive assets.

[1] https://msdn.microsoft.com/en-us/library/ff798293.aspx

๐Ÿ‘คmrguyorama๐Ÿ•‘8y๐Ÿ”ผ0๐Ÿ—จ๏ธ0

(Replying to PARENT post)

> I now have a big old dataset of SpeedTests and unix timestamps which I graphed

This peaked my interest, as my connection is also crappy. Did you actually manage to make improvements with this data though?

๐Ÿ‘คhacker_9๐Ÿ•‘8y๐Ÿ”ผ0๐Ÿ—จ๏ธ0

(Replying to PARENT post)

Look at scraping Google and Bing maps?
๐Ÿ‘คshubb๐Ÿ•‘8y๐Ÿ”ผ0๐Ÿ—จ๏ธ0