PizzagateBot v2.0: Updates and Feedbacks ( pizzagate )

submitted by PizzagateBot

Hey folks, just wanted to make a post to update a couple things you will see from PizzagateBot and get some feedbacks since its been active for 30 days now and I have some free time this week.

Improvements:

1.) Added function to verify that page found is actually valid by opening the page and finding the original link. Sometimes I would get results that were searched properly quoted but got ignored so you may of clicked on a post link and were not able to find the link or how it was related.

2.) Added function to get the post date and include in table column PostDate.

3.) Added function to determine if the link I found is from the OP or if it is from a comment, and, if from comment, post a link to it under table column LinkOrigin.

Here is an example for your reference - https://voat.co/v/pizzagate/1755477/8598437

Feedbacks:

Appreciate any feedbacks, but specifically looking to answer these questions...

1.) I can make pizzagatebot do anything you can do with a web browser. Is there anything PG research related that you find yourself doing repetitively that you'd like to see automated?

2.) Is there anything you would consider broken at the moment?

3.) Anything you don't like in particular about the results? (if you find a bad result, please reply to the comment instead of just downvoating, I will look into it)

4.) What search engine do you use, besides Google, to research PG? I'm going to do side-by-side comparison next.

Thanks and cheers!

PizzagateBot

It's a good idea. I checked a while ago and you need to register a phone number to your account in order to get an API key. I need to buy a burner phone first.

NoBS

As a tech-tard I forgot it takes a dedicated human to make bots play nice. Hat tip Mr Bot Master.

jangles

I noticed a great change THIS IS AWESOME

privatepizza

Love PizzagateBot. Thanks for all your work on this, it's fantastic.

anonOpenPress

1) Automated archiving & automated webpage screenshot

PizzagateBot

:)

I'm about 10k posts into a full archive of some 23k posts that I discovered by crawling ~400k possible ID numbers and checking if they are posts or not. I'll upload it somewhere once done, it saves css in each WARC so total itll be bloated like 8GB, and maybe seed a torrent of it.

I am using the WARC proxy that webrecorder.io provides because it is the first proxy that archives everything including javascript so that you can click on and expand comments. I tried others who claim the same thing but never worked. Also, I like that webrecorder.io allows you to download the WARC so I could include a WARC link in each post and if you thought that post was interesting or worth saving, you could just click and download the WARC.

I use webarchiveplayer - https://github.com/ikreymer/webarchiveplayer - to replay WARC files like this - https://files.catbox.moe/vgd5z5.png and I can combine multiple WARC into a single index so that it is searchable by link like this - https://files.catbox.moe/g8tiqd.png so to open a browser of the index of WARCs that I merged for testing, I just ran this command

webarchiveplayer WARCMerge20170329215008930044.warc

crystalclear

Love the pizzagatebot. I use duck duck go for research instead of Google most times. If there was a way to open up a page with a search or a news search that would be cool but I don't even know if I'm accurately describing what i'd like.

ThePuppetShow

You rock!

Can you have it search out additional social media accounts if someone posts a social media account in the OP? Or would it be too hard to verify you have the same person?

Edit: Thinking about it, that sounds like an impossible task. How about searching out social media accounts from an email address?

PizzagateBot

How about searching out social media accounts from an email address?

I don't know anything about social media. I've been on the front lines of IT for the past decade and I still have never registered a myspace/facebook/twitter/etc. I'm old fashioned and wan't nothing to do with that mess so I'm ignorant to the inner workings.

Cursory google search shows there this a service like this https://www.social-searcher.com/ and http://com.lullar.com/ but the bot would need to recognize names in text, I have tested that with NLTK to find proper nouns in text, which works pretty good, but I'm wondering how often does a new suspect come up that you need it to be automated.

If you use those search engines and end up coming with useful results manually, show me what you found and I can think about it.

Cheers!

ThePuppetShow

You're right, that would have been a good option 3 months ago. It's a little slow now with new names.

NoBS

This bot is a sex toy for the mind and soul of my humanity.