Is it just me or is the recent surge of DataNitro scripts just proof of concepts? I don't really want to do get Facebook Likes in Excel or play Blackjack. They added a Python handler in Excel to do this, it isn't native.
My guess is advertising. HN is their exact demographic / target users, and if you check the front page of "classic" HN where only old accounts can vote:
http://news.ycombinator.org/classic
It's gone. Someone gamed this to the front page
For me, it's still "proof of concept" right now. It's interesting because Python is much easier to use than VBA. Where I work, we use a lot of Excel and data from various sources. The ability to leverage Python code can be a big deal for me.
It's been possible to script Excel (and the rest of Office) using Python for probably 15 years now. I'm pretty sure the COM package is included with all the Win32 Python builds, but if not it's just a package install.
It's a bit disappointing that so many people don't learn about the tools they're using.
Competitor analysis with simple Excel tools. Hm.. I like that ! I wonder if you could build a whole Excel application for social media tracking. There are companies out there that are very successful in this space. Perhaps with your superior technology you could outdo them!
The script is not getting 'likes,' it's pulling the number of 'shares' for a particular URL.
If you want to get to a brand's like-count, you'll need to feed the Graph API with either the Facebook Username of the account (e.g. "nike", not "http://www.nike.com "), or with the Facebook ID# of the brand (e.g. "15087023444").
This can be misleading though. This script does get the count you will see on a Facebook 'Like' button which is actually the sum of the 'like','share', and 'comment' counts.
For example, the script gets a count of 1,257,333 for www.apple.com:
So whilst maybe not 100% accurate in some regards, it is certainly what many people would refer to as the 'like' count (and they'd be confused if you gave a lower number which isn't what the like button on the page showed). This tool is aimed more at URLs than brands.
My question is, why is this interesting?