Hiya folks!

This is going to be a much smaller and quicker article for y’all.

Have you ever wondered how to make and utilize better social media links? Like having control over the preview box that pops up when you share something on Facebook, Twitter, or LinkedIn?

That’s Facebook’s Open Graph protocol which you can see explained here.

Now, you may be wondering what this has to do with data science since I normally focus on that! Well, one thing you should keep in mind as a data scientist is your end user. No matter the visualization or model result, you have to make it easy for the “customer”. Business-minded folk may not worry too much about how many iterations of the model you needed as long as you delivered results on schedule, and software engineers or other data scientists may struggle with your model if it can’t be validated or deployed easily.

So if you’re making or have made a full stack project, you want it to present well and be a pleasant experience for the user. And this is where I failed.

I did not understand how Open Graph worked, and MeaLeon (my recipe recommender powered by machine learning) relied entirely upon something outside my control whenever I shared a link to it. I noticed this because my LinkedIn links only showed broken thumbnails, and it bugged me that I couldn’t figure out why it wasn’t working.

When I did initial searches to figure out why, people mentioned Open Graph tags. However, posts I found about it seemed to neglect one small part mentioned on Facebook’s documentation: you NEED at least 4 tags to get it to work! I had only been putting one (og:image) and it seemingly resulted in the tag not being used.

#full-stack #quarantine #diy #data analysis