Hacker News new | past | comments | ask | show | jobs | submit login

On my own blog, whenever I do a data science writeup, I make sure to do it on a topic that hasn't been done before, with explicit detail and open-sourced code.

...this has had the consequence of getting no pageviews since it removes the mystique of data science. :\




I just had a look at your NCAA visualisation. It was great, and just like your description, not something everyone would search for. So what?

It's super interesting, BECAUSE it's the first time I'm seeing those stats in the form of a 'heatmap'.

You're optimising for quality. And I just happen to think that's right.

Cookie cutter data intro's that walk you through installing anaconda....& and end up with some plt.plot(x,y) might gather some views. But they don't add anything to the ambient noise in the room.

They only encourage the (I'm assuming beginner) blogger to publicise their future work, as they've now experienced 'blogging'.

That's the only good outcome I see.

Perhaps I'm wrong.


Thanks for the compliment on that post! It's true I optimize for quality, but pageviews are good too!

Ideally I'd optimize for a weighted average between the two, but it's tricky without playing a bit dirty.


btw the idea is to share the code and the details bu into chunks instead. I found this more helpful and less overwhelming for newcomers




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: