search

jsoma / data-studio-projects

12 stars 18 forks source link

The Ultimate Ultimate Guitar Story #233

Open kevinlitman-navarro opened 6 years ago

kevinlitman-navarro commented 6 years ago

Pitch

Summary

So there is the website called Ultimate Guitar that every single person who is learning songs on guitar (or piano, or uke, or other things) uses for tabs and chord charts. I decided to scrape some metadata on the tabs on this website to see what I could find out. My question is along the lines of -- what is the best guitar song?

But... I didn't feel like scraping all 1,200,000 tabs (I think I could tho), so I looked at their list of the 100 top rated (which factors in hits as well) and then some bands that I was interested in.

Details

Possible headline(s): People Really Want To Woo Like Ed Sheeran

Data set(s): Scraped ultimate guitar

Code repository: https://github.com/kevinlitman-navarro/DataStudio/tree/master/Code/04_guitartabs

Possible problems/fears/questions: man i can't get away from bar charts, because I keep choosing categorical data and i'm starting to feel like a loser

Work so far

A note about the color scheme -- I thought it would be fun to match ultimate guitar's site, even if that isn't normally a look I would go for.

Here are some the things I made:

this is for the top 100 how-many-tabs-top-100 popular-keys

The rest are for my sample of bands, which includes 23,000 tabs how-many-tabs whos-tabs-are-best

Checklist

This checklist must be completed before you submit your draft.

Palarisk commented 6 years ago

I really love your "Popular keys in popular tabs " chart. I'm not usually huge fan of these area charts (too imprecise) but here it works fantastically, the subject matter is light enough. And the colors are great.

If you're going to keep the third chart, one suggestion: please explain briefly how there can be over 4,500 tabs for Beatles (and thousands for other bands) when they didn't record that many songs. I guess they are just different version, but would love to have the reason in your chart/text.

adrianblanco commented 6 years ago

Cool!

Some suggestions:

kevinlitman-navarro commented 6 years ago

Update

Your project content: images/words/etc

Made some design tweaks, and charted some timeseries that show another way of tracking interest in these bands. Not sure how interesting this is to other people, but i'm enjoying using tabs as a metric for popularity among a certain group of people.

For my next draft, I'm going to add annotations to the timeseries pointing out when albums were released and other milestones that explain the change in interest.

Adrian raises a good point. I might try to do ai to html this time, in which case I will follow his alignment advice, but if I don't have time to make that work, I will just do my titles on the website and they will be properly aligned.

Lots of charts incoming, ignore at will

beatles-tabs-timeseries blink-tabs-timeseries how-many-tabs-revision-1 how-many-tabs-top-100-revision-1 most-tabbed-beatles-revision-1 mychem-tabs-timeseries popular-keys-revision-1 sheeran-tabs-timeseries tayswift-tabs-timeseries whos-tabs-are-best-revision-1

Any changes in direction or topic?

Not really

Problems/Questions

Missing data is a bummer. Would love to look at more keys, like I did with the top 100, but the rest aren't tagged as well.

Checklist

sarahslo commented 6 years ago

to make these fever line charts comparable, the uploads, they all need to be on the same scale. right now you are skewing the data with all these different scales. that's an unfair representation. whenever you put your hand in the data that way you have an impact of how the trends appear.

put them all on one scale in the same chart and change the color of the fever line to match the artist. i'd really like to see how these all compare to each other!

kevinlitman-navarro commented 6 years ago

Update

Your project content: images/words/etc

blink-tabs-timeseries-revision-3

Spent time in illustrator trying out different design iterations on my uploads charts. Didn't get to Sarah's advice yet, but I'll make on master chart for the final draft.

Any changes in direction or topic?

Nope

Problems/Questions

Designing stuff takes forever, I went through like five different styles before settling on this one and was it worth it? Who knows

Checklist

kevinlitman-navarro commented 6 years ago

Final

Project visuals/text

Here as some screenshots, bc I used ai2html to put everything on my page, which is also why my webpage might be a lil wacky

screen shot 2018-08-10 at 10 27 33 pm screen shot 2018-08-10 at 10 27 41 pm screen shot 2018-08-10 at 10 27 50 pm screen shot 2018-08-10 at 10 28 04 pm screen shot 2018-08-10 at 10 28 12 pm screen shot 2018-08-10 at 10 28 17 pm

Details

Headline: The Ultimate Ultimate Guitar Story

Published website version: this guy

Code repository: this guy

Final data set(s): this guy, scraped from the website

What did you find to be the most difficult part of this project?

Spent alot of time on minute design decisions, not sure if I got them right or what

Are you satisfied with what you produced? Is there anything you would like to change or improve?

Satisfied-ish. I didn't really answer the question I set out to - something along the lines of what is the most quintessential guitar song - but I would need to scrape way more to do that. Also, I was only able to get the hits for the Beatles, not sure why my scraper wasn't working but could fix that with more time.

Checklist