Review repo's home page.
Github is how you should get and keep the course files up to date. Make an account, if you haven't.
I recommend you clone my repo, so you can repull it as I add stuff. You will do git clone [copied path]
(https is fine for now). I'll show you in class.
Every week you should do git pull
to get my updates. Do this from inside the directory made by cloning. If you want to make changes to class files, save them with new names or you will have merge conflicts. No one survives that.
You'll be doing homework in gists. This makes it easier for me to review and check it and find issues. Also, the bl.ocks.org site makes things easy for visual display of gists. See: http://bl.ocks.org/.
Make a directory for your files for a gist NOT INSIDE THE COURSE FILES REPO DIRECTORY and upload those by hand from the web page, or using the command line tool gistup. If you use the command line, make a directory for each gist, NOT INSIDE THE CLASS REPO YOU CLONED. (Your gists can't be inside another repo.)
Resources and tips on using gists:
You can see all your gists by looking here - where you fill in your own user name where mine is: https://gist.github.com/arnicas. If you want to edit or delete it, click on the header for one from the list.
Important Tips:
<script>
tag, you want to reference: <script type="text/javascript" src="//cdnjs.cloudflare.com/ajax/libs/d3/3.5.6/d3.min.js"></script>
instead of a local path. A CDN is an online reference to a library file.Demo making a gist for a CSV file...
You need one. Otherwise Javascript code won't execute on your pages. So
set up your servers. If you haven't got one running (WAMP, MAMP, python, or otherwise), stick around and we'll set it up at the end. Use this to help: https://github.com/arnicas/d3-faq#running-a-server.
I expect everyone to be proficient with HTML and CSS and to know at least a bit of Javascript. If you don't, this class will be very hard. If you're already great at JS, I'll try to keep you entertained anyway.
Let's play with console_javascript.html now.
Resources:
A good place to do an initial formatting of data for use in web apps is Excel (or Google Sheets). We often want to use csv files as data sources.
We'll go through a download and csv formatting example in class. You'll need to do the same, so take notes! Let's look at the infant_deaths data set from WHO first. (Walkthrough how I formatted it for use in javascript.)
Things to remember about data in CSV files for use in the browser:
Read this for some tips: http://learnjsdata.com/read_data.html.
Comment from Scott Murray in the Knight D3 online class:
I strongly recommend simplifying column titles so they do not include spaces.
Why? JavaScript object notation is simpler without spaces. Let me give you an example.
Say you define an object like so:
var object = {
Country: “USA”,
“Number of Bananas per Capita (millions)”: “556”,
”Total Length of Peels if Laid End to End (kilometers)”: “412415123”
}
To get at the value “556”, you would need to type:
object[“Number of Bananas per Capita (millions)”]
However, if you excluded spaces from your column titles, so your object
looked like this:
var object = {
country: “USA”,
numBananas: “556”,
peelLength: “412415123”
}
…well first of all, that is so much easier to read, but also then you can
use simplified object notation:
object.numBananas //returns “556”
See, you don’t need the quotation marks or the brackets. So eliminating
spaces just saves you a lot of typing and headache later.
You can read this for more info on reading data in D3: http://learnjsdata.com/read_data.html.
A next step for data work is to do exploration with graphs in Excel or a tool like Tableau. We'll go over those in a couple weeks.
Read in data with D3 and verify it's there in the console: d3_load_csv_json.html. JSON objects (javascript object notation) should be there for each row of your data set.
The contents of each data set are printed to the screen using a simple jquery text operation and "JSON.stringify()."
Homework is due at 5pm on Monday the night before the next class. I have office hours on Monday if you have struggled!
Readings:
Read Chapter 3 of IDVW: http://chimera.labs.oreilly.com/books/1230000000345/ch02.html. You can stop at the SVG part at the end of chapter 3, for now. (Chapter 2 may be nice background but is optional.)
Loading data with D3, at http://learnjsdata.com/read_data.html.
Homework: Fix js_errors.html so it displays correctly. Send me the edited file, plus a screen cap of your console with the results of all the stages correctly showing. Subject line: "Week 1, JS errors". If you can't do it all, send me as far as you got. (20pt)
Homework: Download a data set of interest to you and format in Excel or Google sheets for use in visualization. Follow the advice above. Keep a copy of the Excel file(s) for yourself, but save out a CSV file that can be loaded into javascript. Make a gist in which you load the CSV file and print it to the screen the way I did in d3_load_csv_json.html. (You don't need to do a JSON version.) Make a gist with a page called index.html(!) and your data files send me the link: 'Week 1: CSV data.' You'll need this file next week. (12pt)
Homework: Set up your server, github accounts, etc!