Perhaps a "next" and "previous" link

that's a clusterfuck.

NYT and Stamen do consistently great work. Linking to these might motivate the dataviz tutorial more:

http://www.nytimes.com/interactive/2009/11/06/business/economy/unemployment-lines.html

http://stamen.com/clients/mysociety

Some other highlights:
http://worrydream.com/#!/bartwidget

https://flowcharts.llnl.gov/content/energy/energy_archive/energy_flow_2011/LLNLUSEnergy2011.png

I can pick some different ones if you want to hit a particular subject area or motivation.

<3 me some dataviz.

Would it be possible to replace the current Zlib License with a very similar, copyfree-certified license? The Zlib License has been rejected for copyfree certification because of what appears to be a bookkeeping clause:

Zlib License restriction 2 contains the requirement "Altered source versions must be plainly marked as such," which could reasonably be interpreted by the courts in a manner that violates Copyfree Standard Definition point 3. Free Modification and Derivation. Alternatively, it may be interpreted as merely requiring that a derivative or modified work be renamed, given the concluding clause, "must not be misrepresented as being the original software." Until such issues are clarified, particularly as to the intent of the license authors and the legal applicability of these terms as explained by relevant legal professionals, the Zlib License must be regarded as contrary in some respects to the requirements of the Copyfree Standard Definition.

The most similar copyfree-certified licenses appear to be the General Attribution License and the Software Attribution License. The two are almost identical, with the exception that the SAL very specifically mentions "software" and "source", while the GAL is . . . well, more general, and probably preferable overall.

There is a section in dataviz/tutorial.md that talks about *args and **kwargs without really explaining them (i.e. you can pass rotation=90 even though the method does not take a rotation argument because it accepts varkwargs, i.e. arbitrary named parameters).

We could rephrase that part, or keep the mention of **kwargs (it’s useful) and link to the Python docs (where some effort has been put recently to distinguish the parameters definition style from the argument passing style (e.g. a parameter always has a name, even if you pass it as positional argument in one call)).

Reported by @marundumuturi on twitter: https://twitter.com/MarunduMuturi/status/451641295749283840

Got to code somewhere...

There are line numbers, should just probably use <code> instead of <pre><code> or something like that. Also add numbered steps/check boxes to follow along easier.

Simply a recommendation to use/teach Celery rather than Cron for the scrape tutorial.

Specification: http://geojson.org/geojson-spec.html

GitHub announcement: https://github.com/blog/1541-geojson-rendering-improvements

Bad link -- http://www.python.org/getit/mac/
Good link -- https://www.python.org/download/mac

website/_posts/Begin/2013-09-29-Setup-your-machine.md
website/_site/begin/setup-your-machine/index.html

Currently it's just in the DataViz tutorial, then the other tutorials assume you have already git-cloned the repo. Need to move it to the generic "set up your machine" part.

Some sections seem to provide some useful extra explanation, that isn't strictly necessary. The heading suggests it could be skipped, and a separate style to set these sections aside might be a good idea to separate the primary and secondary content.

something simple, with key resources like SO etc

When I try to upload the file I get the message that only CSV or TSV are supported.

Because I completely forgot about that... /sigh.

While in the virtualenv, sudo apt-get build-dep python-matplotlib followed by pip install matplotlib worked, not just pip install matplotlib.

http://stackoverflow.com/questions/9829175/pip-install-matplotlib-error-with-virtualenv

Recommend mingw and/or visual studio express (offer both, give a quick reason why you'd choose either)

When trying to run pip install -r requirements.txtin the scrape tutorial on Mac OS X 10.9, the following error returns - and is obviously related to psycogs2:

Downloading/unpacking psycopg2==2.4.6 (from -r requirements.txt (line 5))
  Downloading psycopg2-2.4.6.tar.gz (667kB): 667kB downloaded
  Running setup.py egg_info for package psycopg2

    Error: pg_config executable not found.

    Please add the directory containing pg_config to the PATH
    or specify the full executable path with the option:

        python setup.py build_ext --pg-config /path/to/pg_config build ...

    or with the pg_config option in 'setup.cfg'.
    Complete output from command python setup.py egg_info:
    running egg_info

creating pip-egg-info/psycopg2.egg-info

writing pip-egg-info/psycopg2.egg-info/PKG-INFO

writing top-level names to pip-egg-info/psycopg2.egg-info/top_level.txt

writing dependency_links to pip-egg-info/psycopg2.egg-info/dependency_links.txt

writing manifest file 'pip-egg-info/psycopg2.egg-info/SOURCES.txt'

warning: manifest_maker: standard file '-c' not found



Error: pg_config executable not found.



Please add the directory containing pg_config to the PATH

or specify the full executable path with the option:



    python setup.py build_ext --pg-config /path/to/pg_config build ...



or with the pg_config option in 'setup.cfg'.

----------------------------------------
Cleaning up...

Note that I do have gcc installed to compile, as gcc --version returns the following:

$ gcc --version
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.9.sdk/usr/include/c++/4.2.1
Apple LLVM version 5.0 (clang-500.2.79) (based on LLVM 3.3svn)
Target: x86_64-apple-darwin13.0.0
Thread model: posix

Add directions to find the script file after pip installs it.

Mostly issues with Ubuntu, and Mac w/ more than one Python version.

re: http://newcoder.io/Part-2-Graph

Python 2.7.3 (default, Sep 26 2012, 21:53:58)
[GCC 4.7.2] on linux2

OS: Ubuntu 12.10

Packages in virtualenv: pip freeze
argparse==1.2.1
ipython==0.13.1
matplotlib==1.2.0
numpy==1.7.0
wsgiref==0.1.2

works w/o virtualenv activated - weird.

I'm at
http://newcoder.io/begin/setup-your-machine/
clicking on "Save your progress."
Which links to
http://newcoder.io/begin/Save-your-progress
which shows a 404.

:)

Traceback (most recent call last):
File "peadarapi.py", line 405, in
main()
File "peadarapi.py", line 385, in main
adjusted_price = cpi_data.get_adjusted_price(price, year)
File "peadarapi.py", line 131, in get_adjusted_price
current_cpi = self.year_cpi[current_year]

I like static/mynt because it's simple HTML from Markdown. I use it for roguelynn.com and pyladies.com

But I want this project to grow, so I'm not sure how the best to "scale" when wanting to add more tutorials, offering more communications (e.g. forums for folks to ask questions publicly), perhaps more features (e.g. a voting system on the next new tutorial).

One reason I chose mynt is that it's Python, so I want to maintain being "pythonic" in terms of the tools used to generate site content.

I'm not really wanting to go through a WYSIWYG editor for posting new tutorials/content (although I can see how something like that would simplify maintenance in fixing typos/links/etc).

I want to see other people contributing to the production of this site (as opposed to the consumption), so I'm hoping to get some public thought on this as a maintainer point of view.

If the answer is 'stick with it' I'm fine with that.

Clarify the need to clone and have the code locally.

Virtualenvwrapper is installed elsewhere when having at least more than one Python version installed (at least on Mac OS X)! so need to source elsewhere.

Width of each bar

width = 0.5

Assign data to a bar plot

plt.bar(xlocations, counter.values(), width=width)

• why GCC compiler
• why do we need a C compiler at all
• why Clang versus xcode
• why mingw versus visual studio

Virtualenvwrapper isn't prepared for windows you can use instead virtualenvwrapper-win. Also during the installation of the libraries which are based on c modules, a windows user can have problems if they doesn't have visual studio or at least its libraries. They can use instead precompiled modules from here http://www.lfd.uci.edu/~gohlke/pythonlibs/ however they won't have a virtual environment.

I want tests for both sample code on the site and installation environments. Alerts when package dependencies break (when not pinned)

When doing source /usr/local/bin/virtualenvwrapper.sh for virtualenvwrapper - sometimes it doesn't work as it depends on where the OS installs it.

I like your tutorials, but I wonder why you chose PostgreSQL for the scraping tutorial.

From what I see a simple SQLite database would be sufficient for a local scraping script. PostgreSQL is a lot more complicated to set up and maintain.

But probably you have good reason for your choice. Let me know!

PostgreSQL cannot get installed on 10.9, neither via brew install postgresql nor via http://postgresapp.com/. Hence, one cannot follow the scraping tutorial fully.

This may help a lot of folks: http://superuser.com/questions/242190/how-to-install-matplotlib-on-os-x/267204#267204

The README could have more details about where the content source files are located, where the template theme is (in case there are template fixes to do), etc.

There are some repeat steps for each setup. Let's just separate that out and then leave the setup to be the virtualenvs that are specific to each tutorial.

Including PDFs like newcoder.io/pdf, how to run/setup the workshop, etc.

On part 3 of the scraper tutorial, it says to enter psql -h localhost. I tried this and I was getting a connection refused. It turns out that you need to have postgresql running before giving that command. Maybe add a note about how to start postgres from the command line.

Hi,
I was going through setting up machine on linux, and I have one question. There is python-pip package in both fedora and ubuntu. So to get pip on those systems it is enough to run:

sudo [yum|apt-get] install python-pip

But instruction says to install pip as on mac, where there is no (deb/rpm) packaging system.

In reading through Dataviz Part 1 - Parse, I noticed there were a number of typos and other problems that have been fixed in the Github repo, but never pushed to the static site. For example, the site still has:

with open(raw_file) as opened_file:

which doesn't match the final versions of the script. (That syntax can cause problems when playing on the CLI if the user doesn't put the rest of the csv.reader code in a block under the line, as it automatically closes the file and therefore the csv stream.) There are other examples as well:

def parse(raw_file, delimiter):
    """Parses a raw CSV file to a JSON-line object"""

    # Open CSV file
    open_file = open(raw_file)

    # Read the CSV data
    csv_data = csv.reader(opened_file, delimiter=delimiter)

There's a typo, with open_file and opened_file describing the same object.

There's also no lib directory in the Github repo. I imagine in the rest of the tutorials there may be similar problems solved by pushing the fixed markdown files to your site generator.

I like them very much, BTW!

Because I want people like @merwok contributing.

source /usr/local/bin/virtualenvwrapper.sh
on arch, virtualenvwrapper.sh doesn't seem to be located in /usr/local/bin, but rather both /usr/bin and /bin