post about blog itself - Githubissues

stharrold commented 8 years ago

[ ] Pixel settings for devices: http://mydevice.io/devices/
[ ] Use device pixel settings for image width, e.g. width="320"
[ ] Test appearance for different resolutions: https://chrome.google.com/webstore/detail/resolution-test/idhfcdbheobinplaamokffboaccidbal

stharrold commented 8 years ago

stharrold commented 8 years ago

stharrold commented 8 years ago

[ ] structure of dsdemos, testing, tagging, versioning (semantic versions?)
[ ] removed liquid_tags for notebook formatting, also collapse cells does work with pelican-bootstrap3 dependencies
[ ] workflow with git-flow: (feature, start, pelicanconf.py, finish), (release, start, pelicanconf.py, publishconf.py, finish, ghp-import), (hotfix, start, status published, pelicanconf.py, publishconf.py, ghp-import)
[ ] git-flow helps act as a safety measure and prevents merges with src_master that haven't run pelican --settings publishconf.py, otherwise the merge would delete the atom/rss feeds.
[ ] i save terminal output (from cloud9 with caret) for future reference. provide examples of each workflow.
[ ] dates on pages are deadlines then replaced with tag when the page went live. modification dates are tags.
[ ] protected branches src_develop, src_master, master
[ ] how to remove and rename git branches local and remote
[ ] Book "Guide to LaTeX" for math
[ ] pelican-plugin render_math
[ ] nohup vs disown for python -m http.server (http://unix.stackexchange.com/questions/3886/difference-between-nohup-disown-and)
[ ] configure site searchability with google search console
[ ] add robots.txt with sitemaps: https://github.com/getpelican/pelican/wiki/Tips-n-Tricks, https://support.google.com/webmasters/answer/183668?hl=en
robots.txt:

Sitemap: https://stharrold.github.io/feeds/all.atom.xml

stharrold commented 8 years ago

[ ] have lots of static content with exported ipynbs etc so put all under stharrold.github.io/content/static
[ ] i use timestamps for blog tags since i'm not maintaining a backwards compatible api. compare to dsdemos, which does.
[ ] I use ISO8601 timestamps since they're machine readable and can be parsed as a single field in the post slug.

stharrold commented 8 years ago

After each new post, in Google Search Console:
- [ ] "Crawl" > "Fetch as Google" > "Fetch and render" the relative path to the post. Click ">>" to check that it renders correctly, then click "Submit to index". Because I include a lot of references in my posts, I choose "Crawl this URL and its direct links" as the submit method, which Google permits up to 10 times per month for free.
- [ ] "Crawl" > "Sitemaps" > I resubmit both /feeds/all.atom.xml and /feeds/all.rss.xml so that all pages are indexed for both the Atom and RSS services.
- [ ] I publicize the post on LinkedIn, Google+, and Medium. I experimented with tweeting in 2011/2012 but found that it was time-consuming for me to maintain engaged followers. My posts to "Data Science Demos" are abbreviated tutorials, and didn't seem well-suited to the medium.
[ ] I publish my posts as soon as they're good enough, and publicize the posts on the post "Date".
[ ] My aim is to be useful. For novice readers, I provide as many links and footnotes as I recall were useful to me. For advanced readers, I try to be brief move the introductory material into the footnotes or within links.
[ ] include ipynb to html export lines
[ ] mention embedding html, issue #5
[ ] Blog stack: http://stackshare.io/stharrold/data-science-demos/details

stharrold / stharrold.github.io