========= Ideas ========= .. contents:: Themes ======= Data * URIs, HTTP, HTML, HTML5 * Linked Data (RDF, RDFa, JSONLD) * Graphs, Nodes, Edges seeAlso: * https://wrdrd.github.io/docs/consulting/knowledge-engineering.html Projects ========= Resume projects: * https://westurner.github.io/resume/html/resume.html#projects Projects I have created: * https://westurner.github.io/wiki/projects * Src: https://github.com/westurner/wiki/wiki/projects Open Source Projects I have contributed to: * https://westurner.github.io/wiki/contributions * Src: https://github.com/westurner/wiki/wiki/contributions Twedges -------- * (Directed) edges between tweets, later * Visualize "View Conversation" directed edges * Add edges with attributes and provenance metadata * Add a hashtag (recent/all tags w/ checkboxes) * Use case: I am developing a report on [Self Directed Learning], and have a few topics I'd like to discover and curate, mashup, remix * [ ] Twitter #JSONLD @context * Additional properties: * Expanded links * Picture URIs * Twitter Card Metadata * OpenGraph Metadata * Extracted RDF * Parse Errors * Additional edge classes: * TopicEdge * ReplyEdge * ReplyWithCommentEdge * FOAF, SIOC, Schema.org * Extract links, count uniques * #Nanotation RDF microsyntax (seeAlso: `Sphinx Extensions`) * Tab-completion for subjects, predicates, objects? * see http://json-ld.org/playground/ Trails ------- * Sequences of actions * https://github.com/westurner/brw * https://bitbucket.org/westurner/browselinks GrannyTweet.com ----------------- - "Because think first" - Concept stage Overlay -------- * EAV visualization RDFLib Extensions ------------------- * RDFLib IPython notebook * https://bitbucket.org/westurner/rdflib_notebooks * RDF exploration CLI * https://github.com/jonathanslenders/python-prompt-toolkit * schema-aware tab-completion Sphinx Extensions ------------------- * sphinxcontrib-RDF * https://github.com/westurner/sphinxcontrib-rdf * Directive to include an RDF syntax w/ pygments highlighting (see: healthref (ttl)) * Extract triples from an RST (docutils) node graph * Role w/ context? (like RDFa) * sphinxcontrib-ipython * run sphinx docs with ``.. ipython::`` directives through IPython interactively (e.g. w/ tmpipnb) * ipythonify tests (ipython_nose + what API w/ a notebook?) * sphinxcontrib-srclinks * Add links to {Source, Edit, History, Annotate} with {GitHub, BitBucket,} to the Sphinx HTML template sidebar. * Source: https://github.com/westurner/sphinxcontrib-srclinks * jinja2 template and contexts * sphinx extensions * Add an arrow indicating the currently selected heading (⬅) * Add an arrow indicating the currently selected TOC heading (⬅) * Scroll the sidebar to the selected TOC heading (¶) * "Open link in new tab" * "Show visited links" * [ ] Factor into sphinxcontrib. plugins * Source: https://github.com/wrdrd/docs/tree/master/docs/_static/js Open Source Mailing List Extractor ---------------------------------- Use cases: * https://mail.python.org/pipermail/python-ideas/2014-December/030228.html * incentivization of actionable crossreferences * PEP research * "is this actionable?" * "are we voting?" * Crawl/parse/extract links and +1 from given thread(s) * Detect a few standard link types: * Issue * Src * Doc * Ref * x-link * +1s with expandable snippets? (like ``grep -C``) * There could be configurable per-list link heuristics: * http[s] * Issue: https://bugs.python.org/issue(\d+) * Src: https://hg.python.org// * Src: https://github.com/// * Src: https://bitbucket.org/// * Patch/Attachment: https://bugs.python.org/(file[\d]+)/ * Doc: https://docs.python.org// * Wiki: https://wiki.python.org/moin/ * Homepage: https://www.python.org/ * Build: http://buildbot.python.org/all/builders/AMD64%20Ubuntu%20LTS%203.4/builds/771 * Packages + PyPI pkg: https://pypi.python.org/pypi/ + Warehouse pkg: https://warehouse.python.org/project/ * Wikipedia: https://[lang].wikipedia.org/wiki/ --> (dbpedia:) * This could - most efficiently - be added to mailman (e.g. in Postorious and/or HyperKitty) * http://mailman-bundler.readthedocs.org/en/latest/ * http://pythonhosted.org//mailman/ * https://mail.python.org/mailman/listinfo/mailman-developers Path / URI / URN and the __div__ operator ------------------------------------------- * build URI paths with ``/`` * URL encoding * some combination of operators could be useful:: scheme://user@host.netloc:port/path/?query#fragment from _ import URI, https, //, http https // 'user@host.netloc:port' / path ** query % fragment * __div__ / * __floordiv__ // * __and__ && * __mod__ % * __invert__ ~ * __or__ | * __xor__ ^ * __and__ + * __mul__ * * __pow__ ** * __lshift__ << * __rshift__ >> References ~~~~~~~~~~~~ * https://pypi.python.org/pypi/path.py -- __div__ * https://pypi.python.org/pypi/pathlib (stdlib) * https://pypi.python.org/pypi/furl * https://pypi.python.org/pypi/URLObject * https://pypi.python.org/pypi/rdflib schema.org/Animal, AnimalShelter, and ImageNet ----------------------------------------------- | tags: #linkeddata #schemaorg https://lists.w3.org/Archives/Public/public-vocabs/2015Feb/0121.html * http://schema.org/Animal ( http://schema.org/AnimalShelter ) * Labeling (e.g. ImageNet animals) * Matching / gamification schema.org/Course ------------------- | tags: #linkeddata #schemaorg - https://schema.org/Course - https://pending.schema.org/Course - https://github.com/schemaorg/schemaorg/issues/195 . - What is schema.org? https://wrdrd.github.io/docs/consulting/knowledge-engineering#schema-org ThingSequence, CourseSequence, CourseUnit ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ | tags: #linkeddata #schemaorg - Multi-Track Video Editing - https://github.com/westurner/brw (IFrame carousel timer) - https://github.com/westurner/pbm (Bookmarks JSON CLI & Tornado app w/ brw) - https://lists.w3.org/Archives/Public/public-schema-course-extend/2016Jan/0020.html - https://twitter.com/westurner/status/738471673767165953 - Fading, Beatmatching, Event spike synchronicity (cuepoints) - JSONLD (schema.org RDF) - learning from prepared materials / learning by watching / learning by doing LinkedReproducibility ---------------------------- | tags: #linkedreproducibility & #linkedrepro | tags: #linkeddata #schemaorg | Homepage: https://wrdrd.github.io/docs/consulting/linkedreproducibility | Src: - Linked- - StudyGraph - StructuredPremises - LinkedMetaAnalyses Links: | Hashtag:``#LinkedReproducibility`` - https://twitter.com/hashtag/linkedreproducibility - https://wrdrd.github.io/docs/consulting/data-science#linked-reproducibility - https://wrdrd.github.io/docs/consulting/linkedreproducibility - https://news.ycombinator.com/item?id=11685292 - https://news.ycombinator.com/item?id=12195302 Jupyter Notebook Meta object ----------------------------- .. note:: See: https://github.com/westurner/nbmeta/issues/1 - Jupyter Notebook - https://wrdrd.github.io/docs/tools/#jupyter-notebook - https://wrdrd.github.io/docs/tools/#nbformat - https://wrdrd.github.io/docs/consulting/education-technology#jupyter-and-reproducibility - ``_repr_html_`` => RDFa **and/or** JSONLD - Challenges / Objectives - I want some metadata to be displayed at the top; but also to add metadata through the notebook - Metadata about the JupyterNotebook - "Add JSONLD @context to the top level .ipynb node " https://github.com/jupyter/nbformat/issues/44 - Example JupyterNotebook metadata: .. code:: json {"@context": {"schema": "http://schema.org/"}, "@type": ["schema:JupyterNotebook", "schema:ScholarlyArticle", "schema:DataCatalog"], # dataset/hasPart ? "name": "Notebook Name", "author": { "@type": "schema:Person", "givenName": "Wesley", "familyName": "Turner", "url": "https://westurner.github.io/" }, "dateCreated": "2016-09-20" } - Metadata from the JupyterNotebook .. code:: json {"@context": { "schema": "http://schema.org/", "prov": "http://www.w3.org/ns/prov#"}, ..., "hasPart": { "@type": "schema:Dataset", "name": "Months of the Year", "distribution": { "@type": "schema:DataDownload", "contentUrl": "http://...", "encodingFormat": "CSV", # URI? } } } - SoftwareApplication versions - version_information, watermark - requirements.txt [EDIT] schema:JupyterNotebook, SoftwareEnvironment ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ see: - https://github.com/jupyter/nbformat/issues/44#issuecomment-275954832 schema.org classes for ``@type=[":JupyterNotebook",]`` : - https://schema.org/CreativeWork - https://schema.org/DigitalDocument - https://schema.org/Article - https://schema.org/ScholarlyArticle - https://schema.org/CreativeWorkSeries - https://schema.org/ItemList nbmeta ~~~~~~~ | Src: https://github.com/westurner/nbmeta | Src: https://github.com/westurner/nbmeta/tree/master/nb | Roadmap: https://github.com/westurner/nbmeta/issues/1 nbmeta is a research project for structured data in Jupyter (IPython) Notebooks. Objectives - Store and publish JSONLD and RDFa - https://github.com/westurner/nbmeta/issues/1 nbmeta and #LinkedReproducibility - StudyGraph -- things with structured edges (:CreativeWork with :name, :url, :authors, :funders, :about) - StructuredPremises -- premises with structured edges, satisfaction criteria - LinkedMetaAnalyses -- shared structured metadata, variable inclusion criteria Tools Vocabularies ===================== * See: RDFJS pad re "Tools Schema" Classes and Properties: https://text.allmende.io/p/rdfjs westurner.github.io/tools ---------------------------- | https://westurner.github.io/tools/ | Src: https://github.com/westurner/tools | Docs: https://westurner.github.io/wiki/tools.html#tool-collections Central tools.rst [ ... -> tools.rdf, tools.ttl, tools.jsonld ] * 1. https://westurner.github.io/provis/tools.html * 2. https://westurner.github.io/dotfiles/tools.html * 3. https://westurner.github.io/tools.html ( `westurner.github.io/tools`_ ) - Projects that link with `westurner.github.io/tools`: provis, dotfiles, https://wrdrd.github.io/docs/tools