<!-- TITLE: I want to make really big, detailed maps (of sociology) --> # Ideas + Are there any direct contradictions, clear cut disagreements, either over time or at a single time? What does the map of these contradictions look like? # Maps exist... [Eigenfactor](http://eigenfactor.org/about.php) makes some amazing maps! [like this one](http://eigenfactor.org/projects/well-formed/map.html), [this one](http://eigenfactor.org/projects/well-formed/radial.html), and [this one](http://eigenfactor.org/projects/well-formed/treemap.html). Don't forget to use your mouse! + many analyses on this website are [broken](http://eigenfactor.org/projects/motionGraph/) + this all points to an [amazing knowledge concentration](https://datalab.ischool.uw.edu/) at [UW](https://ischool.uw.edu/) + pls note that this is replicable + you can just copy down the JS files ([e.g.](view-source:http://eigenfactor.org/projects/well-formed/eigenfactor.js)), which link to the data files ([e.g.](http://eigenfactor.org/projects/well-formed/data/mapgml.json)) + this makes it somewhat simple to customize and repurpose this tool! [Eugene garfield's](http://www.garfield.library.upenn.edu/papers/mapsciworld.html) original [phanerozoic](http://www.chronozoom.com/#/t00000000-0000-0000-0000-000000000000/t48fbb8a8-7c5d-49c3-83e1-98939ae2ae67/td4809be4-3cf9-4ddd-9703-3ca24e4d3a26/t096717fd-64db-4416-92a7-6d35c83777d5@x=1.1037855897460738e-16&y=0&w=1.0239952009598081&h=2.113584907619056) And an even better timeline tool, [timeline JS](https://timeline.knightlab.com/#make) AND I discovered [this guy](http://truth-and-beauty.net/) + holy poo: [OECD better life index](http://www.oecdbetterlifeindex.org/countries/south-africa/) Don't forget [scimaps](http://scimaps.org/iteration). Led to [Kim](https://kimalbrecht.com/vis/#all), [flowing data](https://flowingdata.com/), [Outlier conference](https://www.outlierconf.com/speakers), [this amazing viz](http://scimaps.org/mapdetail/making_sense_of_skil_236) For more, look to [viz research](https://github.com/mathisonian/awesome-visualization-research) and [a list of frameworks](https://github.com/fasouto/awesome-dataviz). Thinking about using [Grasshopper](https://www.youtube.com/watch?v=Y66bSN4QFUE) for creating the shapes, but looks overly [complicated](https://st12.ning.com/topology/rest/1.0/file/get/2768810768?profile=original) and only [barely useful](https://www.youtube.com/watch?v=nZWzB_79xs0&t=182s) for my purposes. ## Linked jazz [Linked Jazz](https://linkedjazz.org/), e.g. [Art Tatum](https://linkedjazz.org/network/?person=Art_Tatum). Hosted now by [semantic lab](https://semlab.io/). Very cool... [This one](https://dmpublisher.s3-us-west-2.amazonaws.com/old-dm/directionsmag.com/ee/images/newsletter/2005/03_23/Jazz_Map1_lg.jpg) looks cool, but low quality. # Misc [ray shading the antarctic peninsula](https://twitter.com/mdsumner/status/1040126980279001088/photo/1) Puzzle printing: [The tweet](https://twitter.com/dataandme/status/1321792371898617857), and [the marketing](https://www.c82.net/euclid/puzzles/), and [the printing / shipping / etc](https://www.createjigsawpuzzles.com/sell/marketplace/byrnes-euclid.html) # Tiling / Rendering / Whatever [polymaps](http://polymaps.org/) seems to make amazing maps ([e.g.](http://polymaps.org/ex/flickr.html), [e.g.](http://opencyclemap.org/)) + uses [TileStache](http://tilestache.org/) server + rendered with [mapnik](https://mapnik.org/), "Generates vector tiles from OGR datasources in GeoJSON usable in Polymaps.", etc. This just looks like the answer. So what is the format I need to generate to render my map? + [What is a shapefile?](https://www.youtube.com/watch?v=PfbjoAcQp8I) (it's the thing you want to use) + [QGIS](https://qgis.org/en/site/forusers/download.html) is amazing + But, honestly it's just a friggin [pain](https://docs.qgis.org/2.14/en/docs/user_manual/working_with_projections/working_with_projections.html) to work with.... Maybe generate the shapefiles in python, [PyShp](https://github.com/GeospatialPython/pyshp#examples), and [pip:pyshp](https://pypi.org/project/pyshp/#adding-records)? + You [can script](https://docs.qgis.org/3.16/en/docs/pyqgis_developer_cookbook/intro.html#scripting-in-the-python-console) directly in QGIS... + You shouldn't need to know much about [map projections](https://docs.qgis.org/3.4/en/docs/gentle_gis_introduction/coordinate_reference_systems.html), since you're not mapping on a sphere. And there's a catesian grid, e.g. [EPSG:27700](https://gis.stackexchange.com/questions/311604/creating-a-project-without-projection-simple-cartesian-coordinates) <hr> + There are algorithms for constrained graphing, [e.g.](https://github.com/schochastics/graphlayouts), [e.g.](https://github.com/bdcht/grandalf), [e.g.](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.217.4984&rep=rep1&type=pdf), but these seem to all be for super small graphs. They're really cool however, and if I can use them for large graphs, I should just do that. + [Physics simulations](https://www.google.com/search?channel=tus2&client=firefox-b-1-d&q=physics+simulation+arbitrary+constraints) could provide the solution, as long as I can specify individual forces however I like (specify a field equation, e.g.) + [Matter.js is SO COOL](https://brm.io/matter-js/demo/#terrain). This could be the whole thing, if it were scalable (which it CERTAINLY is not). <hr> +[Embedding spaces](https://github.com/lferry007/LargeVis) of some sort could be the answer... + Maybe Ben would be able to understand and guide me on [this](http://vlado.fmf.uni-lj.si/pub/networks/doc/project/large0.pdf) kind of stuff. + [Look at this!](https://towardsdatascience.com/large-graph-visualization-tools-and-approaches-2b8758a1cd59) ## actually producing the tiles? + [maperitive](http://maperitive.net/docs/TenMinutesIntro.html) is a gui application for this, but seems a bit clunky + [export OSM bits to play around with](https://www.openstreetmap.org/export#map=19/38.03683/12.87095) + [metalist of rendering from geodata methods](https://wiki.openstreetmap.org/wiki/Rendering)... still haven't found anything which seems "simple" lol, and a [description](https://wiki.openstreetmap.org/wiki/Creating_your_own_tiles) from OpenStreetMap + [umap](http://umap.openstreetmap.fr/en/map/new/#14/54.8338/-7.4650) good for geographic but not custom... still need to serve my own tiles. perhaps once I have the tiles // lots of types of openGeo additions + another similar tool [geojson](http://geojson.io) + [awesome-geojson](https://github.com/tmcw/awesome-geojson) perhaps there's more to mine here ## seadragon [OpenSeadragon](https://openseadragon.github.io/) seems like a useful tool. Still not exactly sure how to modify what happens at different zoom levels (or indeed whether this is even possible) but it's easy & infinitely zoomable... e.g. this [blockchain viz](https://www.doc.ic.ac.uk/~dmcginn/adjmat.html) or infinite zoom on an image just using google maps like the [sky map](http://deepspacemap.com/#bookmark) this [bitcoin map](https://www.doc.ic.ac.uk/~dmcginn/adjmat.html) uses seadragon and this [cell zoom](https://www.allencell.org/deep-cell-zoom.html) super [deep](https://map.projectzomboid.com/#0.7689250336682178,0.24651287963535667,0.9) zoom another [deep one](https://embers.nicejacket.cc/known-eilarun.html) the only [network seadragon](https://research.cchmc.org/od/02/) I could find purportedly Gephi will [export](https://gephi.wordpress.com/2010/07/26/graph-visualization-on-the-web-with-gephi-and-seadragon/) to Seadragon... sounds dumb # Animation (using [Unreal Engine](https://www.unrealengine.com)) + [How to make Mechanical Joints with Physics Constraints in Unreal Engine 4 ](https://www.youtube.com/watch?v=bMEp9tAcisU) + [Physics: Pendulus and Spring](https://www.youtube.com/watch?v=y2wfX1Ud4uk) + [How to make Magnets using Physics Constraints in Unreal Engine 4](https://www.youtube.com/watch?v=--xR0aQ2rB0) + [Physics: Impulse and Force](https://www.youtube.com/watch?v=hNkjJFiymuc&t=380s) # point clouds & giant 3d viz + [infinite fractal zoom](https://peabrainiac.github.io/webglFractal3D/main.html) for inspiration + [potree](http://potree.org/potree/examples/viewer.html) for efficient in-browser point cloud viz + [this](http://creatingdata.us/techne/deep_scatterplots/#) had some giant point cloud visualization, like I'm trying to build. clunkyish but has an API // "Deep scatterplots" ([code](https://github.com/ColinEberhardt/d3fc-webgl-hathi-explorer), [explorer](https://colineberhardt.github.io/d3fc-webgl-hathi-explorer/#), [blog](https://blog.scottlogic.com/2020/05/01/rendering-one-million-points-with-d3.html)) + [with svg](https://d3fc.io/examples/webgl-with-svg-overlay/) + [Barnes-Hut](https://en.wikipedia.org/wiki/Barnes%E2%80%93Hut_simulation) always useful & important to learn, along with an [interface](https://github.com/d3/d3-quadtree) (d3) for it, [tutorial](https://medium.com/@waleoyediran/spatial-indexing-with-quadtrees-b998ae49336) + [this](https://github.com/PAIR-code/scatter-gl) could be useful for 50k interactive points (3d or 2d)Interactive 3D / 2D webgl-accelerated scatter plot point renderer. Core functionality from the embedding projector, capable of rendering and interacting with tens of thousands of points.