{"id":424,"date":"2014-06-24T23:54:48","date_gmt":"2014-06-25T04:54:48","guid":{"rendered":"http:\/\/bitquill.net\/blog\/?p=424"},"modified":"2016-05-12T14:31:08","modified_gmt":"2016-05-12T19:31:08","slug":"towards-laws-of-the-3d-printable-design-web","status":"publish","type":"post","link":"https:\/\/bitquill.net\/blog\/towards-laws-of-the-3d-printable-design-web\/","title":{"rendered":"Towards laws of the 3D printable design web"},"content":{"rendered":"

With the\u00c2\u00a0explosive growth of 3D printing, and rapid manufacturing at the consumer level in general, physical\u00c2\u00a0objects can be designed and manipulated in a computer. However, like other forms of digital content (e.g., documents, software, music), this is only part of the story: digital representation also enables online sharing and collaboration (as Chris Anderson has pointed out). A prime example of the potential of all these technologies combined with online sharing and collaboration is the initial design of consumer-grade 3D printers themselves\u00c2\u00a0which, perhaps unsurprisingly, was what many early adopters of the technology used it for<\/a>. \u00c2\u00a0Considering that the rest of us is where those early adopters were five or more years ago, the future should be interesting.<\/p>\n

Despite hearing about 3D printing daily, very few studies have looked at the digital content of physical things, and the processes that generate it. I collected data some time ago, and started off with this visualization<\/a>, which I wrote about before<\/a>. A further initial analysis of the data has some interesting stories to tell.<\/p>\n

<\/p>\n

Exponential growth rates.<\/strong><\/em>\u00c2\u00a0 The total number of things over time\u00c2\u00a0(blue) exhibits an exponential growing, with a compound doubling time\u00c2\u00a0of 6.1 months. Furthermore, if we consider only remixes (green), then\u00c2\u00a0the growth rate far outpaces the overall rate, with a compound\u00c2\u00a0doubling time of 4.6 months. Consequently, the relative ratio of remixes is also growing at an exponential pace (red) and, although obviously this cannot continue forever, there is little evidence that the growth rate of remixing is abating (in fact, after the introduction of the Thingiverse\u00c2\u00a0Customizer, which is excluded from this plot, the rate has picked up even further).<\/p>\n

\"thingiverse_growth_log\"<\/a><\/p>\n

\n
\n
\n

Popularity: views vs. likes vs. makes<\/span><\/strong>.<\/strong>\u00c2\u00a0\u00c2\u00a0<\/span>The following table summarizes the results of\u00c2\u00a0least-squares regression on measures of user actions, showing\u00c2\u00a0the top-3 best predictive features (\\(p < 0.01\\), ranked by \\(t\\)-test scores) with 95% confidence intervals of the corresponding\u00c2\u00a0regression coefficients, as well as the bottom-2 worst features.<\/p>\n<\/div>\n<\/div>\n<\/div>\n\n\n\n\n\n\n\n\n\n
Variable<\/th>\nBest predictors<\/th>\nWorst predictors<\/th>\n<\/tr>\n<\/thead>\n
\\(\\mathit{\\#Views}\\)<\/td>\n\\(\\mathit{\\#Likes}\\!: 43.1\\text{–}44.6, \\mathit{\\#DLs}\\!: 0.35\\text{–}0.38, \\mathit{\\#Views}’\\!: 0.28\\text{–}0.31\\)<\/td>\n\\(\\mathit{\\#Make}’\\, (p=0.48), \\mathit{\\#Remix}’\\, (p=0.06)\\)<\/td>\n<\/tr>\n
\\(\\mathit{\\#DLs}\\)<\/td>\n\\(\\mathit{\\#Likes}\\!: 43.1\\text{–}44.6, \\mathit{\\#DLs}\\!: 0.35\\text{–}0.38, \\mathit{\\#Views}’\\!: 0.28\\text{–}0.31\\)<\/td>\n\\(\\mathit{\\#Remix}\\, (p=0.66), \\mathit{\\#Remix}’\\, (p=0.51)\\)<\/td>\n<\/tr>\n
\\(\\mathit{\\#Likes}\\)<\/td>\n\\(\\mathit{\\#Views}\\!: 0.006, \\mathit{\\#Make}\\!: 2.72\\text{–}2.83, \\mathit{\\#Likes}’\\!: 0.42\\text{–}0.46\\)<\/td>\n\\(\\mathit{\\#Remix}’\\, (p=0.59), \u00c2\u00a0\\mathit{\\#DLs}’\\, (p=0.27)\\)<\/td>\n<\/tr>\n
\\(\\mathit{\\#Makes}\\)<\/td>\n\\(\\mathit{\\#Likes}\\!: 0.074\\text{–}0.077, \\mathit{\\bf\\#Files}\\!: -0.13\\text{–}0.11, \\mathit{\\#Makes}’\\!: 0.28\\text{–}0.33\\)<\/td>\n\\(\\mathit{\\#Remix}’\\, (p=0.99), \u00c2\u00a0\\mathit{\\#DLs}’\\, (p=0.51)\\)<\/td>\n<\/tr>\n
\\(\\mathit{\\#Remix}\\)<\/td>\n\\(\\mathit{\\#Views}\\!: 0.0003, \\mathit{\\bf\\#Remix}’\\!: 0.18\\text{–}0.27, \\mathit{\\bf\\#Sources}\\!: 0.19\\text{–}0.39\\)<\/td>\n\\(\\mathit{\\bf\\#Make}’\\, (p=0.71), \u00c2\u00a0\\mathit{\\#DLs}\\, (p=0.66)\\)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

The relative incidence of user actions depends on the relative effort\u00c2\u00a0required to take those actions. Therefore, we observe that roughly\u00c2\u00a0(order of magnitude) 100 views “contribute” one like in our linear\u00c2\u00a0models, and roughly 10 likes “contribute” a make. The first is not\u00c2\u00a0particularly surprising. However, the fact that only 10\\(\\times\\) likes\u00c2\u00a0contribute a make seems to suggest that users are actively seeking\u00c2\u00a0things, and have the means and motivation to actually print<\/em>\u00c2\u00a0things that they have liked.<\/p>\n

Another intuitive, in retrospect, observation is that the number of\u00c2\u00a0files has a negative<\/em> effect on makes. This provides evidence\u00c2\u00a0for the hypothesis that simpler things (consisting of fewer parts) are\u00c2\u00a0more likely to be made.<\/p>\n

Sublinearities and power-laws.<\/strong><\/em>\u00c2\u00a0 The first figure below shows the number of likes vs. makes, and the second figure shows views vs. likes (both smoothed using exponential-size buckets). \u00c2\u00a0The emerging relationships are that\u00c2\u00a0\\(\\mathit{\\#Likes} \\propto\u00c2\u00a0\\mathit{\\#Makes}^{0.70}\\) and \\(\\mathit{\\#Views} \\propto\u00c2\u00a0\\mathit{\\#Likes}^{0.85}\\). \u00c2\u00a0Similar relationships have been observed in other domains. \u00c2\u00a0However, if we look at remixes vs. makes, no such pattern emerges, which brings us to a last point.<\/p>\n

\"likes-vs-makes-expbucket\"<\/a>\"views-vs-likes-expbucket\"<\/a>\"remix-vs-makes-expbucket\"<\/a><\/p>\n

Popular vs. Generative.<\/strong><\/em> \u00c2\u00a0Perhaps the most surprising observation is that typical measures of general\u00c2\u00a0popularity have little relation to whether a thing is remixed or not:\u00c2\u00a0(i) makes are, in fact, the worst predictor of number of remixes (table and last figure above); and (ii) in fact, the number of\u00c2\u00a0remixes is a bad<\/em> predictor of almost everything, except<\/em>\u00c2\u00a0of other remixes (table above). This suggests that\u00c2\u00a0aspects of a design that make it broadly appealing are distinct from\u00c2\u00a0aspects that make it inspiring and, furthermore, agrees with the\u00c2\u00a0author’s personal experience that following remix links is more useful\u00c2\u00a0when looking for ideas, than when looking for utilitarian or fun\u00c2\u00a0things to print.<\/p>\n

What next?<\/strong><\/em> \u00c2\u00a0As a “bonus”, here is a visualization of the evolution of the largest connected component of the remix graph (with Customizer outputs excluded). \u00c2\u00a0The last frame is essentially the same data as in our interactive visualization<\/a>. \u00c2\u00a0This video was hacked together using Matplotlib’s basic\u00c2\u00a0animation facilities and layed out using a simple breadth-first traversal of the graph. \u00c2\u00a0Not as pretty as it can\u00c2\u00a0be, but it still shows an interesting picture.<\/p>\n

http:\/\/bitquill.net\/blog\/wp-content\/uploads\/2014\/03\/rxg_animation_slower2.mp4<\/a><\/video><\/div>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

With the\u00c2\u00a0explosive growth of 3D printing, and rapid manufacturing at the consumer level in general, physical\u00c2\u00a0objects can be designed and manipulated in a computer. However, like other forms of digital content (e.g., documents, software, music), this is only part of the story: digital representation also enables online sharing and collaboration (as Chris Anderson has pointed […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":false,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[67,45],"tags":[69,32,55,41],"class_list":["post-424","post","type-post","status-publish","format-standard","hentry","category-make","category-scitech","tag-3d-print","tag-data-mining","tag-social-media","tag-social-networks"],"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_shortlink":"https:\/\/wp.me\/p7x9xm-6Q","jetpack-related-posts":[],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/posts\/424","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/comments?post=424"}],"version-history":[{"count":50,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/posts\/424\/revisions"}],"predecessor-version":[{"id":699,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/posts\/424\/revisions\/699"}],"wp:attachment":[{"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/media?parent=424"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/categories?post=424"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/bitquill.net\/blog\/wp-json\/wp\/v2\/tags?post=424"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}