<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Matti Schneider on Medium]]></title>
        <description><![CDATA[Stories by Matti Schneider on Medium]]></description>
        <link>https://medium.com/@MattiSG?source=rss-166622c8ac77------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*5B728Knz1H7cXAQmrAoMmQ.jpeg</url>
            <title>Stories by Matti Schneider on Medium</title>
            <link>https://medium.com/@MattiSG?source=rss-166622c8ac77------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 18:06:21 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@MattiSG/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Jekyll Lean Startup template]]></title>
            <link>https://medium.com/@MattiSG/jekyll-lean-startup-template-ec6f8d94211c?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/ec6f8d94211c</guid>
            <category><![CDATA[coding]]></category>
            <category><![CDATA[collaboration]]></category>
            <category><![CDATA[jekyll]]></category>
            <category><![CDATA[cms]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Thu, 23 Apr 2020 08:31:00 GMT</pubDate>
            <atom:updated>2020-04-23T08:31:00.920Z</atom:updated>
            <content:encoded><![CDATA[<h4>A template for small teams to deploy landing pages and small websites cheap and fast, engage non-technical contributors, and architect information over time.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pU0IfavpA7gG7S4IGMU2nQ.jpeg" /></figure><p>I regrouped the best practices I have identified over years of using the static website generator <a href="https://jekyllrb.org/">Jekyll</a> to deliver landing pages under half a day that evolve into full-fledged production websites where non-technical team members contribute directly to the content through online code editors.</p><p>The developer experience is most of the time terrible with content management systems that ship WYSIWYG editors, and learning HTML and managing metadata is too much for the vast majority of non-technical users. <strong>I aim for the sweet spot where efficient contribution is possible both for developers and non-technical users.</strong></p><p>This approach has enabled me to build products such as:</p><ul><li><a href="https://beta.gouv.fr/">beta.gouv.fr</a>, <strong>the website of the French government public digital services incubator</strong>, which uses Jekyll both as CMS and HR system and has <a href="https://github.com/betagouv/beta.gouv.fr/">over 370 contributors</a> of all sorts.</li><li><a href="https://pariscall.international/">pariscall.international</a>, <strong>the public website for the Paris Call for Trust and Security in Cyberspace</strong>, which uses Jekyll to maintain a <a href="https://github.com/ambanum/pariscall.international/tree/master/_supporters">list of signatories</a> and relies on an additional <a href="https://github.com/ambanum/pariscall.international-backend">backend</a> to <a href="https://github.com/ambanum/pariscall.international/commits?author=AmbNum-Bot">create files</a> from survey answers, making it easy for anyone to add files and for non-technical maintainers to update the data over time.</li><li><a href="https://disinfo.quaidorsay.fr/">disinfo.quaidorsay.fr</a>, a resource <strong>supporting actors countering information manipulation</strong> (aka “fake news”), which uses Jekyll for a <a href="https://github.com/ambanum/disinfo.quaidorsay.fr/tree/master/_pages">multilingual</a>, <a href="https://github.com/ambanum/disinfo.quaidorsay.fr/tree/master/_pages/en/encyclopedia">collaborative encyclopedia</a> of tools and practices.</li><li><a href="http://lesquais.voyage/">lesquais.voyage</a>, the digital extension of an artistic installation.</li></ul><p>This template recommends the use of collections over simple pages, styling through collection names, local mirroring of GitHub Pages production versions…</p><p><strong>See the feature list and the code on </strong><a href="https://github.com/MattiSG/jekyll-template"><strong>github.com/MattiSG/jekyll-template</strong></a><strong>.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ec6f8d94211c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Counter disinformation and improve lockdown adherence by transforming rules into code]]></title>
            <link>https://medium.com/@MattiSG/counter-disinformation-and-improve-lockdown-adherence-by-transforming-rules-into-code-85639b7e45d9?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/85639b7e45d9</guid>
            <category><![CDATA[rules-as-code]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[code-is-law]]></category>
            <category><![CDATA[government-as-a-platform]]></category>
            <category><![CDATA[disinformation]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Fri, 10 Apr 2020 13:57:41 GMT</pubDate>
            <atom:updated>2020-04-10T13:57:59.230Z</atom:updated>
            <content:encoded><![CDATA[<h3>Countering disinformation and improving lockdown adherence by transforming rules into code</h3><h4>Broadcast official recommendations with an API to support the fight against Covid-19</h4><p><em>French version: </em><a href="https://medium.com/@matti_sg_fr/lutter-contre-la-d%C3%A9sinformation-et-suivre-le-confinement-en-transformant-la-loi-en-code-9c765d875668"><em>Lutter contre la désinformation et suivre le confinement en transformant la loi en code</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XlEhZrzBnW_N_iBPMESSsQ.png" /></figure><p><strong>In the face of the COVID-19 epidemic, civil society responded as usual with a speed and strength initially exceeding that of government services. </strong>Even before the pandemic status was announced, articles written <a href="https://medium.com/@tomaspueyo/coronavirus-act-today-or-people-will-die-f4d3d9cd99ca">by individuals</a> gave recommendations for containment based on Asian examples, collectives shared predictions of the health impact, <a href="https://epidemicprotect.com/">organisations compiled</a> self-diagnosis techniques… Official speeches and actions only amplified this movement and initiatives have multiplied: compilation of epidemiological data, mapping, dissemination of official recommendations by memes, self-diagnosis tools to relieve medical services…</p><p>In a connected society, information spreads rapidly and widely, usually with slight changes at each transmission. In the case of official announcements accompanied by regulatory texts, these progressive alterations of the message are due to the nature of the media: a video and an explanatory text can easily be shared between humans. However, the only authentic source, the regulatory text, which is supposedly unambiguous, is unsuitable for mass information: the drafting format is often complex to grasp, and is therefore mostly ignored by the general public. It is thus rephrased for communication, with a risk of alteration. Unless a system of mass monitoring and censorship is to be put in place, it is impossible (and probably even then not desirable) to control dissemination by third parties: sharing on WhatsApp with one’s family, summarising over Messenger with friends, publishing a website listing current regulation , installing an app to monitor the situation…</p><p>The vast majority of these initiatives are carried out by volunteers who master digital technology and collaborate with experts in other fields (such as, obviously, the medical field) in a spirit of solidarity, in order to contribute to national and global efforts. <strong>This <em>multitude</em> will by definition always be more powerful than public agencies</strong>, since in a democracy there are more forces outside than inside the State (cf. <a href="https://www.armand-colin.com/lage-de-la-multitude-2e-ed-entreprendre-et-gouverner-apres-la-revolution-numerique-9782200601447"><em>L’Âge de la Multitude</em></a>).</p><p>However, it is true that multiplying sources of information does raise an issue: initiatives follow one another and duplicate one another, the error of one leads to a cascade in others, and above all the speed at which the situation evolves makes some obsolete without their creators caring enough —or being able— to correct their services fast enough. <strong>The once-useful information they contributed to disseminate has a persistence, which once outdated can become dangerous.</strong></p><blockquote>For example, in France, government recommendations have rapidly evolved from calling a specific phone number to a different one in case of Covid-19 suspicions, from consulting at a medical clinic to remote interviews, from allowing to prohibiting to allowing again a digital version of mandatory travel documents… Many tools are not yet up to date with all these changes and the initial goodwill of the people who created them may have been eroded, diminishing their commitment to keeping them up to date and effectively adding to ambient noise.</blockquote><p>Taking into account the delay in disseminating and referencing instructions and the persistence of obsolete instructions, it is foreseeable that uncertainty will follow, for example, the announcement of new consultation modalities, passage to higher stages of lockdown, or the enactment of additional movement restrictions, causing a loss of responsiveness and making the crisis increasingly difficult to manage over time.</p><h3>The usual response: forbid alternatives</h3><p>Faced with this reality, <strong>the administration’s reflex is most often to <em>oppose</em> this creativity</strong> by reminding people of the official channels of information: only the PDF file of the Ministry of the Interior must be used for lockdown derogations; only the social account of the Ministry of Health must be relied upon for medical calling instructions; etc.</p><p><strong>By wanting to be the only one to speak, the State has a deeply counterproductive action.</strong> First of all, by depriving itself of precious amplifiers: those who could relay its messages most faithfully and by means inaccessible to the administration are inhibited by at least partial compliance with its request not to do so. Conversely, bad actors who seek to manipulate information don’t give a damn of these instructions and benefit from the reduction of information channels: their messages are counterbalanced only by official channels and no longer by a multitude of channels that disseminate convergent information.</p><p><strong>Full control is not achievable in a democracy, and the current situation favours the few bad actors who deliberately disseminate false information </strong>—be it for their own benefit, out of misunderstanding or by want of chaos— by inhibiting civil society’s ability to generate its own antibodies.</p><p>It should be noted that public actors are much more accustomed to this role being played by the media —even though information is distorted there too. It is unimaginable that the State would prohibit the rephrasing of its recommendations to newspapers and television channels and enforce the pure diffusion of official speeches without comment or explanation.</p><h3>Support broadcasting official recommendations</h3><p><strong>The most effective way to reduce information entropy is to minimise the friction to replicating the original information and to maximise the number of channels disseminating it.</strong> Quite the opposite of the current dominant approach of minimising channels of dissemination, which leads by construction to a minimisation of the dissemination itself; and which is made even worse in an adversarial context where there are other sources disseminating disinformation, less strong but more numerous. Copying and sharing should be supported, and duplication encouraged as long as means are provided to reflect updates without delay.</p><blockquote>To put it another way: since a self-diagnosis chatbot on Messenger will be created anyway, and the Ministry of Health does not have the means to make one and to get it adopted it as fast as civil society, then it should be based on reliable information whose update can be done at zero cost for its creators. The State’s resources will thus be much better invested in trivially certifying tools rather than fighting against their existence.</blockquote><p>This proposal therefore aims to <strong>build an ontology of government recommendations and disseminate them through an API</strong>. It is a matter of equipping digital makers in order to leverage their ability to create services that quickly cover the entire population; enabling them to obtain symbolic and even financial recognition for their work; all the while standardising the broadcast messages.</p><p>By making the effort to standardize its recommendations and to represent them in a machine-readable way, the administration merely updates its ability to issue press releases in the digital age, with the unique opportunity of increased responsiveness in crisis situations.</p><blockquote>Let us take the example of rail transport. Communicating timetables, delays and cancellations by a rail operator shares a lot with communicating health measures by an official source, in particular requirements of large scale and short timeframe.</blockquote><blockquote>To achieve this ubiquity, the rail operator relies first and foremost on communication through machines: official website, ticket resellers, mobile applications, amateur sites, ticket counters, ticket control terminals and station display panels, all are updated within minutes.</blockquote><blockquote>What if a health measure with national scope could be displayed in all stations a few moments after the decree is published?</blockquote><p>As an illustration of its feasibility and usefulness, the same ideas were implemented for the <a href="https://github.com/Delegation-numerique-en-sante/covid19-algorithme-orientation">French self-diagnosis algorithm</a>, leading in a few days to <a href="https://sante.fr/covid-numerique">many solutions</a> rather than a <a href="https://www.maladiecoronavirus.fr/">single web application</a>.</p><h3>International Collaboration</h3><p><strong>This standardisation will also have the side-effect of facilitating international collaboration</strong> and benchmarking. By creating a shared representation of measures already taken by other countries, one can anticipate measures that will likely be taken, and allow the sharing and reuse of tools by countries that will soon be affected. Instead of a thousand efforts spread over a thousand divergent initiatives, we could join efforts on one initiative —or at least on interoperable ones. The database created by consolidating the description of government responses will be valuable for social science and crisis management research.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ClKAYR9Dxe8hOXd7T7JYrQ.png" /></figure><p><em>This proposal was </em><a href="https://twitter.com/matti_sg_fr/status/1241357602879127552"><em>published on March 21</em></a><em> and then submitted after feedback and team building to the French </em>Agence Innovation Défense<em> as part of its </em><a href="https://www.defense.gouv.fr/aid/appels-a-projets/appel-a-projets-lutte-covid-19"><em>Call for Proposals for Innovative Solutions to Combat Covid-19</em></a><em>. It was turned down on April 7, 2020.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=85639b7e45d9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Post post mortem]]></title>
            <link>https://medium.com/@MattiSG/post-post-mortem-8aeb7ba1009e?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/8aeb7ba1009e</guid>
            <category><![CDATA[diversity-in-tech]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Thu, 07 Feb 2019 18:47:19 GMT</pubDate>
            <atom:updated>2019-02-07T18:47:19.736Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<h4>A short story on engineering blinkers</h4><p>Last year, on a Friday, I had a presentation of the benefits assessment tool I was a Product Manager for. It was its third “official opening”, but hey you don’t complain when you get visibility 😉 I got two ministers there, along with all that ensues at that level: local officials and, obviously, journalists.</p><p>It just happened that the weekend right after I had to help friends to empty a house full of movie shooting gear, so I was about to leave for three days of cleaning up in the countryside.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OmU0yJZXigmUj4vQPO94Ow.jpeg" /><figcaption>That’s the kind of stuff we had to clear. It did take a while!</figcaption></figure><p>I had given the usual presentation and shaken hands and given a couple interviews while the ministers were doing their job (bringing people in and congratulating officials for stuff they have just been briefed about).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*gMjEljwlAJImdovEJ0phyw.png" /><figcaption>The things you do to satisfy partners…</figcaption></figure><p>All of that was combined with the opening of a nationwide ad campaign. After 2 years of trying, we had finally nailed radio spots and sponsored posts on social networks. My team had to create a <a href="https://github.com/betagouv/mes-aides-campaign">specific subdomain</a> to add required third-party trackers without contradicting our privacy policy and it had been quite a bit of work but once again: you don’t complain.</p><p>So we’re moving all that heavy cinema equipment around and we have only two days to finish and I’m nosebleeding and obviously that’s the BEST MOMENT to get paged at 1pm on a weekend.</p><p>Two of my sensors are down, the app doesn’t seem to come back up on its own, I’m in a place with lousy internet. I try to load the website immediately, it does render. But a couple minutes later, I get a second text notification confirming that the computation component is still down… I immediately call another teammate and let him finish lunch while I take a first look, and we agree to rotate throughout the afternoon.</p><p>So there I was, spending 20 minutes foraging through logs while bringing the computation engine back up, writing a collaborative assessment and future postmortem amidst cardboard packaging and film equipment. The outage was about 15 minutes long total, in two sessions, and was invisible to most users as our caching strategy was valid and the UI was still served; some users only got an error telling them that the computation was not possible at that time. The logs showed that we had had as many hits in 10 minutes as we were used to in a day.</p><p>But for us, there was still no way to understand what happened and why there was a peak in activity, so it was likely to happen again anytime. I had the sensors dashboard full size on my screen for the whole afternoon and evening, checking it every time I was moving stuff in and out of the room, ready to manually restart the server anytime.</p><p>We managed to clean everything, there was no other outage over the weekend, and it was all fine even though it was stressful. My team was all up and available, we protected the one teammember who was off, our community manager did a great job at keeping users updated, and we made sure to have at least one person pageable with a reliable internet connection at any time of the day or night.</p><h3>The After</h3><p>I spent the best of Monday analysing logs to determine if we had suffered from a DoS attack, a DDoS, a major component failure… I measured packet frequency, payload origin, found no pattern, checked worst offenders against known blacklists… And then, mid afternoon, our series-watching, glitter-loving, TV shows-fan CM casually drops:</p><blockquote>“Hey, I think I found it. We’ve been on national TV news.”</blockquote><h3>🤦‍♀️</h3><p>I watch so little TV I did not even know one of the main channels does their noon news edition 15 minutes earlier than others. While I was raging out loud and looking for attacks and technical failures, my CM was watching the replays of all TV news of the day we had the outage inbetween replying to users, and he found the reason without any effort. We had just been hit by thousands of very legit users, drawn to our app by a mention on mass media.</p><p>So hey, that’s one small illustration of why it’s worth hiring for diversity. I don’t know shit about TV, and apparently neither did the other engineers in my team. I could have wasted even more time and energy trying to figure out what happened: I had eliminated that option as the hours did not match, and I was looking in the dimensions that were the most tangible for me and was expecting our press partners to let me know if we had major outlets appearances. They failed, but my team won.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/391/1*7SyNaT5IuMusYVnxXkT0Fw.png" /><figcaption><a href="https://github.com/betagouv/mes-aides-ui/pull/493/files">That</a>’s called “scaling”, apparently.</figcaption></figure><p>And that was the entirety of the necessary technical fix.<br>👈</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8aeb7ba1009e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[“HTTPS considered harmful”, yes, but isn’t HTTP too?]]></title>
            <link>https://medium.com/@MattiSG/https-considered-harmful-yes-but-isnt-http-too-1ee1f4a36358?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/1ee1f4a36358</guid>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[https]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Thu, 11 Jan 2018 21:25:01 GMT</pubDate>
            <atom:updated>2018-01-11T21:25:01.218Z</atom:updated>
            <cc:license>http://creativecommons.org/licenses/by/4.0/</cc:license>
            <content:encoded><![CDATA[<p><em>Since </em><a href="https://larlet.fr/"><em>David</em></a><em> doesn’t allow public responses, I’ll try and follow his way of publishing letters in personal spaces.</em></p><p>Dear David,</p><p>Following on our conversation regarding your point that <a href="https://larlet.fr/david/stream/2018/01/06/">HTTPS is harmful</a>:</p><blockquote>Encouraging everybody to switch to HTTPS promotes strong dependency to a third-party mafia, increases load time, makes your content inaccessible if you have any trouble reconducting your certificate, avoids migrating easily from one hosting platform to another, forces upgrading on a lot more security issues if you are hosting yourself. Even worse, when you switch there is no harmless turning back! <strong>That’s not the Web I’m aiming for.</strong></blockquote><p>To which I replied earlier a technical answer:</p><h3>Matti Schneider on Twitter</h3><p>@davidbgk Re: HTTPS. - strong dependency to a mafia: #LetsEncrypt? - increases load time: #HTTP2 improves them. - inaccessibility if trouble reconducting: don&#39;t use HSTS! - avoids migrating easily: why? 301 are honored over HTTPS. - forces upgrading: certbot autorenew works well.</p><p>There is now a <a href="https://larlet.fr/david/stream/2018/01/10/">new set of arguments</a> you’re putting forward. Here is my reply:</p><blockquote><em>When you turn an oligopole into a monopole, it cannot be a mafia anymore, heh.</em></blockquote><p>Let’s Encrypt not being a mafia does not mean it’s not a SPOF. The centralised model is bad. Question is not “is HTTPS perfect”, it is “is HTTPS better than HTTP”. Maybe we do not mean the same thing by “mafia”.</p><blockquote><em>“0-RTT will reduce initial load time.” One day, maybe. But for now it’s quite limited to say the least.</em></blockquote><p>We’ll see. TLS 1.2 did get a nice push forward thanks to Let’s Encrypt. With more cloud providers, docker images, shared Ansible configurations, and default Nginx setups, upgrading server setups goes faster than it used to.</p><blockquote><em>“HTTP2 is good for performances.” […] HTTPS highly impacts my First Byte Time though.</em></blockquote><p>Now we go into the details of what <em>is</em> performance. Of course if you consider TTFB, HTTP+TLS 1.2 is slower than HTTP. No argument here. Have you measured the difference in <a href="https://developers.google.com/web/tools/lighthouse/audits/first-meaningful-paint">time to first meaningful paint</a> though? So often, that entails loading some CSS and images.</p><p>HTTP2 multiplexing allows me to keep CSS files split by components with no additional request cost, which then allows me to leverage cache at a very granular level, highly speeding up navigation.</p><blockquote><em>“You have the guarantee your content is not altered.” Except if done once downloaded.</em></blockquote><p>I don’t understand your point. Views being rendered on the client-side from controlled code is one thing, intermediaries injecting trackers or serving ads within your code is another.</p><blockquote><em>“Don’t use HSTS!” I don’t get the point of providing content over HTTPS if you do not force it somehow</em></blockquote><p>“somehow”, yes. A 301 or 302 is not the same as HSTS. That was a reply to your impression that having a certificate “makes your content inaccessible if you have any trouble reconducting your certificate”. Refusing to serve over HTTP but being able to do so if needed is not the same as explicitly forbidding recovery in case you cannot renew. One should use HSTS only if one has the resources to maintain that infrastructure properly, with recovery keys.</p><p>I don’t argue HTTPS is overkill for many uses, especially for websites that provide read-only, low-importance information, and I do agree with your underlying expectations of simplicity and performance. My replies are only there because you used several technical arguments that I consider slightly exaggerated. You mention “not in my case”, then it would be worth describing that case more precisely in the article (though I think I see the kind of small, server-rendered, simple website you’re talking about and that I also tend to ship).</p><p>I’d argue that there’s a flip side to the ethical arguments too, though, and these are the ones that personally convinced me to take the burden of the added complexity.</p><p>You’re right, “<em>That’s not the Web I’m aiming for</em>.”. But the Web where corps <a href="https://www.eff.org/deeplinks/2014/11/verizon-x-uidh">inject ad trackers</a> in traffic and where government agencies massively <a href="http://static.guim.co.uk/sys-images/Guardian/Pix/audio/video/2013/7/31/1375269604628/KS8-001.jpg">tap HTTP</a> isn’t either. And in this game, as a professional developer, I prefer to take on the complexity and work harder to optimise my content for performance in order to protect my users, and the users of the Web in general. For sure, certificate issuers are now honeypots, and we need to keep pushing for decentralised trust. But <a href="http://mattischneider.fr/h4yr/">the IX were already</a>, and it has become way too easy for intermediaries to do anything they want on clear text to refuse taking care of the privacy of our users. I am glad Wikipedia is served over HTTPS, not because I’m afraid my ISP will change the birth date of Napoleon, but because I don’t want it to know which articles I’m reading at what time. The profiling power of aggregated metadata is too strong for us not to make it as hard as we can for spies to leverage it against our fellow netizens.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1ee1f4a36358" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Stupid Science: I compared 23 sticky notes to help you spare wallet and planet]]></title>
            <link>https://medium.com/@MattiSG/stupid-science-i-compared-23-sticky-notes-to-help-you-spare-wallet-and-planet-fc9b97d88503?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/fc9b97d88503</guid>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[collective-intelligence]]></category>
            <category><![CDATA[sticky-notes]]></category>
            <category><![CDATA[tools-for-design]]></category>
            <category><![CDATA[research-reports]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Thu, 23 Nov 2017 10:42:31 GMT</pubDate>
            <atom:updated>2023-11-10T20:49:16.813Z</atom:updated>
            <cc:license>https://creativecommons.org/licenses/by-nc-nd/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eyQAUbfrzlQ6WUT7wwWMig.jpeg" /></figure><p>Over the past months, I’ve done short instructional videos on <a href="https://www.youtube.com/watch?v=owWMB6In9LQ&amp;list=PLhQ0uCLmau9E7LCbA4LD47z-fcJxAlr8U">how to use sticky notes</a> (in French, with English subtitles — you’re welcome to <a href="http://www.youtube.com/timedtext_cs_panel?c=UCacJcmB2wsdLD6c6eECl6bg&amp;tab=2">help in subtitling</a> in other languages!). The main goals are to share best practices and make it easier for professional facilitators to ready participants.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fvideoseries%3Flist%3DPLhQ0uCLmau9E7LCbA4LD47z-fcJxAlr8U&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DowWMB6In9LQ&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FowWMB6In9LQ%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/c3deae05c45ac89de3660e8d34714747/href">https://medium.com/media/c3deae05c45ac89de3660e8d34714747/href</a></iframe><p>One of the announced videos was “which sticky note model should I choose?”. It took a bit longer than expected to deliver. There was good reason for that delay though: <a href="https://twitter.com/matti_sg/status/895010249945600002">I went overboard</a> with scope in order to answer the question properly 😁</p><p>Thus, I present: “<strong>which sticky notes should you choose out of 23, based on way too much data</strong>”. <em>(Want to get straight to the result? Ah, you’re missing on all the fun… just scroll down to the end </em>😉<em>)</em></p><h3>Dimensions</h3><p>I compared products based on the issues most often encountered when using sticky notes in a professional agile setting:</p><ol><li>How well does the sticky stay on a wall over time?</li><li>How likely is the ink to leak through a note?</li><li>How hard is it to open the pack?</li><li>How expensive are packs?</li><li>How socially and environmentally sustainable is the product?</li><li>How bad does the pack smell when you open it? 😄</li></ol><p>After specific experiments, I rated each product on a 0-to-5 index for each of these dimensions.</p><h3>Choosing the models to compare</h3><p>I just took all the sticky notes I could get my hands on! And then, I thought 8 wasn’t worth it to run one-month-long experiments, so I just hunted and bought <em>all</em> the standard size, acceptably priced stickies on Amazon France.</p><h3>Matti Schneider (FR) on Twitter</h3><p>Tes post-its, tu les veux en chocolat (https://t.co/XLsll45KIx), salami (https://t.co/EzevXzAenk) ou cupcake (https://t.co/zS9XObytG6) ?</p><p>I want to thank the <a href="http://agile-france.org/">AgileFrance association</a> for supporting my efforts advancing the world’s knowledge on such a critical topic: if I was able to compare 23 models and not 8, that’s because I got a grant for it as a community member. Volunteering time + sharing financial resources = 💖</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zP0gxXX3Sze1Djk9VzxIpA.jpeg" /><figcaption>All the compared models next to each other</figcaption></figure><h4>23 models is not 23 brands</h4><p>So, what do we have on trial? The 3M common models: <em>Post-It, Post-It Mini, Post-It Super Sticky</em> (normal and <em>XL</em>),<em> Post-It Recycled </em>and the<em> Tesa </em>global competitor.</p><p>Some more local models. Common in France: <em>Lyreco, Office Depot Recycled, Apli Classic, Apli Néon</em>. I also discovered Germany produces some:<em> Idena, Kyome Haftnotizklotz</em>,<em> Paperfoxx</em>. And obviously, China and Taiwan: <em>AB.M Idea, Cosanter JinXin, Stick’N Pop-up notes, Tiger Block.</em></p><p>Finally, a few almost “noname” from big EU distributors: <em>5 stars</em>,<em> Q-Connect</em>, <em>Snopake</em>, <em>NiceDay</em>, <em>Tartan</em>; and an actual white label: a <em>Red Cross</em>-provided pack that was just lying around and that I included as a comparison element.</p><p>Enough with the context already, let’s make some science out of these pieces of paper! 😁</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jsxzL_BBPDwpS4wNAH1RLw.jpeg" /><figcaption>Before unwrapping.</figcaption></figure><h3>Stickiness</h3><p>How many times have you moved a sticky note around, clustered it with some others, only to find it on the ground when you go back to the whiteboard after lunch break?</p><blockquote>Since spatial position usually encodes information (or at least it should <a href="https://speakerdeck.com/mattisg/entangling-artefacts">because human brains will believe it does anyway</a>), falling stickies means lost information.</blockquote><h4>Experimental process</h4><p>I set up two different testing grounds to simulate common situations.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fr5Iv8EL0tvo%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dr5Iv8EL0tvo&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fr5Iv8EL0tvo%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/469a243d6b27cf0ad89afa03e120e09d/href">https://medium.com/media/469a243d6b27cf0ad89afa03e120e09d/href</a></iframe><p>One on a South-facing modern glass window: perfectly flat surface exposed to direct sunlight (Paris in August, 15–32°C, 160h of cumulated sunlight), ventilation holes and window next to it always open to ensure <a href="https://youtu.be/sSRAmZhbHlA">humidity variation</a>. Since humidity is more controlled in offices, this simulates a <strong>whiteboard </strong>over a longer time</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/804/1*lYasqEpLOPX4cUA1rPMyFQ.png" /><figcaption>Yellow is hours of sunlight, red is max temperature, blue is min temperature. <a href="http://www.meteofrance.com/climat/france/paris/75114001/releves#">Data from MeteoFrance</a>.</figcaption></figure><p>The other trial surface was a painted wall in the same room: same humidity variations but never exposed to direct sunlight, and slightly grainier. This simulates an <strong>office wall</strong> over a longer time.</p><p>I obviously stuck notes following <a href="https://www.youtube.com/watch?v=gdBDOaZSZBg&amp;list=PLhQ0uCLmau9E7LCbA4LD47z-fcJxAlr8U&amp;index=6">best practice</a>, ensured equal force was applied on all of them, and thoroughly cleaned both surfaces beforehand.</p><h4>Discussion</h4><p>All stickies were much more stable than expected. None fell over 30 days! I suspect this is the consequence of using my full weight to stick them, as that was the only way to ensure consistent force.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fg5H4NfyBljo%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dg5H4NfyBljo&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fg5H4NfyBljo%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="640" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/b6abd326ad58a2e207e2f20b6d733431/href">https://medium.com/media/b6abd326ad58a2e207e2f20b6d733431/href</a></iframe><p><a href="https://twitter.com/Vavoor/status/895160823022333952">Another hypothesis</a> was that they went straight from pack to wall and the glue was more efficient this way, but I ran a parallel experiment over two hours trying to measure after how many sticking-unsticking cycles the glue would run out, and it turns out the limit is above 150. I could measure no difference in falling tendencies. Even with a fan under the note for 5 minutes. Crazy, right.</p><p>It also appears that there is no perceivable change in stickiness between 20 days and 30 days. In a similar fashion, the window surface increases overall stickiness, but products fared comparably on both surfaces. Put otherwise, <strong>either the glue is poor right from the start and for all surfaces, or it has good chances to stay for long</strong>.</p><h4>Results</h4><p>Since I received each model on a different date, I could not have 30 days of exposure for all of them. However, since days 20 and 30 had no difference in measured stickiness for models that were exposed that long, and since wall surface simply made it easier to spot glue failure without changing tendencies, it is sufficient to observe the wall at day 20.</p><p>I rated each model based on the following scale:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/643/1*hGBMKplWL3mdEyirc6uP_A.png" /><figcaption>Adherence distribution</figcaption></figure><ul><li>5 = perfect adherence</li><li>4 = tooting visible on detailed inspection</li><li>3 = tooting visible 1 meter afar</li><li>2 = not readable from all angles</li><li>1 = about to fall</li></ul><p>You should thus prefer one of the 8 products with the best glue, especially if you intend to leave your stickies on the wall for a long time: the 3M <em>Post-It Recycled</em> or <em>Super Sticky</em> models, the <em>Cosanter JinXin, Lyreco</em>, <em>Q-Connect</em>, <em>Stick’N</em> or <em>Tartan</em>. Other brands can still have good sticking power, but I won’t list all of them inline. You can <a href="https://www.icloud.com/numbers/0NlGpPJ2vXY5uZzwIczCyscTA">look at the data</a> if you’re curious 😉</p><h3>Ink penetration</h3><p>Write down your idea on that sticky note with a marker (<a href="https://www.youtube.com/watch?v=m6xkXNd1VLI&amp;list=PLhQ0uCLmau9E7LCbA4LD47z-fcJxAlr8U&amp;index=3">properly</a>), remove it from the pack and stick it on the wall (<a href="https://www.youtube.com/watch?v=gdBDOaZSZBg&amp;list=PLhQ0uCLmau9E7LCbA4LD47z-fcJxAlr8U&amp;index=6">properly</a>), correct a typo on it, realise that you stained both the wall and the second sticky. Rage.</p><blockquote>Sticky notes that let ink go through are at best environmental and financial waste, at worst used as an unreadable medium by participants.</blockquote><h4>Experimental process</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MHk0SUk4ndWJHcBc3nvh7Q.jpeg" /></figure><p>Markers have different inks and solvents, ranging from very light (like a pen) to very strong (like a paint marker). In order to represent that range, I <a href="https://youtu.be/XkB4WJv1lH0">taped</a> 6 of them together, ensured their tips were perfectly aligned, and pushed that tool on the stickies packs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2ffW8mpvbOD0WhEyzQNSHA.jpeg" /></figure><p>I hit stickies packs strongly with that tool from about 10 cm above, and kept pushing down for 5 seconds.</p><p>I then counted the visible dots on each following note. In order to reduce variation, I repeated that process 3 times for each product.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a9ZcpWJO2LjxKCyhHpDjGQ.jpeg" /></figure><p>I counted dots on each note, and took the mean of the three runs. For example, for the series on the left, each column is a run. The first note (obviously) has all 6 tips visible. Two tips always leak on the sticky that comes next on the pack, thus mean is 2 for note 2. The third note has one run with 0 dots, one with 1 dot, one with 2 dots. The mean is thus (0+1+2)/3 = 1.</p><p>I counted means this way until note 6 (no model leaked beyond that point, which is already considerable). It is worth noting, however, that leaking in this stress-test trial does not mean leakage would be systematically observed in standard usage: the pressure applied is much stronger and much longer.</p><h4>Results</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/514/1*cxFXa55NYrF4_sF0lNDnxw.png" /><figcaption>Distribution of mean ink penetration by note depth. Colour intensity represents number of models leaking that number of dots on that note.</figcaption></figure><p>Even with the importance of the applied stress, the range of possibilities is very wide, as shown in the distribution graph. Two models never leaked beyond note 2 (<em>3M Recycled Post-It</em> and <em>Stick’N Pop-up Notes</em>), while some kept leaking until note 6 (notably <em>NiceDay</em>, which consistently let one marker type through).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/981/1*XcfzHnMmJyZyuCkWOIu6cQ.png" /><figcaption>Green means no leakage, yellow means lower-than-median leakage, red means above or equal leakage.</figcaption></figure><p>In order to ease comparison across dimensions, I computed an index for the “ink penetration” dimension. This index normalises the means for all notes to a mark from 0 to 5. <a href="https://www.icloud.com/numbers/0NlGpPJ2vXY5uZzwIczCyscTA">Look at the data</a> if you want the exact formula.</p><h3>Ease of opening</h3><p>The brainstorming session starts. One participant runs out of sticky notes. As you’re well prepared, you take a new pack out of your bag.</p><blockquote>“Does anyone have scissors?” Nope. Two minutes of fighting with plastic wrapping ensue. It was a fun break, but the participant lost focus and forgot their idea.</blockquote><h4>Experimental process</h4><p>Purely human-based and biaised. I rated each pack from 1 (impossible to open without a cutting utensil) to 5 (excellent affordance and usage of the self-opening device). I did not get all models as new packs, some are consequently not rated in this dimension.</p><h4>Results</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/555/1*JLqOaPh41TpwkCMCxJu6FA.png" /><figcaption>Distribution of ease of opening the pack</figcaption></figure><p>Packaging does not seem to be an integral part of product design for most brands. The ones that are awarded a 5 clearly show a precut easy opening (<em>3M Post-it Recycled</em>), or a ribbon to tear the plastic wrapping : <em>Cosanter, Idena, Q-Connect, Paperfoxx </em>and <em>Tesa</em>.</p><p>The worst rankings were for a few brands seemingly taking pleasure in using the most shreddable plastic wrapping with strong glue (<em>Snopake</em>), or in suggesting ways to open that are even less productive than tearing the packaging apart (<em>Stick’N Pop-up notes</em>).</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FowWMB6In9LQ%3Fstart%3D40%26feature%3Doembed%26start%3D40&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DowWMB6In9LQ&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FowWMB6In9LQ%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/2f571d5fef14f10cf8383ff08fa8e58e/href">https://medium.com/media/2f571d5fef14f10cf8383ff08fa8e58e/href</a></iframe><p>The case of 3M plastic-wrapped products (i.e. all but <em>Recycled Post-it Notes</em>) is interesting. Indeed, the precutting does work very well <em>if you know how to use it</em> (see video). Functionality is good, but discoverability is so bad this package-opening tutorial is the most popular video in the series!</p><h3>Price</h3><h4>Experimental process</h4><p>Very simple: divide the price of a pack by the number of stickies in it 😉 I did not take transportation costs into account as they are not linear, and can be offset if you buy other products in the same order.</p><h4>Discussion</h4><p>Prices can vary over time due to dynamic pricing in online retailers. It is also possible that you will get different prices depending on the retailer, or get a sale or a discount. I minimised this effect by recording only the standard prices of a single retailer (Amazon), over a single week. The main aim being to compare prices relatively to each other, I believe this allows for a fair evaluation of the incurred costs.</p><h4>Results</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/123/1*fJ4SQTDAxarDco3QmVfSmw.png" /><figcaption>Distribution of the prices of a single sticky note</figcaption></figure><p>Prices vary wildly from the cheapest Chinese product with notes worth less than a cent to the fanciest at a sixth of an euro a piece.</p><p>I did not buy the few even more expensive products for this study, as they are too expensive to be considered in a professional usage anyway.</p><p>The median price is 0.02 €, and anything more expensive than 0.05 € a note is clearly a ripoff.</p><p>I normalised these results on a 0 to 5 scale simply by linearly rating the price from most expensive (0) to least expensive (5).</p><h3>Smell</h3><p>A displeasing smell is not only a matter of comfort. Smell indeed mostly comes from solvent, which means it can be a sign of harmful chemicals used in the adhesive that you’re going to put your hands on and let evaporate in your working space for days.</p><blockquote>Beyond the (admittedly limited) risk to health of small doses, an uncomfortable smell on pack opening is always a distraction to the group, and can induce headaches in the most sensitive participants.</blockquote><h4>Experimental process</h4><p>Once again, easy and biased. If I don’t smell anything when I open the pack, I award 5. The worse and longer it smells, the less points.</p><h4>Results</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/603/1*SRejfUj0YQ2fmH9YYlKWMw.png" /><figcaption>Distribution of (lack of) smell</figcaption></figure><p>The vast majority of products have no smell at all when opened. But the worst, <em>Tiger Block</em>, had me breathing through the mouth for about 10 minutes.</p><p>The case of the <em>Tartan</em> brand is interesting. Indeed, it disappoints in the smell dimension, while it scored well everywhere else. Concurrently, while this brand is basically nonexistent on the internet, its production address is… the same as 3M’s! I thus suspected this product to be a “noname” cheaper version of the 3M Post-it Notes, but at least the solvent used is different.</p><h3>Sustainability</h3><p>We know what stickies are for, but we don’t often think about where they come from and where they go when they die 😉</p><blockquote>As agile or collective intelligence professionals, wouldn’t it be fitting that we ensure the sustainability of the tools we use to help others reach sustainable efficiency?</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wHvW4_Ydp2omR2FwvpB6Sw.jpeg" /><figcaption>What kind of waste does one brainstorming session create?</figcaption></figure><h4>Experimental process</h4><p>I researched each brand for its certifications and policies, and awarded points on the following elements:</p><ul><li>2.5 points for paper: recycled and FSC certified sustainable sources give points, the higher proportion the better.</li><li>Half a point for both sustainable adhesive sources and solvent-less glue.</li><li>1 point for carbon footprint: 0.75 for producing in the EU (if you’re on another continent, this should be awarded to producers in your vicinity); 0.25 for assessing lifecycle carbon footprint.</li><li>Half a point for both recyclable packaging and guaranteed recyclability.</li></ul><h4>Results</h4><p>Three products stands out. The <em>3M Recycled Post-it</em> reaches a high score thanks to its 100% recycled paper and recycled cardboard packaging. The only other brand with a recycled cardboard packaging, Stick’N, lost the associated points because said package also contained a glued sponge (!) used to lift up the notes from beneath. 3M products also get a further bonus thanks to their <a href="http://multimedia.3m.com/mws/media/1033586O/pulp-and-paper-sourcing-policy.pdf">Pulp and Paper Sourcing Policy</a>, explicitly asking since 2015 all 3M provider chain to respect not only the natural enviroment, but also the rights of workers, indigenous people and rural communities.</p><p>An equivalent socially responsible policy is claimed by <a href="http://www.globalnotes.de/Policy-Statement/">GlobalNotes</a>, the German manufacturer for both <em>Paperfoxx</em> and <a href="http://geni.us/6GML7gB"><em>Tesa</em></a>. They are also obviously advantaged for producing in the EU (if you’re in the US, 3M is thus even better as they produce in Kentucky). But beyond that, though the notes are not made from 100% recycled paper, their durable paper sourcing is certified by the <a href="https://ic.fsc.org/en/choosing-fsc/fsc-labels">Forest Stewardship Council</a>. These brands don’t benefit from their plastic wrapping, but compensate by using <a href="http://www.globalnotes.de/Water-based-glue/Enviroment-Water-based-glue">water-based solvents</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/249/1*3ARghvgyC2waKdsruAKE_g.png" /><figcaption>Distribution of sustainability indexes</figcaption></figure><p>This is what the German brand <em>Idena</em> lacks, along with a socially responsible sourcing policy.</p><p>The 3M <em>Super Sticky</em> models follow, thanks to their durable adhesive (<a href="https://www.post-it.com/3M/en_US/post-it/contact-us/about-us/">67% by weight</a> coming from annual plants, as opposed to chemical production). The attention to adhesive also gives a boost to Apli, a Spanish brand not using solvents in its glue.</p><p>Even the least specific 3M models get acceptable ratings thanks in part to their incorporation of recycled material (0 to 30% depending on the colour), but also because of the care the brand takes in guaranteeing the recyclability of its products: the adhesive and inks have been tested for dissolving in the dye-removal part of the recycling process. This is the only brand in this study claiming to have conducted such tests.</p><p>It is saddening to see that most brands do not give any message about sustainability in their products. It is also worth noting that no product reaches even 4 out of 5 points in this dimension: no sticky note seems to have the “holy grail” <a href="https://ic.fsc.org/en/choosing-fsc/fsc-labels">FSC 100%</a> (or 100% recycled) certification for their papers, and even 3M does not (yet?) incorporate its Super Sticky adhesive innovations in all its products.</p><h3>And the winner is…</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mqvbuZILss86Sluiz0AHDw.jpeg" /></figure><p>The <a href="http://geni.us/3MRecycled"><strong>3M Recycled Post-it</strong></a>!</p><p>Classic brand, but a somehow unexpected model. I did not expect the Recycled version to be not only one of the most sustainable, but also so much better at stickiness than its “standard” counterpart, and to outperform all other models in the ink penetration test. Maybe its heavier fabric is to thank.</p><p><strong>I am thrilled to conclude that the most efficient and cost-effective choice is also the most sustainable!</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lYIioVp5MExNPaxF6mEDLw.png" /></figure><h3>Parting words</h3><p>I hope this piece of research was useful and brought some smiles along the way. Beyond producing “stupid science”, having fun and taking agile way too seriously, I believe we have a strong responsibility in educating as “change agents”, “coaches” or whatever we like to call ourselves these days. If you consider switching to this model, even if on pure efficiency or cost-effectiveness ground, use it as an opportunity for a discussion on sustainability with yoursef, your clients, participants and fellow coaches.</p><blockquote>Through questioning our practices thoroughly, we might well discover other occasions on which the best thing to do for ourselves is also the best for the rest of the world.</blockquote><p>Please challenge or support this piece based on your experience! 🙂 Science is only reliable when it is reproducible, which is why I tried to describe as precisely as possible everything I did and opened the source data.</p><p>If you enjoyed this article, <strong>clapping, sharing and subscribing</strong> makes me feel good and helps the content reach other readers 🙂⬇️</p><p>Oh, and please let me know if you do change habits. If this work has any impact, I’d love to know! 😄</p><h4>Recommended reads</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*MLX_RWRP2zCQ4nxeuQxoWw.png" /><figcaption>“<a href="https://medium.com/graphicfacilitation/all-you-need-to-know-about-whiteboard-markers-2a25d1249911">All you need to know about whiteboard markers</a>” by <a href="https://medium.com/graphicfacilitation">Yuri Malishenko</a>.</figcaption></figure><p><a href="https://medium.com/graphicfacilitation">Yuri Malishenko</a> compared 9 brands of <a href="https://medium.com/graphicfacilitation/all-you-need-to-know-about-whiteboard-markers-2a25d1249911"><strong>whiteboard markers</strong></a> over ink quality, erasability, shape, smell, availability, eco-friendliness…</p><p><em>Know of another crazy comparison? Let me know!</em></p><h4>Acknowledgements</h4><p><em>This piece of stupid science would not have been the same without AgileFrance, Nicolas, Thomas, Alban, Raphaël, Yannick, Colin, Guillaume, Julien, Laurent and Anouchka. Thank you all so much for supporting me in such endeavours</em> 💝</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fc9b97d88503" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I’ve read all United Nations working papers on autonomous weapons, and you should feel concerned]]></title>
            <link>https://medium.com/@MattiSG/ive-read-all-united-nations-working-papers-on-autonomous-weapons-and-you-should-feel-concerned-59e2c32ac6da?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/59e2c32ac6da</guid>
            <category><![CDATA[autonomous-weapons]]></category>
            <category><![CDATA[robotics]]></category>
            <category><![CDATA[united-nations]]></category>
            <category><![CDATA[personal-responsibility]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Wed, 15 Nov 2017 15:01:01 GMT</pubDate>
            <atom:updated>2017-11-17T16:38:46.366Z</atom:updated>
            <cc:license>https://creativecommons.org/licenses/by-nc-sa/4.0/</cc:license>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*71q6FKNfpAbz4Ck530W2jw.png" /><figcaption><a href="https://vimeo.com/209734673">The Punishment</a>, art installation by <a href="http://www.filipevilasboas.com/">Filipe Vilas-Boas</a> &amp; Paul Coudamy, Futur en Seine 2017</figcaption></figure><p>This week, the United Nation’s Group of Governmental Experts on Lethal Autonomous Weapons Systems (LAWS) meet. This post is a recap of the context and content of the discussion.</p><h3>Here is the short-term future we want to avoid</h3><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F9CO6M2HsoIA%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D9CO6M2HsoIA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F9CO6M2HsoIA%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/39da0f5b4c5103c246ab4f8059220bd3/href">https://medium.com/media/39da0f5b4c5103c246ab4f8059220bd3/href</a></iframe><p>Worries about autonomous weapons were already outlined in August in an <a href="https://futureoflife.org/autonomous-weapons-open-letter-2017">open letter</a> signed by actors such as Stephen Hawking and Elon Musk.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fc277ynyRPgs%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dc277ynyRPgs&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fc277ynyRPgs%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/4d9257dfcfaab21fba915ec7fb2d73ed/href">https://medium.com/media/4d9257dfcfaab21fba915ec7fb2d73ed/href</a></iframe><p>Basically, you should not worry about Terminator. You should worry about current technology being weaponised, States deploying it on the field, and it being misused and leaked to more rogue actors.</p><p>Think chemical weapons giving a strong advantage when used for the first time, then all States stockpiling death gas, and years later sarin killing in the metro, or dictatorships using chlorine against its own people.</p><h3>Summary of States positions</h3><p>Feeling concern about this topic, <strong>I’ve read through all of the member states published working papers.</strong> Below is my own, biased, summary of their positions on autonomous weapons, in a (new, 280-chars) tweet each.</p><p><a href="http://www.unmultimedia.org/avlibrary/asset/2034/2034445/">GENEVA / LETHAL AUTONOMOUS WEAPONS</a></p><p>You can check the original papers for yourself on the <a href="https://www.unog.ch/80256EE600585943/(httpPages)/F027DAA4966EB9C7C12580CD0039D7B5?OpenDocument">United Nations’ working group page</a>. Most documents are about 3 pages long (with one much longer exception, see below).</p><h4>US (working paper 6)</h4><p>Law that should be applied to autonomous weapons is that which applies to States. States have responsibility on them. Since autonomy is software, <strong>responsibility could be extended to development</strong> (not only use), but “based on info available at that time”.</p><h4>Switzerland &amp; Netherlands</h4><p><strong>International law already explained everything</strong> [12 pages of legal review dating back to 1863 follow]. Basically, each State has the responsibility to review if their weapons and their use of them are legal. We’re done here.</p><h4>US <em>(working paper 7)</em></h4><p>We lack a proper definition. Here’s ours, from DoD. Also, please stop with the academic stuff and sci-fi fearmongering. <strong>We need to focus on the intent to design good international law</strong>, not on the intrinsic characteristics of autonomous weapons.</p><h4>France &amp; Germany</h4><p>We lack a proper definition. We make a political statement: “<strong>humans should continue to be able to make ultimate decisions</strong> with regard to the use of lethal force and should continue to exert sufficient control over lethal weapons”.</p><h4>Netherlands</h4><p>Humans still decide of targets, area and time for attacks, and assess damage. That is “meaningful control” and <strong>current law applies</strong>. Why on earth would any state want to develop fully autonomous weapons? We reject that but don’t ask for a moratorium: stop what?</p><h4>Russia</h4><p>We lack a proper definition. Everybody agrees losing “meaningful control” is unacceptable, but “meaningful” is politics, and ain’t nobody got time for that here. <strong>We need to have actual examples of autonomous weapons</strong> before being able to qualify what to ban.</p><h4>Belgium</h4><p>About that definition: actually, it’s not so much about what autonomous weapons do or don’t. <strong>It’s about the level of uncertainty</strong>. If at some point you can’t predict exactly its behaviour yet it can kill, then we need to define law on hacking, disabling, zoning it…</p><h4>Switzerland</h4><p>Let’s define “autonomous weapons” as systems that <em>could</em> breach international humanitarian law of war without explicit human command. Then, we just reaffirm, translate and <strong>disambiguate the principles of law in the case of robots through standards and policies</strong>.</p><h4>Non-aligned countries</h4><p>Yeah, we should really keep discussing about all that stuff, in a transparent and participatory way. The discussions we had already were pretty good.</p><h3>What that means</h3><h4><strong>Not a single one of the parties is in favour of a ban</strong></h4><p>They certainly do have a point that a proper definition of what to ban needs to be found before a ban can be decided, but the gap in addressing the concerns of scientists, producers and users of AI is impressive. In many papers, the amount of legalese shows just how much only jurists have been involved in the topic. The main argument beyond the lack of a definition is the impression that banning autonomous weapons would hinder civilian AI development.</p><p>I feel worried of the —certainly pragmatic— point that both Russia and the USA consider nothing fundamental changes until they have observed an actual autonomous weapon kill people in a problematic way. In the EU, there seems to be an agreement on the need for a political statement and a working definition as a starting point.</p><h4>What can be done</h4><p>At that point, it is pretty hard for an outsider to guess whether the lack of consideration for the scientific and industrial community raising alarms is due to good military lobbying, or genuine lack of understanding of the speed and versatility of the digital industry. In either case, based on my limited understanding of the processes in place, it seems impossible to achieve a ban before the technology is actually carried out.</p><p>What can still be done, though, is:</p><ol><li>Stay informed, join the #BanKillerRobots petition and take action on <a href="http://autonomousweapons.org/">autonomousweapons.org</a>.</li><li>Pressure your government to get involved in the topic at the UN level.</li><li>Request national legislation to preemptively ban producing, buying and selling autonomous weapons.</li><li>Convince your friends who work in the (para)military industry to put their skills to a better use.</li><li>If you’re in the AI field, bring ethical considerations to the forefront in your workplace, especially considering potential reuses of your work. Calls to a higher authority to prevent bad stuff that could be done with our work do not free us from personal responsibility. Especially when it becomes so clear that we can’t rely on others to behave like we’d expect them to.</li><li>And obviously, share these elements within your different circles.</li></ol><h3>Epilogue</h3><p>I am curious about one thing. Several NGOs have affiliated themselves with the “<a href="http://www.stopkillerrobots.org/">Campaign to Stop Killer Robots</a>”. Yet, I couldn’t find which joined the UN working group, while it explicitly states:</p><blockquote>Who can participate in a CCW GGE? A CCW GGE is not the same as a GGE established under the General Assembly. It is open to all High Contracting Parties and non-States parties, international organisations and non-governmental organisations.</blockquote><p>I mean, I’m sure making scary videos, nice websites and catchy hashtags gets you more donations and subscribers than participating in boring meetings in Geneva, but considering how important the topic is, I am sure they would <em>also</em> love to be where the decisions will be made, right?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*y6xLykgrzYVqXIvHWPJDcQ.jpeg" /><figcaption><a href="https://vimeo.com/209734673">The Punishment</a>, art installation by <a href="http://www.filipevilasboas.com/">Filipe Vilas-Boas</a> &amp; Paul Coudamy, Futur en Seine 2017.</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=59e2c32ac6da" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When should I use a double-bang?]]></title>
            <link>https://medium.com/@MattiSG/when-should-i-use-a-double-bang-1672bf8e677f?source=rss-166622c8ac77------2</link>
            <guid isPermaLink="false">https://medium.com/p/1672bf8e677f</guid>
            <category><![CDATA[best-practices]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[javascript]]></category>
            <dc:creator><![CDATA[Matti Schneider]]></dc:creator>
            <pubDate>Sun, 04 Oct 2015 09:20:49 GMT</pubDate>
            <atom:updated>2015-10-04T09:20:49.849Z</atom:updated>
            <cc:license>https://creativecommons.org/licenses/by-nc-sa/4.0/</cc:license>
            <content:encoded><![CDATA[<p>I recently reviewed code from a coworker, in which I suggested removing the famous double-bang (`!!`) construct from the return value of a function named `hasFeature()`. The following discussion ensued.</p><p><em>NB: the double-bang construct allows one to coerce (“cast”) a value to a Boolean value, through double negation. It is not an operator in itself, but the repetition of the unary `!` operator, which negates the truthiness of its operand.</em></p><pre>&#39;truthy&#39;; // &#39;truthy&#39;<br>! &#39;truthy&#39;; // false<br>!! &#39;truthy&#39;; // ! false ⇒ true</pre><blockquote>So in JS, as long as the code is fine with it, you shouldn’t cast ?</blockquote><p>Well, it is a matter of consistency, of course. If your codebase relies on `=== true`, you should obviously cast. On the other hand however, if you do take advantage of truthies and falsies and write `if (hasFeature())`, what’s the point of casting it? Wasting (admittedly very little) CPU and memory?</p><p>As soon as you understand that <em>everything</em> in JS can be evaluated to a boolean value, there is little need for actual `Boolean` types, except in domain representations and specific cases where you would have a different treatment for different falsies (such as `undefined` meaning the value has not yet been initialized, as opposed to a `false` value being a domain value).</p><p>This doesn’t mean you should <em>avoid</em> `Boolean` values. It simply means `return myVal` has <em>exactly</em> the same truthiness value as `return !! myVal`. The fact that the function name suggests it returns a `true`/`false` value actually suggests it is very safe to return truthies/falsies, as its return value would be only used in a comparison or a condition.</p><p>Then again, consistency above all. But if you want to return a `Boolean`, I’d go with a clear, readable cast with `Boolean(result)` where even a JS newbie understands what goes on rather than the cryptic `!! result`.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1672bf8e677f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>