<?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[Computable Blog - Medium]]></title>
        <description><![CDATA[Computable Blog - Medium]]></description>
        <link>https://medium.com/computable-blog?source=rss----638ab2c4f2bb---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Computable Blog - Medium</title>
            <link>https://medium.com/computable-blog?source=rss----638ab2c4f2bb---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 03 Jun 2020 10:10:45 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/computable-blog" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com" rel="hub"/>
        <item>
            <title><![CDATA[Why we won’t apply for PPP]]></title>
            <link>https://medium.com/computable-blog/why-we-wont-apply-for-ppp-60ccde1c9ec2?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/60ccde1c9ec2</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[venture-capital]]></category>
            <dc:creator><![CDATA[Roger Chen]]></dc:creator>
            <pubDate>Tue, 07 Apr 2020 19:56:07 GMT</pubDate>
            <atom:updated>2020-04-14T22:31:39.500Z</atom:updated>
            <content:encoded><![CDATA[<h4>It wasn’t meant for us. Is it meant for you?</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/813/0*9DOEs06yL0S3m9Ox.JPG" /><figcaption>JP Morgan Chase Institute report on small businesses</figcaption></figure><p>Take a moment to think about the restaurants, coffee shops, and neighborhood stores that you frequent. Most small businesses without venture capital support have a month or less of cash reserves. By the time the coronavirus dust settles, there’s a good chance that your usual spots won’t exist anymore. Some of those shops might even be owned by family members or friends. My mother, for example, called the other day to let me know that her restaurant has a 50–50 chance of surviving all this.</p><p>When the Paycheck Protection Program (PPP) was recently announced, it was intended to preserve jobs by giving lifelines to small businesses that might otherwise shutter. It’s currently becoming a money grab instead, and we’ve decided we won’t be part of it. I don’t expect other startup founders to do the same. Startups are small businesses too, and some founders truly need that lifeline. Decisions like these depend on circumstance, so the best I can do is to share ours with the hope that it helps others reach their own conclusions.</p><p>Like most startup founders, I feel a relentless urge to constantly improve the financial position of our company. My instinctive reaction when I first heard that PPP loans could be forgiven: ”Everyone will go after this money, so we should too.” Yet, something just didn’t feel right. It became clear why after I listed our reasons for and against pursuing PPP.</p><p>Why we should apply:</p><ul><li>We may increase our runway by 1 month plus change.</li><li>My team and investors might question why I didn’t apply.</li></ul><p>Why we shouldn’t apply:</p><ul><li>We have over a year of runway to achieve milestones with our current staff and payroll. If we can’t make that work, then we probably haven’t earned the right to stay in business.</li><li>By taking the PPP loan, we are possibly snuffing out other companies — ones with clearly more need than us and for whom PPP was intended in spirit.</li><li>Everyone has been impacted including us, but we are among those less affected. We are (1) pre-revenue and focused on early product development, (2) an all-remote team working on software technology, and (3) able to continue commercial development and activity online</li><li>It’s not a given that PPP will work out for us anyway. Expected value is probably less than a month of runway.</li><li>We incur significant legal and social risk even if we apply in good faith given the high uncertainty of present guidelines. It’s not “free money.”</li></ul><p>Maybe my mind was already made up when I wrote these lists, but they certainly made things clearer for me. The reasons for applying just aren’t good enough to justify the risks we incur as a company and the costs we pass on to the broader business community.</p><p>I’ll reiterate here that our decision works for us and by no means will it work for everyone else. Even well-funded startups should seriously consider PPP if COVID-19 truly presents existential risk. For example, I know of some venture-backed companies that saw double-digit 2020 sales projections evaporate to zero. Meanwhile some tech firms in particularly hard hit categories like travel or retail are hurting just as much as bootstrapped small businesses. I just urge that if you can actually make do, then please try to make do, because there are plenty of others that simply can’t right now. As an entrepreneur, I’m all for jostling for market advantages, but that just isn’t what this program is about.</p><p>I’m sure many won’t agree with our views, and that’s fine with me. When we’re all facing an invisible enemy, it’s moot to debate who’s right or wrong. We can only stand by our values, and this is just how we chose to express ours. Business leaders have a responsibility to look after their companies to be sure. In times of crisis, we should look after our communities too.</p><p>Stay safe out there and be well.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=60ccde1c9ec2" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/why-we-wont-apply-for-ppp-60ccde1c9ec2">Why we won’t apply for PPP</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How we killed our own project because of privacy concerns]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://medium.com/computable-blog/how-we-killed-our-own-project-because-of-privacy-concerns-ed61d97cf000?source=rss----638ab2c4f2bb---4"><img src="https://cdn-images-1.medium.com/max/2600/1*M73fKjqNANxZcOGdguGmUw.png" width="2880"></a></p><p class="medium-feed-snippet">We built a data collective for drivers. Our prototype enabled us to test abstract ideas for a product; it also pushed back against our&#x2026;</p><p class="medium-feed-link"><a href="https://medium.com/computable-blog/how-we-killed-our-own-project-because-of-privacy-concerns-ed61d97cf000?source=rss----638ab2c4f2bb---4">Continue reading on Computable Blog »</a></p></div>]]></description>
            <link>https://medium.com/computable-blog/how-we-killed-our-own-project-because-of-privacy-concerns-ed61d97cf000?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/ed61d97cf000</guid>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[uber]]></category>
            <category><![CDATA[lyft]]></category>
            <category><![CDATA[privacy]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Reid Williams]]></dc:creator>
            <pubDate>Wed, 04 Mar 2020 04:02:26 GMT</pubDate>
            <atom:updated>2020-03-04T04:02:26.648Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[On-chain data cooperatives reach mainnet]]></title>
            <link>https://medium.com/computable-blog/on-chain-data-cooperatives-reach-mainnet-913a986bdfba?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/913a986bdfba</guid>
            <category><![CDATA[ethereum]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Roger Chen]]></dc:creator>
            <pubDate>Tue, 11 Feb 2020 18:59:41 GMT</pubDate>
            <atom:updated>2020-02-11T18:59:41.496Z</atom:updated>
            <content:encoded><![CDATA[<h4>Introducing a new protocol for community-governed datasets</h4><p><strong><em>Imagine this:</em></strong> George, a machine learning engineer, sinks into his seat, takes a sip of coffee, and connects to his favorite coffee shop’s WiFi network (via VPN of course). Knees weak, palms sweaty, he can’t wait to explore a new idea he recently conceived. MRI images alone provide an incomplete picture of how the brain works, but could multimodal deep learning across other types of data characterizing neural activity — such as EEG, MEG, fMRI, etc — unlock interesting new insights? He didn’t know, but he sure was eager to find out.</p><ul><li>Instant access to a cloud computing instance — check.</li><li>Finding open source machine learning models to adapt — easy.</li><li>Getting said data to experiment with — stumped.</li></ul><p>A couple of years ago, I had this exact experience, and many others I know have had similar ones as well. It got me thinking. If we can unlock the world’s data, how much more innovation, progress, and productivity can we drive by unlocking the innovators themselves? While not the only reason, this question heavily influenced the core thesis behind why we started <a href="https://computable.io/">Computable</a> in the first place.</p><h3><strong>Data has a governance problem</strong></h3><p>The Internet simplified access to high performance and affordable computing resources. Open source software fostered a culture of sharing the latest advances in algorithms. Yet, neither really solved the fact that data remains in silos. Which makes sense because data introduces a wholly new set of unique challenges that open source culture and the Internet can’t solve alone. For one, not all data are created equal; namely, the usefulness of data hinges on standardization, quality control, and labeling. Furthermore, those who possess valuable data are often those who feel most incentivized to hoard it, even if that value can be better derived by sharing with others. (For example, the scientific community’s growing <a href="https://www.nature.com/articles/s41467-018-05227-z">willingness to share data</a> has already started generating important novel findings that might otherwise have been discovered years later.) Finally, technology-driven capitalism has raised serious societal questions about how to best <a href="https://www.nytimes.com/interactive/2018/12/10/business/location-data-privacy-apps.html">respect user consent and privacy</a>. In a nutshell, <em>data has a serious governance problem</em>. We need a more cooperative structure that incentivizes data sharing while also fairly distributing value and control.</p><p>Today the Computable team is elated to share a new milestone towards that vision — <strong>Ethereum </strong><a href="https://etherscan.io/address/0x2c5358c9573ad83ceb510982a438edcd5b0e0a81"><strong>mainnet alpha deployment</strong></a><strong> of the first decentralized and on-chain data cooperative.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cd9S_s-eviBtSkMcXICDIA.png" /><figcaption><em>Computable’s values built in code, not just words.</em></figcaption></figure><h3><strong>Introducing on-chain data cooperatives</strong></h3><p>A data cooperative is a collaborative dataset bound to a common use case and supported by a community of participants. An on-chain data cooperative is one that is governed via smart contracts — in this case, the <a href="https://github.com/computablelabs/computable">Computable protocol</a>.</p><p>Smart contracts combine the best attributes of the Internet and open source software to help us reimagine governance. They foster transparency and security by implementing permissionless governance functions in auditable open source code. We’ve seen this already with how projects like <a href="https://makerdao.com/en/">MakerDAO</a> and <a href="https://compound.finance/">Compound Finance</a> are pioneering a new financial technology stack centered on open participation. Anyone can get fair access to the same lending products and vote on the parameters governing them, e.g. stability fees, savings rates.</p><p>The Computable protocol brings similar attributes to the world of data. In particular, any member of a data cooperative can unilaterally opt in or out of sharing, receive fair compensation when their data is used, collaboratively curate the quality of a dataset to boost its collective value, and vote on how that value is distributed via tokens. The result is an egalitarian yet market-driven approach to incentivize data sharing by both producing financial reward and tackling trust issues around data.</p><h3><strong>What’s next</strong></h3><p>Our mainnet alpha is the first deployment of our contracts, but not the last. By construction, a new set of contracts (or a new data cooperative) can and should be launched for each application with a unique focus. After all, governance would and should look differently for a mapping dataset versus a rare disease one — in terms of how data is standardized, curated, and rewarded. Different data cooperatives will likely also coalesce unique social communities around their unique causes. We have already begun market experiments to explore what some of these applications and communities will look like. Ongoing work is also happening to create rich user experiences for community participation — a significant challenge facing decentralized application development in general.</p><p>Stay tuned for more updates. In the meantime:</p><ul><li>Check out the <a href="https://github.com/computablelabs/computable-deploy/blob/master/mainnet/info.txt">deployed contracts</a>.</li><li>Learn how the protocol works <a href="https://computablelabs.github.io/compspec/">here</a>.</li><li>Review our <a href="https://github.com/computablelabs/computable/blob/master/audit/trail_of_bits_audit_8_30_2019.pdf">security audit</a>.</li><li>Read the <a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">whitepaper</a>.</li><li>Questions, comments, feedback? Get in touch via <a href="https://twitter.com/computable_io">Twitter</a>, <a href="https://discordapp.com/invite/5A9G6hd">Discord</a>, or <a href="https://forum.computable.io/">Discourse</a>!</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=913a986bdfba" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/on-chain-data-cooperatives-reach-mainnet-913a986bdfba">On-chain data cooperatives reach mainnet</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Weekly Development Update — November 22nd 2019]]></title>
            <link>https://medium.com/computable-blog/weekly-development-update-november-22nd-2019-e8868344a0e1?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/e8868344a0e1</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Bharath Ramsundar]]></dc:creator>
            <pubDate>Mon, 25 Nov 2019 21:22:24 GMT</pubDate>
            <atom:updated>2019-11-25T21:22:07.757Z</atom:updated>
            <content:encoded><![CDATA[<h3>Weekly Development Update — November 22nd 2019</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/320/0*maWN9FPhZOccPnp9.jpeg" /></figure><p>Hi Computable Community!</p><p>We’ve got a number of exciting updates this week. To catch-up on what we’ve been up to, check out <a href="https://forum.computable.io/t/weekly-development-update-november-15th-2019/100">last week’s update</a>.</p><h3>Last Week’s Updates</h3><p>We made a major push the week previous to get contracts up onto mainnet, so we ended up taking a little time to rest and recover after our big push. This meant this week was a little slower than usual, but helped us build up strength for our big push moving forward.</p><h4>Debugging Mainnet Capi and Frontend</h4><p>We took some time this week to debug our mainnet Capi and frontend deployments. There were a number of subtle interaction and timing errors that we had to iron out as we spent more time testing the system.</p><h3>Next Week’s Priorities</h3><h4>Making Deployments More Robust</h4><p>As we get closer to getting our systems ready for use by more folks, we’re planning on spending a good chunk of time making the systems more robust and ready for broader use. Expect to see more of this coming up soon!</p><h4>Getting Involved</h4><p>We’re excited to grow the Computable community. If you’d like to get more involved, follow us on social media or chat with our devs on Discord. Here are some references to help you get started:</p><p><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></p><p><a href="https://github.com/computablelabs">Github</a></p><p><a href="https://www.computable.io/">Website</a></p><p><a href="https://discord.gg/3mFF8zy">Discord</a></p><p><a href="https://twitter.com/computable_io">Twitter</a></p><p><a href="https://t.me/computablelabs">Telegram</a></p><p><em>Originally published at </em><a href="https://forum.computable.io/t/weekly-development-update-november-22nd-2019/101"><em>https://forum.computable.io</em></a><em> on November 25, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e8868344a0e1" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/weekly-development-update-november-22nd-2019-e8868344a0e1">Weekly Development Update — November 22nd 2019</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Weekly Development Update — November 15th 2019]]></title>
            <link>https://medium.com/computable-blog/weekly-development-update-november-15th-2019-e931c66a6d45?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/e931c66a6d45</guid>
            <category><![CDATA[comptable]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[crypto]]></category>
            <dc:creator><![CDATA[Bharath Ramsundar]]></dc:creator>
            <pubDate>Wed, 20 Nov 2019 20:54:08 GMT</pubDate>
            <atom:updated>2019-11-20T20:53:40.337Z</atom:updated>
            <content:encoded><![CDATA[<h3>Weekly Development Update — November 15th 2019</h3><p>Hi Computable Community!</p><p>We’ve got a number of exciting updates from last week. To catch-up on what we’ve been up to, check out <a href="https://forum.computable.io/t/weekly-development-update-november-8th-2019/99">the previous week’s update</a>.</p><h3>Last Week’s Updates</h3><h4>Contracts shipped to mainnet!</h4><p>Following up from our deployment of contracts to Rinkeby last week, we deployed our first data market contracts to mainnet! This was an exciting milestone for us, since it marks the first time a data market is up and running with live funds. We’re not going to share the address of these contracts just yet, since we’re still doing final testing, but we hope to be able to share more information about the mainnet deployment soon.</p><h4>Using Compas to interact with mainnet contracts</h4><p>One of our worries earlier was that our development tooling might break as we transition from our private and public test networks to the main network. Part of the reason for this has to do with timing. Our test networks are proof-of-authority, not proof-of-work, and tend to confirm transactions more quickly. It seemed possible that differences in timing could lead to tricky errors between testnet and mainnet. However, we were able to boot up Compas and connect it to our mainnet contracts just fine! This was a nice validation for us of our engineering process and the effort we’ve put into building solid tools.</p><h4>Deploying Capi for Rinkeby and Mainnet</h4><p>We worked this week on getting Capi instances up and running to serve the Rinkeby testnet and mainnet contracts. This process worked smoothly for the most part, but required us to clean up our config file handling and add some improvements on how we estimate gas price for mainnet. We’ve pushed these changes in and have deployed a Capi instance for Rinkeby and are almost done deploying for Mainnet as well.</p><p>We’re still focused on working through some timeout issues as well. Sometimes, Capi hangs when talking to Rinkeby/mainnet and we think it might have to do with some of our worker threads timing out. We’ll have more to say on this next week.</p><h4>Deploying frontend for Rinkeby</h4><p>We also improved our deployment infrastructure for our web UI and now have a deployed version of our web UI that talks to the Rinkeby contracts. This version is almost fully functional, but we’re working through some last bugs on the integration here. Once we have these bugs ironed out, we’re going to work on connecting our web frontend to the mainnet contracts.</p><h4>Picture Time</h4><p>We took a good chunk of time this week to improve and polish our web UI handling. Here’s a sneak peek at some new screenshots from the UI. Here’s the page where you can choose to share data</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*8sRn3XlCwZwrvlIQ" /></figure><p>Clicking through, you get to the listing upload page where you can upload an interesting file to share. You can then upload this to the datatrust.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Lg8aBzhvC78jmr2u" /></figure><p>We’ll have more pictures for you next week as we continue refining the web UI!</p><h3>Next Week’s Priorities</h3><h4>Polishing the Rinkeby and Mainnet Deployments</h4><p>We made a push this week to get our first set of contracts out onto mainnet. We want to finish the process of connecting these contracts to Capi and to our web UI. We also have some minor bugs which we need to clean up. We’ll get to these next week.</p><h3>Getting Involved</h3><p>We’re excited to grow the Computable community. If you’d like to get more involved, follow us on social media or chat with our devs on Discord. Here are some references to help you get started:</p><p><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></p><p><a href="https://github.com/computablelabs">Github</a></p><p><a href="https://www.computable.io/">Website</a></p><p><a href="https://discord.gg/3mFF8zy">Discord</a></p><p><a href="https://twitter.com/computable_io">Twitter</a></p><p><a href="https://t.me/computablelabs">Telegram</a></p><p><em>Originally published at </em><a href="https://forum.computable.io/t/weekly-development-update-november-15th-2019/100"><em>https://forum.computable.io</em></a><em> on November 20, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e931c66a6d45" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/weekly-development-update-november-15th-2019-e931c66a6d45">Weekly Development Update — November 15th 2019</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Weekly Development Update — November 8th 2019]]></title>
            <link>https://medium.com/computable-blog/weekly-development-update-november-8th-2019-cbc7b3074718?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/cbc7b3074718</guid>
            <category><![CDATA[computable]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Bharath Ramsundar]]></dc:creator>
            <pubDate>Tue, 12 Nov 2019 03:58:19 GMT</pubDate>
            <atom:updated>2019-11-12T03:58:19.692Z</atom:updated>
            <content:encoded><![CDATA[<h3>Weekly Development Update — November 8th 2019</h3><p>Hi Computable Community!</p><p>We’ve got a number of exciting updates this week. To catch-up on what we’ve been up to, check out <a href="https://forum.computable.io/t/weekly-development-update-november-1st-2019/98">last week’s update</a>. We’re going to continue cross-posting our updates on our forums here and on Medium.</p><h3>Last Week’s Updates</h3><h4>Deployed on Rinkeby Testnet</h4><p>Our biggest update this week is that we now have a version of our contracts deployed onto the Rinkeby testnet! Check out this picture of one of the Parameterizer contract on rinkeby viewed through EtherScan</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Y1eX1Ub2bFNEMTR2" /></figure><p>Getting onto Rinkeby posed some fun challenges. We had to get some Rinkeby ETH from the faucet, which required us to post some Tweets online.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*yQt8Pvuz058MurxL" /></figure><p>We then used Compas to interact with the contracts on Rinkeby. The dashboard worked beautifully, letting us obtain MarketToken within minutes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*KId6MFYMysathK19" /></figure><h4>Deliveries Now Functional!</h4><p>After a good chunk of debugging, we can now perform deliveries through our frontend! This means that it’s possible to purchase data through the frontend and get the delivery of the data once purchased.</p><h4>Support and Withdraw through Frontend Now Works</h4><p>We’ve got the capability to now support/withdraw markets through the frontend.</p><h4>Solidifying Frontend Infrastructure</h4><p>We continued solidifying our frontend infrastructure, working on cleaning out some technical debt.</p><h4>Groundwork to Deploy on Mainnet</h4><p>We’re working on setting up infrastructure for our mainnet deployments now that the tokens are on public testnet.</p><h3>Next Week’s Priorities</h3><h4>Shipping to Mainnet</h4><p>We’re aiming to deploy a version of our contracts onto mainnet to provide an alpha developer playground.</p><h3>Getting Involved</h3><p>We’re excited to grow the Computable community. If you’d like to get more involved, follow us on social media or chat with our devs on Discord. Here are some references to help you get started:</p><p><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></p><p><a href="https://github.com/computablelabs">Github</a></p><p><a href="https://www.computable.io/">Website</a></p><p><a href="https://discord.gg/3mFF8zy">Discord</a></p><p><a href="https://twitter.com/computable_io">Twitter</a></p><p><a href="https://t.me/computablelabs">Telegram</a></p><p><em>Originally published at </em><a href="https://forum.computable.io/t/weekly-development-update-november-8th-2019/99"><em>https://forum.computable.io</em></a><em> on November 12, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cbc7b3074718" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/weekly-development-update-november-8th-2019-cbc7b3074718">Weekly Development Update — November 8th 2019</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Weekly Development Update — November 1st 2019]]></title>
            <link>https://medium.com/computable-blog/weekly-development-update-november-1st-2019-569ba1a03d12?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/569ba1a03d12</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[blockchain]]></category>
            <dc:creator><![CDATA[Bharath Ramsundar]]></dc:creator>
            <pubDate>Fri, 01 Nov 2019 22:41:52 GMT</pubDate>
            <atom:updated>2019-11-01T22:41:51.768Z</atom:updated>
            <content:encoded><![CDATA[<h3>Weekly Development Update — November 1st 2019</h3><p>Hi Computable Community!</p><p>We’ve got a number of exciting updates this week. To catch-up on what we’ve been up to, check out <a href="https://forum.computable.io/t/weekly-development-update-october-25th-2019/95">last week’s update</a>. This week, we’re going to expand our distribution of our weekly updates by cross-posting them on both our <a href="https://forum.computable.io/">forums</a> and on our <a href="https://medium.com/computable-blog/">medium</a>.</p><h3>Last Week’s Updates</h3><h3>Computable Terminal UI now at stable v1</h3><p>The <a href="https://github.com/computablelabs/compas">Compas</a> UI is now at a stable v1! We held a small internal tutorial session where we on-boarded our team into using Compas for debugging and internal data market governance purposes. The session was very successful, with attendees able to use the terminal UI to rapidly interact with data markets. Previously we’d used a web-based wallet UI to directly interact with contracts, a method which was buggy and required many clicks to execute simple interactions. But with Compas, we have a light-weight, interactive view of the global market state. Check out this top level control panel:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*joJNXwk0QTrJRKsa" /></figure><p>It’s possible to expand this panel out to gain more information. Recall that a “listing” is a chunk of data in a data market. It’s possible to use Compas to “double click” on a particular listing to gain more information about its current state:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/967/0*4RbfSHMBHBnsMaHv" /></figure><h3>Debugging Data Purchase Through Front End</h3><p>We laid the groundwork for purchasing data through the frontend last week and this week we iterated on trying to get data purchases up and running. We found a few bugs, where we had some subtle error conditions with transactions and file delivery. We’ve made progress towards smoothing these out and managed to get the delivery partially functional through the frontend. We still have a few encoding problems, but we’re hoping we get these cleared out son.</p><h3>Second Pass through Web Frontend to Stabilize and Improve</h3><p>We spent a good chunk of this week cleaning up our web frontend. With the exception of data delivery, all of our core flows are functional, but in MVP form. This week we hammered down a bunch of loose corners and took steps towards making the frontend more robust and usable. We’ll have more to report (hopefully with pictures), next week.</p><h3>Groundwork to Deploy on Public Testnet and Mainnet</h3><p>We did some early scoping this week to work out our process for deploying our contracts to public testnet and mainnet. We’ll start executing on these plans next week.</p><h3>Improving Frontend Deployment Infrastructure</h3><p>As we gear up to launch, we took some time to make our frontend infrastructure more stable. We set up continuous integration, and created an automatic deployment process that will push the latest development version of the frontend up and connect it with our private testnet.</p><p>We’re beginning to realize that having a deployment pipeline, where “dev” is the private testnet, “staging” is the public testnet, and “prod” is the mainnet is a paradigm that might work for our frontend infrastructure. We’re doing more work to understand how to execute on this insight.</p><h3>Next Week’s Priorities</h3><h3>Shipping to Public Testnet</h3><p>A big milestone on the way to mainnet is being able to deploy contracts onto a public testnet. We’re planning on deploying a public testnet version of our release candidate contracts and connecting them to our frontend next week.</p><h3>Testing Deployment to Mainnet</h3><p>As part of our gearing up towards a mainnet deployment, we’re planning on running some simple tests of our mainnet deployment process. We’ll probably ship a simple ERC20 contract onto mainnet to verify that our process for deployment is running smoothly.</p><h3>Completing Delivery through Frontend</h3><p>We’re getting very close to completing deliveries through the frontend. We’re going to push hard on this and get the web UI functional.</p><h3>Getting Involved</h3><p>We’re excited to grow the Computable community. If you’d like to get more involved, follow us on social media or chat with our devs on Discord. Here are some references to help you get started:</p><p><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></p><p><a href="https://github.com/computablelabs">Github</a></p><p><a href="https://www.computable.io/">Website</a></p><p><a href="https://discord.gg/3mFF8zy">Discord</a></p><p><a href="https://twitter.com/computable_io">Twitter</a></p><p><a href="https://t.me/computablelabs">Telegram</a></p><p><em>Originally published at </em><a href="https://forum.computable.io/t/weekly-development-update-november-1st-2019/98"><em>https://forum.computable.io</em></a><em> on November 1, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=569ba1a03d12" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/weekly-development-update-november-1st-2019-569ba1a03d12">Weekly Development Update — November 1st 2019</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Computable Contract Audit]]></title>
            <link>https://medium.com/computable-blog/computable-contract-audit-771e3d39ea7?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/771e3d39ea7</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[audit]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <dc:creator><![CDATA[Bharath Ramsundar]]></dc:creator>
            <pubDate>Tue, 29 Oct 2019 17:02:34 GMT</pubDate>
            <atom:updated>2019-10-29T17:02:34.810Z</atom:updated>
            <content:encoded><![CDATA[<p>At Computable, our mission is to democratize access to data and create a fair economic system around data ownership. A few months ago we released our whitepaper, <a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Fair value and decentralized governance of data </a>which introduced our design of a decentralized protocol for fairly valuing and governing datasets. We also <a href="https://www.computable.io/blog/computable-whitepaper-a-fair-market-protocol-for-data">announced</a> that we’d built a system of smart contracts that implemented our data governance protocol.</p><p>We’ve spent the last few weeks working with the talented team at <a href="https://www.trailofbits.com/">Trail of Bits </a>to audit our smart contract system. Auditing is essential to creating secure software, and this is especially true for smart contract code. Smart contracts tend to implement core infrastructure, control the flow of money, are difficult to upgrade, and have functions that are publicly callable, making them especially attractive to attack.</p><p>Trail of Bits has produced a <a href="https://github.com/computablelabs/computable/blob/master/audit/trail_of_bits_audit_8_30_2019.pdf">detailed report </a>which details the security analysis of our code. Quoting from the executive summary:</p><blockquote>The overall quality of the code base is good. The architecture is suitable and avoids unnecessary complexity. Component interactions are well defined. The functions are small, clean, and easy to understand.</blockquote><p>Our contract system has a number of unique features that made it an interesting project for both of our teams. First, our contracts are written entirely in <a href="https://github.com/ethereum/vyper">Vyper</a>. In fact, we believe our codebase is the largest Vyper smart contract system to undergo an extensive security review thus far. Even though Vyper is a new language, its stronger types and improved numerical support made the switch from Solidity worth it for us. We’ll say more in a future blog post.</p><p>The system consists of <a href="https://github.com/computablelabs/computable">7 interlocking contracts </a>which together define the functionality of a decentralized data market. We worked hard to make our code clean, modular, and easy to understand, making it relatively straightforward to fix issues surfaced by the audit.</p><p>Being meticulous and human-centered about readability also means that everyone on our team and in our community can use the code itself as the source of truth for how the system will behave. In Dapp development, protocol features and their behavior impact a diversity of stakeholders from other smart contract developers to product designers. Docs are important at fleshing out a system’s rationale and behavior, but nothing substitutes for readable code, and smart contracts are no exception.</p><p>Our design faced some unique challenges. First, managing off-chain state through on-chain code is difficult since a smart contract can’t easily enforce that some action has been taken in the physical world. We’ve made a set of structured trade-offs (such as having the datatrust operator be a semi-trusted participant) that permitted us to build a pragmatic, functional system, but we’re upfront that we haven’t entirely cracked the problem of off-chain data governance yet. We’re working hard towards this and have some research <a href="https://forum.computable.io/t/towards-atomic-delivery-of-data/62">ideas</a> we’re pursuing. Second, on-chain governance systems are vulnerable to attacks from “whales,” rich individuals who can use their buying power to take control of a market. Based on feedback from the audit, we’ve made changes to the smart contract system which make it harder (but not impossible) for malicious entities to destructively alter market parameters.</p><p>Another known issue is the possibility for front-running and denial-of-service attacks. A sophisticated adversary could with sufficient effort, slow down transactions in a market and make participating in a market difficult for stakeholders. However, sustaining such an attack would be challenging, and there are steps participants can take to make such attacks harder (such as raising their gas fees). We’re planning to actively watch for such sophisticated attacks in practice and will take steps to educate and mitigate risk to our users. If you’d like to read about the trade-offs we made, we encourage you to read our <a href="https://forum.computable.io/t/detailed-trail-of-bits-audit-response-october-28th-2019/97">detailed response </a>to the audit report.</p><p>All in all, we’re confident that we’ve found a unique and interesting point in the smart contract design space that will serve as a foundation for future work on the fair governance of data. We’d love to hear from you on Twitter <a href="https://twitter.com/computable_io">@computable_io </a>, on our <a href="https://discord.gg/3mFF8zy">discord</a>, or in our <a href="https://forum.computable.io/">forums</a></p><p><strong>Resources</strong></p><p><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></p><p><a href="https://github.com/computablelabs">GitHub</a></p><p><a href="https://www.computable.io/">Website</a></p><p><a href="https://twitter.com/computable_io">Twitter</a></p><p><a href="https://forum.computable.io/">Forum</a></p><p><a href="https://discord.gg/3mFF8zy">Discord</a></p><p><em>Originally published at </em><a href="https://www.computable.io/blog/2019/10/28/computable-contract-audit"><em>https://www.computable.io</em></a><em> on October 28, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=771e3d39ea7" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/computable-contract-audit-771e3d39ea7">Computable Contract Audit</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Computable Whitepaper: A fair market protocol for data]]></title>
            <link>https://medium.com/computable-blog/computable-whitepaper-a-fair-market-protocol-for-data-f56f0b4143f8?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/f56f0b4143f8</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[cryptocurrency]]></category>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[decentralization]]></category>
            <category><![CDATA[privacy]]></category>
            <dc:creator><![CDATA[Roger Chen]]></dc:creator>
            <pubDate>Mon, 10 Jun 2019 20:52:21 GMT</pubDate>
            <atom:updated>2019-06-10T20:52:21.017Z</atom:updated>
            <content:encoded><![CDATA[<h4>Announcing our new paper on <a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Fair value and decentralized governance of data</a></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E8yjeNE4JhH6q79t2F3hvQ.png" /></figure><p>There has been a lot of backlash towards large technology firms recently. Some of it is well deserved, and some of it feels a bit harsh. The truth is that aggregating data is a necessary component for data-driven innovation. For instance, we benefit from countless products, such as web maps and navigation, that large Internet companies have been able to develop because of the wealth of data they possess. However, it shouldn’t <em>just </em>be large tech companies that control the world’s data. The wealth distribution of data is at present dangerously skewed, and we need to establish fairer practices in how our data is controlled and used.</p><p>Given its subjectivity, fairness can of course be hard to define, but we can get a pretty good start by laying down some sensible ground principles. At Computable, here are just some of the things we believe in:</p><ul><li>People should have the right to keep their data private</li><li>If they choose to make it public, they should know how their data is used</li><li>They deserve basic rights even if their data is in the public domain, including the right to compensation and attribution based on their terms of use</li><li>In the open market, more valuable data receives more usage</li><li>Because of replication, data lineage is key to establishing data ownership</li><li>The source of truth for data lineage must be transparent and trustless</li></ul><p>Using these guiding ideas, the Computable team has been hard at work developing a new kind of protocol for fairly sharing data. Today I’m excited to share a new whitepaper that discusses the economic and governance framework behind the protocol. <a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf"><em>Fair value and decentralized governance of data</em></a><em> </em>is meant to be a resource to the community of innovators hoping to create a better way to share and monetize data while respecting personal rights. We hope the ideas presented will spark some of your own. If so, please leave us a message on our <a href="https://forum.computable.io/">new forum</a> or talk to us on <a href="https://discordapp.com/invite/5A9G6hd">Discord</a>. We’d love to find ways to work together!</p><h3>What’s coming next</h3><p>In the coming weeks, we will deploy our smart contract implementation of this protocol on a public Ethereum testnet. Until then, you can follow our <a href="https://github.com/computablelabs">GitHub repo</a> as we continue to develop in the open. Soon developers can build decentralized data market applications with powerful data lineage properties using our protocol’s functions. Application users will be able to leverage the Ethereum blockchain to record ownership of data and control data access. The trustless and transparent transaction records that result offer a universal source of truth for data lineage and authorized access. When combined with regulatory trends like CCPA and GDPR, this powerfully makes data ownership enforceable. When unauthorized use of data is discovered, owners can point to auditable records to prove that they never gave consent to the violating party and pursue appropriate legal recourse. In the future, we will further help owners protect their data by progressively incorporating powerful privacy technologies.</p><p>Another key feature that the protocol will provide is the fair distribution of value for crowdsourced datasets. People and businesses alike will be able to pool their data together (to compound its collective value) and sell access through decentralized data markets powered by the protocol. As that data resource gets consumed, the protocol tracks how often each contributor’s data is used. It then proportionally rebalances market ownership and distributes payments accordingly. As a result, those who have made the most valuable contributions, as measured by real-world usage, will receive the greatest financial reward, rightfully so. This dynamic ownership scheme also forms the basis for a fair and decentralized governance model. Those who continue to add the most value to a network receive the most voting power in making governance decisions. Those who cease to add value meanwhile see their influence erode, defeating rent seeking behavior and helping to sustain healthy and active governance participation.</p><p>Give the <a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">whitepaper</a> a read to learn more and talk to us on our <a href="https://forum.computable.io/">forums</a> or through our <a href="https://discordapp.com/invite/5A9G6hd">Discord channel</a>. The paper was written so that interested developers have enough detail about the design to begin spec’ing out their own implementation code if they so choose. (Or you can always refer back to <a href="https://github.com/computablelabs">ours</a>!) This paper primarily focuses on the economic and governance design behind the protocol. We tried to simplify where we could, but truthfully the protocol’s mechanics and cryptoeconomics involve a great deal of nuance and complexity. Over the coming months, we will release a series of posts that will unpack and explain these nuances as educational content for the community. We will also create documentation materials for those who are interested in working in, on, and around our implementation code. Finally, we will share more research resources including simulation results and various tools to help developers get to building.</p><p>We’re on a path to create universal access to data while greatly respecting those who create that resource. It is a commitment that requires solving numerous challenges in the years ahead, but we think this protocol offers a significant step forward. We offer it as an open resource, and we can’t wait to see what people will do with it!</p><h3>Resources</h3><ul><li><a href="https://github.com/computablelabs/computable/blob/master/whitepaper/computable_whitepaper.pdf">Whitepaper</a></li><li><a href="https://github.com/computablelabs">GitHub</a></li><li><a href="https://www.computable.io/">Website</a></li><li><a href="https://forum.computable.io/">Forum</a></li><li><a href="https://discordapp.com/invite/5A9G6hd">Discord</a></li></ul><p><em>Originally published at </em><a href="https://www.computable.io/blog/computable-whitepaper-a-fair-market-protocol-for-data"><em>https://www.computable.io</em></a><em> on June 10, 2019.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f56f0b4143f8" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/computable-whitepaper-a-fair-market-protocol-for-data-f56f0b4143f8">Computable Whitepaper: A fair market protocol for data</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Economic obstacles to a decentralized oracle]]></title>
            <link>https://medium.com/computable-blog/economic-obstacles-to-a-decentralized-oracle-c27d74467df8?source=rss----638ab2c4f2bb---4</link>
            <guid isPermaLink="false">https://medium.com/p/c27d74467df8</guid>
            <category><![CDATA[blockchain]]></category>
            <category><![CDATA[dapps]]></category>
            <category><![CDATA[decentralized-oracles]]></category>
            <category><![CDATA[prediction-markets]]></category>
            <dc:creator><![CDATA[Artur Gorokh]]></dc:creator>
            <pubDate>Wed, 31 Oct 2018 17:35:22 GMT</pubDate>
            <atom:updated>2018-10-31T17:35:22.337Z</atom:updated>
            <content:encoded><![CDATA[<p>An oracle is a mechanism that summons real-world data to the blockchain, for example, the usd price of some asset, an outcome of an election, or an output of some costly computation. It’s difficult to imagine blockchain applications breaking out of its speculative nature and doing something truly useful without making use of oracles.</p><p>The existing projects sometimes employ centralized oracles, i.e. a trusted source that commits to sending the correct information to the blockchain. Even the projects that are in part oracles themselves, TrueBit¹ and Augur, still employ a centralized oracle within their design. The reason we’re not seeing applications adopting a fully decentral approach is that the problem of creating a functional decentralized oracle is genuinely tough.</p><p>In this post I look at the subject of oracles from the perspective of mechanism design and describe a couple of problems of economic and game-theoretic nature that I believe to be the primary obstacles to a functional decentralized oracle. The post is intended in part as a motivation to <a href="https://medium.com/@arturgorokh/labor-a-proposal-for-a-decentralized-oracle-5b59ea5e623b">my own proposal</a>.</p><p>I thank <a href="https://www.computable.io/">Computable Labs</a> for giving me an opportunity to work on blockchain research this summer, and for allowing me the intellectual freedom to pursue my interests. (<a href="https://angel.co/computable-labs/jobs">They’re hiring!</a>)</p><p>Let’s start with stating some assumptions that span various possible applications. I’ll use the term <strong>task</strong> for a question that task-creator wants to answer using an oracle (Who won the election? What is the market price of REP?). By <strong>workers</strong> I would mean people or machines that are performing the work of fetching the information to the blockchain.</p><h3>Assumptions</h3><ul><li><strong>Truth-value of tasks is universally accessible </strong>— given the willingness to expend time/resource, all participants will come to the same conclusion about task’s true value. <br>I’ll also assume that no answer to the task is overwhelmingly likely to be correct, to avoid <a href="https://eprint.iacr.org/2015/702.pdf">Verifier’s Dilemma</a>( I suspect that almost every question that doesn’t satisfy this condition has a formulation that does, e.g. instead of asking who won the election you can ask how many states have the candidates won).</li></ul><p>- <strong>Fetching information from the real world is labor</strong> — it entails costs and will be performed only if those costs are outweighed by the benefits. Depending on the application, the cost can be measured in the cpu cycles, time spent looking something up, the capital cost to build/update/run software that interacts with API of an online service, or the opportunity cost of not doing something more profitable.</p><p>- <strong>Workers and task-givers are rational.</strong> Given the mechanism, the costs and the state of the market, participants will do the thing that makes them most money — e.g. splitting their accounts into multiple ones, pulling together with other participants, misreporting or just not participating at all.</p><p>There are also a couple of properties tasks commonly have that I think are of great importance.</p><p>- <strong>Tasks may have large underlying stakes and adversaries.</strong> The outcome of a task may have a financial value for some agents that is significantly larger than the cost of fetching the data. For example an outcome of a prediction market can decide a fate of 50000$ in pot money, but fetching the outcome is a job that people would be willing to perform for about 2$.</p><p>- <strong>Task hardness is variable and invisible.</strong> It takes much more effort to report how many House representatives won their districts by more than 10% margin than it does to report who won the US presidency. But the information about hardness of a given task is itself part of the real world thus inaccessible from the blockchain.</p><h3>Low fees vs Security</h3><p>In terms of core dynamics underlying a decentralized oracle, Schelling point game seems to be the only way to go. If I ask many participants to report some real-world outcome, and reward the ones who voted for the most popular option, the true outcome is a point of salience that is easy to coordinate around. Vitalik describes this idea in detail <a href="https://blog.ethereum.org/2014/03/28/schellingcoin-a-minimal-trust-universal-data-feed/">here</a>.</p><p>The first proposal for a decentralized oracle, SchellingCoin, which is essentially just the simple Schelling point game I described above, lets me illustrate an important issue. Imagine the cost of answering a task is small (for example, 1$ to name the current US president), but the underlying stakes are high (100000$ of stakes on a prediction market). An ideal outcome here is prediction market fetching the truth-value while paying only a few bucks for it, but it doesn’t work out this way in case of SchellingCoin.</p><p>If the task giver sets the reward at 5$, then only a few outsiders are going to participate, as after that the marginal utility of joining is going to be too low to attract more voters. The total amount staked is thus going to be low (say 100$), and so is the amount an adversary needs to shift the outcome of the oracle to the desired option. Alternatively, one can set the reward to 20000$, thus getting many outsiders to participate in the oracle and avoiding subversion by an adversary with limited funds. But this security comes at a hefty price — a large portion of the winnings would go towards oracle fees rather than to market’s winners.</p><p>Large fees are bad in themselves, but they also create an additional hazard of parasitic markets that arises due to blockchain’s transparency: anyone can set up an application (e.g. a prediction market) that uses the oracle’s output without paying the fees. As participants switch to a less costly parasitic application, the imbalance between oracle’s fees and underlying stakes grows larger, thus making the system even less secure.</p><p>We want an oracle that makes tasks <strong>cheap to execute and expensive to subvert.</strong> The only way I see of achieving this property is entangling many tasks together, so that subverting a given task costs as much as the total reward pool of all tasks. This can be done by assigning workers to tasks randomly.</p><h3>Free market vs Security</h3><p>Random assignment of workers to tasks is in the nutshell the <a href="https://github.com/JuliaDKoch/papers/raw/master/truebit_predictable.pdf">recent proposal</a> for a modification of TrueBit. When a new task comes along, the mechanism randomly elects a few workers to submit their answers, with matching probability proportional to worker’s stake. To overwhelm a given task, an adversary needs to win several voting spots in a particular task, and doing so with a non-negligible probability requires stake that is comparable with the total stake in the system, which itself is proportional to the total reward pool of the tasks.</p><p>Random matching, at least when employed at scale, seems to solve the problem of cheap and secure tasks, but another issue now pops up. What do we do with the workers who were assigned to a task but didn’t submit an answer?</p><p>If we punish the slackers by burning their stake (as the paper suggests), then a worker is essentially forced to process all assigned tasks, even if they come with unfair rewards (TrueBit avoids this issue by having a centralized oracle that prescribes a reasonable computation price. )</p><p>On the other hand, if we don’t punish the slackers, task-givers have an incentive to price their tasks fairly — otherwise it wouldn’t get processed. But workers are now encourage to stake as much as they can afford, and only process the tasks that are most profitable for them. This can result in tasks taking forever to resolve (as we need to pick new workers after the current ones have timed out) as well as decreased security.</p><p>This is a trade-off between free-market economy and security: random assignment decreases the threat of adversaries, but strips workers of their freedom to pick their workload, which in turn is crucial for incentivizing fair reward-setting.</p><h3>Social layer dependence</h3><p>I also wanted to discuss how Augur’s oracle works. Although the team initially seemed to consider the random worker-task assignment scheme, they eventually landed onto something different. Every prediction market has a dedicated reporter who is to determine the outcome. Any owner of REP can then challenge the outcome reported, effectively starting a stake-weighted vote open to all REP holders.</p><p>Augur’s mechanism suffers from some of the problems I described above — for example, the appropriate rewards for tasks are hardcoded rather than discovered via a market mechanism.</p><p>But the issue that is unique to Augur is its implicit dependence on social layer. Imagine the platform has grown to having thousands of ongoing markets, and my friend and I create a market on some obscure issue(say, predicting how many restaurants will open in Toronto in the first week of October 2018), assigning my friend as a dedicated reporter.</p><p>A sore loser, he reports the wrong outcome to grab the pot money. If the platform is functioning well I shouldn’t expect some passerby REP holder to check if the reported outcome is correct — doing such spot-checks is not going to be profitable for REP holders. Instead I should somehow find a sufficiently wealthy REP-holder and get them to challenge my friend, and for this I’d have to resort to off-chain communication.</p><p>Such dependence on social layer is not a critical flaw, but it makes the oracle considerably less predictable, especially as the market scales up.</p><h3>Taking the right trade-offs</h3><p>To summarize, the only known way of trustlessly getting stuff onto blockchain is playing schelling point games. There are at least two major economic problems left to resolve then.</p><p>First one is making cost of subverting the task much higher than than an appropriate reward for performing the task. A promising path in achieving this seems to be assigning workers to tasks at random, effectively pulling tasks’ security together. Second challenge then is to simultaneously with this grant workers freedom in picking which tasks to process, as this is essential for incentivizing task-givers to set competitive rewards.</p><p><a href="https://medium.com/@arturgorokh/labor-a-proposal-for-a-decentralized-oracle-5b59ea5e623b">In the next post</a> I describe my own proposal for a decentralized oracle that I believe do just that.</p><p><em>¹ Although TrueBit is not an oracle per se due to its use of verification game, it is useful to think of it as one. More precisely, it can be viewed as an oracle with an additional property that the correct value is established whenever at least one worker reports it (whereas in normal oracle we would need at least the correct outcome to be reported by at least the majority of voters)</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c27d74467df8" width="1" height="1"><hr><p><a href="https://medium.com/computable-blog/economic-obstacles-to-a-decentralized-oracle-c27d74467df8">Economic obstacles to a decentralized oracle</a> was originally published in <a href="https://medium.com/computable-blog">Computable Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>