<?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 Luke Feeney on Medium]]></title>
        <description><![CDATA[Stories by Luke Feeney on Medium]]></description>
        <link>https://medium.com/@luke_feeney?source=rss-5880487fd698------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*Dp3-6f3Ms1Ylt1PH-kZGgg.jpeg</url>
            <title>Stories by Luke Feeney on Medium</title>
            <link>https://medium.com/@luke_feeney?source=rss-5880487fd698------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 06:17:55 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@luke_feeney/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[We Love GPLv3, but are Switching License to Apache 2.0: TerminusDB]]></title>
            <link>https://medium.com/terminusdb/we-love-gplv3-but-are-switching-license-to-apache-2-0-terminusdb-355f5f54ce99?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/355f5f54ce99</guid>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[terminusdb]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[cow-duck]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Tue, 08 Dec 2020 16:37:02 GMT</pubDate>
            <atom:updated>2020-12-08T16:37:02.284Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*D-jMNDodUM1OhQThx4QRJA.jpeg" /></figure><h3>Changing License from GPLv3 to Apache 2.0: TerminusDB</h3><p>We have decided to re-license TerminusDB from GPLv3 to Apache 2.0. We want independent software developers (ISVs) to embed TerminusDB in their applications and those developers feel that Apache is a lower risk option. The substantive points of practical difference are far less important — sufficient people believe it to be true and sufficient lawyers have advised teams to be wary of GPL.</p><p>In our experience, ISVs and devs in large companies/institutions size up their options at project conception and there remains a niggling doubt that ‘GPL might limit commercial prospects and cause me headaches’. The world has changed — and code freedom is being overtaken by developer freedom.</p><p>Open-source software is everywhere. It’s eating the world. In the top 10 databases on <a href="https://db-engines.com/en/ranking">DB-Engines</a>, the remaining proprietary databases were released in 1980 (Oracle), 1983 (IBM Db2) and 1989 (Microsoft SQL Server). It is hard to imagine another non-OSS database ever entering the top 10.</p><p>We had hoped that our association with the principals of the free software movement would result in community adoption and contribution, but that hasn’t really been the case. We see limited community input that relates to our choice of GPL. That might not be too surprising as when you investigate which license to choose on Stackoverflow, you get popular but wrong-headed comments like:</p><p>the GNU/GPL bunch are generally extremists when you encounter them in the wild.</p><p>don’t use GPL if you want your project to be commercial</p><p>With the shift to Apache, TerminusDB is, in a sense, becoming more open source as we are removing restrictions on how you can use the software.</p><h3>Debate</h3><p>The core TerminusDB team had a long debate about licenses before the release of 1.0 last year. The main topics of discussion were:</p><ol><li>The risks of a cloud provider forking the code then hosting the database</li><li>What open source means to TerminusDB as a group</li><li>What we, as a community of devs and users, are most comfortable with</li></ol><h4>1. The big bad cloud providers</h4><p>In the past there was an unwritten rule, that big platforms wouldn’t come along and fork open source code and deliver the same product as a service. Unfortunately, those days are gone. AWS in particular has actively sought to offer very similar services to open source products. This led to <a href="https://www.mongodb.com/licensing/server-side-public-license/faq">MongoDB</a>, <a href="https://www.cockroachlabs.com/blog/oss-relicensing-cockroachdb/">CockroachDB</a> and <a href="https://www.confluent.io/confluent-community-license-faq/">Confluent</a> (among others) changing their licences to variations of ‘server side’ and moving away from the open-source tradition. They try to say ‘we are still open source, we are just forbidding a specific type of action’, but it can feel like window dressing for ‘we want to sweat our assets’.</p><p>Mongo, for example, is hardly suffering — it’s valued at over $15 billion. With such vast resources, they should be (and are) able to compete in the provision of their own database. MongoDB’s technology is more than competitive with AWS’ DocumentDB, and Mongo’s Atlas DBaaS — which runs on AWS infra — has been a huge success.</p><p>I’m sure our perspective will shift over time, but from where we’re standing, having a cloud provider launch a competing service would be a sign of enormous success. (And this is not to say that the cloud providers’ parasitic approach to OSS projects is not a genuine problem, it simply acknowledges that you have to be a widely used OSS project before it *becomes* a problem).</p><h3>2. Free software</h3><p>We don’t think it should be our job to provide corporations with free labor.</p><p>We do think that the software community should be able to access and use TerminusDB.</p><p>In 1974 software became copyrightable in the USA. It subsequently became obvious that researchers were giving out software for free, but businesses were not giving back. GNU/GPL came along to provide a new framework for that interface — the software would be free as in freedom (libre). Everybody would be free to modify and distribute, but proprietary additions would not be allowed.</p><p>The problem of businesses not giving back remains today.</p><p>GPL and copyleft provisions work well when they dominate open-source, but their waning popularity increases the ability for ISVs and corporates to go for more open licensing and for legal teams to write anti-GPL provisions into internal rules. Only significant developer push back can change that reality (and why push back when there are so many OSS options with permissive licenses).</p><p>Maybe if <a href="https://db-engines.com/en/system/MySQL">MySQL</a>, its offshoot <a href="https://db-engines.com/en/system/MariaDB">MariaDB</a>, and our graph brothers <a href="https://db-engines.com/en/system/Neo4j">Neo4j</a> weren’t the only GPL flag flyers in the top 20 databases, it might be easier to gain adoption with GPL; however, the other big OSS players: Postgres, Cassandra, Elastic &amp; Redis all go for less restrictive licenses.</p><p>The Affero GPL (AGPL) is treated as an even greater pariah than the GPLv3. The Google internal policy] bans all use of AGPL:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/636/0*RdFtdyyZuTIAESiz.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/620/0*Uld9qhiDUwBGIU-X.jpg" /></figure><p>It reminds of <a href="https://news.ycombinator.com/item?id=13421608">this HN comment</a> on the <a href="http://www.defmacro.org/2017/01/18/why-rethinkdb-failed.html">excellent postmortem</a> of the demise of RethinkDB (recommended reading for all OSS folk):</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/637/0*jt5KQTDuDQxhwG-O.jpg" /></figure><p>Our graph database brothers and sisters in <a href="https://neo4j.com/open-core-and-neo4j/">Neo4j moved</a> from the AGPL to ‘AGPL with a commons clause’ to GPLv3 to making some of the code proprietary. Getting the right license for an OSS project that also allows for some commercialization is far from straightforward. (As Neo certainly know — check out this <a href="https://public.igovsol.com/neo4j-court-records/graphfoundation/80-main.pdf">current court case about the formerly GPL enterprise code</a>).</p><p>It seems that <a href="https://github.com/graknlabs/grakn">Grakn</a> and <a href="https://github.com/fluree/db">Fluree</a> are seeing some success with the AGPL. We genuinely wish them well as we know it is a hard path to walk.</p><h3>3. Our community</h3><p>We did get some negative feedback about the GPL prior to launch. Some people working in corporates weren’t comfortable and thought that it would prevent them from using Terminus. Our response was — <strong>we didn’t develop for them</strong>. And that remains the case; however, the GPL skeptical environment is pervasive. There is a person who tweets TerminusDB after every release asking when the Apache version will land — I always ask him why he thinks he needs an Apache version and he doesn’t really know… though he has a vague feeling that the application he builds on top will be less valuable.</p><p>We thought we would focus on the cloud offering — <a href="https://terminusdb.com/hub/">TerminusHub</a>, which, as SaaS built on the database, has the benefit of being an in-house ‘product’ that doesn’t worry about the license of the underlying software. However, the community wants to build applications on the software and we want to offer an easy way to build a version control and collaboration layer into ISV applications. We think TerminusDB is the perfect infrastructure to build the next <a href="https://www.notion.so/">Notion</a> or <a href="https://roamresearch.com/">Roam Research</a>. We worry about the syncing, versioning and data collaboration — you worry about your users.</p><h3>Why shift Licenses</h3><p>We do not think that our job is to provide corporations with free labor.</p><p>We still believe that <a href="https://www.gnu.org/gnu/manifesto.en.html">“the fundamental act of friendship among programmers is the sharing of programs.”</a></p><p>We would welcome a GPL fork of the TerminusDB code — we are happy to work with any such project should it emerge.</p><p>But we wish to build a community first and foremost. In order to facilitate that community, we will be moving to Apache 2.0 immediately.</p><p>We are going to continue to focus on creating a great open-source database and allow everybody to use that software in their projects.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=355f5f54ce99" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/we-love-gplv3-but-are-switching-license-to-apache-2-0-terminusdb-355f5f54ce99">We Love GPLv3, but are Switching License to Apache 2.0: TerminusDB</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TerminusDB 4.0 — The Star’s End Release]]></title>
            <link>https://medium.com/terminusdb/terminusdb-4-0-the-stars-end-release-ed9f8f44a2ec?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/ed9f8f44a2ec</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[terminusdb]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[version-control]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Tue, 08 Dec 2020 16:32:10 GMT</pubDate>
            <atom:updated>2020-12-08T16:32:10.117Z</atom:updated>
            <content:encoded><![CDATA[<h3>TerminusDB 4.0 — The Star’s End Release</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vHUxtsAc-Ie2RXa-RVW6uQ.png" /></figure><p><strong>Version 4.0 of TerminusDB — the Star’s End Release</strong> * — is a big step forward in solidifying and extending the functionality of TerminusDB and TerminusHub.</p><p><a href="https://terminusdb.com/">TerminusDB</a> is an <a href="https://github.com/terminusdb/terminusdb">open source</a>, revision control graph database designed for distributed collaboration. TerminusDB allows you to push, pull, time-travel and merge data, much in the way that is possible with code in Git.</p><p>With this release TerminusDB takes a significant step towards our vision of a general-purpose tool for data &amp; content management and collaboration.</p><p>TerminusDB 4.0 extends the revision control features to allow greater visibility and reproducability by allowing you to look inside individual commits to see what changes were made. Provenance, lineage, and compliance are increasingly essential in the management of data and content.</p><p>The <em>major</em> additional features are:</p><h3>Model Building Tool</h3><p>The database ships with an integrated visual schema building tool. It allows users to build complex data models using point and click tooling. This makes building models significantly quicker, more efficient, and inclusive of a broader range of people by lowering the technical bar for the application of business rules to your data.</p><p>It lets you visually design knowledge graphs in the same way they are presented to business users. We have tried to make the tool easy to use, while preserving the ability to model the most complex domains.</p><p>Like everything in TerminusDB, the models are versioned so you can make changes in testing and if it breaks your database, you can easily roll back to an earlier working version. This <strong>collaborative knowledge graph design and implementation functionality</strong> does not exist elsewhere.</p><p>You can currently share your models/schemas through TerminusHub — the next step is to make modelling and collaboration even easier by providing data-libraries on TerminusHub, including schema.org but also basic standard models for common domains such as CRM, accounting, and inventory. Users can then generate and share models with collaborators or the broader public.</p><h3>Document Editor</h3><p>TerminusDB 4.0 has full surfability, clickability and editability of database documents through the console. It is a wiki or catalogue of all your data that you can edit in place. You don’t have to write code or execute a query, just edit the document directly. If you have a collaborative project, this is a useful approach to building and curating data assets.</p><p>The documents link to associated documents delivering a linked data application. You can filter documents, you can find the document you want using search, you can create a new document in the interface — it is a <strong>general-purpose tool for managing your data</strong>.</p><p>This is the beginning of the TerminusDB content management system, which we will be expanding over the coming months — next step is to provide web accessibility to TerminusHub databases and then a publisher API to provide the ability to publish the results of any query as a static content set.</p><h3>CSV Manager</h3><p>With a single click, you can now build a database from a CSV or a group of CSVs. You can easily use TerminusDB and TerminusHub as a place to version and collaborate when working with CSV data. This will be especially useful with fast changing data as frequently used in machine learning and data science projects.</p><p>We have tried to make it as simple as possible for users to interact with the CSV tooling. You can view and edit CSVs directly in the document viewer. You can also add CSVs from this interface. No need to write code.</p><p>The database automatically spots if the CSV is already in the database and will just add the delta. There is an append mode that just adds the data that is not there and an update mode that updates the entire CSV.</p><p><strong>Manage your CSVs and other data via TerminusDB and Hub.</strong></p><h3>Command Line Interface</h3><p>We are excited to launch the TerminusDB CLI. You can connect to TerminusDB with a Git-like CLI to run queries or use the revision control features.</p><p>You can create databases and branches, you can list databases/branches, you can query in the command line and you can load CSVs. You also get full error reporting in the CLI. Use the CLI to import a CSV, commit changes and push the changes to TerminusHub.</p><p>The CLI will continue to develop and add new operations. We will make it easier to execute queries directly from the CLI. We’re also planning to launch CLIs for the JavaScript and Python clients.</p><p>TerminusDB is a big step towards our ambition to be a complete decentralized data and content management system that is accessible to all.</p><p><strong>END</strong></p><ul><li>Star’s End is a reference to the mysterious location of the Second Foundation in the Asimov series of novels. The home of the First Foundation was Terminus.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ikYJCeIkBxw5BFEYQ6GTfQ.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ed9f8f44a2ec" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminusdb-4-0-the-stars-end-release-ed9f8f44a2ec">TerminusDB 4.0 — The Star’s End Release</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My First TerminusDB 3.0 Graph]]></title>
            <link>https://medium.com/@luke_feeney/my-first-terminusdb-3-0-graph-4a0da36bcc9f?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/4a0da36bcc9f</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[frank]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Tue, 01 Sep 2020 15:54:18 GMT</pubDate>
            <atom:updated>2021-10-09T15:46:22.002Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*jI0sTRBxH-9Gtunpwxevmg.jpeg" /></figure><p>THIS IS NOT THE LATEST VERSION OF TERMINUSDB — CHECK OUT OUR BLOG OR DOCS FOR NEWER TUTORIALS: <a href="https://docs.terminusdb.com/v10.0/">https://docs.terminusdb.com/v10.0/</a></p><p>In this tutorial, we will start using TerminusDB 3.0, TerminusDB Console and its query client. This is a step-by-step guide to create your first knowledge graph.</p><p>We’re assuming you already have TerminusDB installed, but if you haven’t go to our <a href="https://terminusdb.com/hub/download">download center</a>. It’s very easy to get started.</p><p>In this tutorial, we are going to import data from CSVs about bike journeys between stations in Washington D.C., USA, build a knowledge graph from the data, analyze and visualize it. The data is taken from Capital Bike Share (available <a href="https://www.capitalbikeshare.com/system-data">here</a> — thanks folks) and the CSV data used this tutorial is available at <a href="https://terminusdb.com/t/data/bike_tutorial.csv">https://terminusdb.com/t/data/bike_tutorial.csv</a></p><h3>Create a Database</h3><p>Open up the TerminusDB console (https://127.0.0.1:6363/ — go to our <a href="https://terminusdb.com/hub/download">download center</a> to get up and running). Click on Create to start.</p><p>Click on Create to start and select &#39;TerminusHub Database’ as we want to share this with the world.</p><p>You can specify an id for the database, to make it memorable, let’s make it 1stdb (note that Terminus IDs are URLs and they cannot have spaces!). As a title, enter the name you want to give your Database, something meaningful like My First Database. Then you can add a short description to your database, like It is my first database using TerminusDB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BPuFS7b6qSs6OoE2U7ms6w.jpeg" /></figure><p>Click the Create On TerminusDB button at the top right, and you’ll automatically go to the main database page. Something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MfsEun-c0QUwABhbnmuyJA.jpeg" /></figure><h3>Let’s Create a Schema</h3><p>The schema allows you to organize data into meaningful objects, and it ensures data integrity — nothing goes into your database that is not in the schema. This is a TerminusDB super power — and ensures you derive long term value from your data.</p><p>The TerminusDB Console provides a schema editor using WOQL.js. Remaining on the query page, copy this WOQL.js query into the text box:</p><pre>WOQL.and(         <br>    WOQL.doctype(&quot;Station&quot;)             <br>        .label(&quot;Bike Station&quot;)             <br>        .description(&quot;A station where bicycles are deposited&quot;),         <br>    WOQL.doctype(&quot;Bicycle&quot;)             <br>        .label(&quot;Bicycle&quot;),         <br>    WOQL.doctype(&quot;Journey&quot;)             <br>        .label(&quot;Journey&quot;)             <br>        .property(&quot;start_station&quot;, &quot;Station&quot;)<br>            .label(&quot;Start Station&quot;)             <br>        .property(&quot;end_station&quot;, &quot;Station&quot;)<br>            .label(&quot;End Station&quot;)                                                <br>        .property(&quot;duration&quot;, &quot;integer&quot;)<br>            .label(&quot;Journey Duration&quot;)         <br>        .property(&quot;start_time&quot;, &quot;dateTime&quot;)<br>            .label(&quot;Time Started&quot;)<br>        .property(&quot;end_time&quot;, &quot;dateTime&quot;)<br>            .label(&quot;Time Ended&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;Bicycle&quot;)<br>            .label(&quot;Bicycle Used&quot;)<br>)</pre><p>Add a commit message in the box above (e.g. “Add Schema”) and click Run Query</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UrntDKOlUGkrzJYzZOCYVA.jpeg" /></figure><p>It worked!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RlxcagQG8S0naUzoPdJURw.jpeg" /></figure><p>Let’s stop to review this schema-building query:</p><p>We perform all operations within the and.</p><p>So here’s the operations we have performed:</p><ol><li>We created three different document types (given by the doctype function): Station , Journey and Bicycle</li><li>We added label (names) or description (short descriptions) to them.</li><li>We created properties for Journey, we do that by using the property function after Journey with the first argument as the name of the property and the second argument as the type (or range) of the property.</li><li>For each property, you have to provide an id and the type of that property in property, as with class you can add a label to it as well.</li></ol><p>We can view the result by clicking the ‘Schema’ tab:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vtK0Va19gcRWFgYeKahTTA.jpeg" /></figure><h3>Let’s Load in the Data</h3><p>Now let’s get to the good part. We’ll now load the data from the CSV. We’re going to progressively extend the query to import the data, cleaning it and matching it as we go. WOQL is a highly composable language, you can combine queries arbitrarily using logical ANDs and ORs.</p><p><strong>Let’s build the next query in steps and only hit </strong>Run Query <strong>at the end of the query (full query is available at the bottom of the section)</strong></p><p>Go back to the Query page, and copy in the following query:</p><pre>//Read data from CSV<br>const csv = WOQL.get(<br>    WOQL.as(&quot;Start station&quot;,&quot;v:Start_Station&quot;)<br>        .as(&quot;End station&quot;, &quot;v:End_Station&quot;)<br>        .as(&quot;Start date&quot;, &quot;v:Start_Time&quot;)<br>        .as(&quot;End date&quot;, &quot;v:End_Time&quot;)<br>        .as(&quot;Duration&quot;, &quot;v:Duration&quot;)<br>        .as(&quot;Start station number&quot;, &quot;v:Start_ID&quot;)<br>        .as(&quot;End station number&quot;, &quot;v:End_ID&quot;)<br>        .as(&quot;Bike number&quot;, &quot;v:Bike&quot;)<br>        .as(&quot;Member type&quot;, &quot;v:Member_Type&quot;)<br>).remote(&quot;https://terminusdb.com/t/data/bike_tutorial.csv&quot;)<br>//Transform data into correct shape for insert<br>const wrangles = [<br>    WOQL.typecast(&quot;v:Duration&quot;, &quot;xsd:integer&quot;, &quot;v:Duration_Cast&quot;),<br>    WOQL.typecast(&quot;v:Bike&quot;, &quot;xsd:string&quot;, &quot;v:Bike_Label&quot;),<br>    WOQL.typecast(&quot;v:Start_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ST_Cast&quot;),<br>    WOQL.typecast(&quot;v:End_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ET_Cast&quot;),<br>    WOQL.typecast(&quot;v:Start_Station&quot;, &quot;xsd:string&quot;, &quot;v:SS_Label&quot;),<br>    WOQL.typecast(&quot;v:End_Station&quot;, &quot;xsd:string&quot;, &quot;v:ES_Label&quot;),<br>    WOQL.idgen(&quot;doc:Journey&quot;,[&quot;v:Start_ID&quot;,&quot;v:Start_Time&quot;,&quot;v:Bike&quot;],&quot;v:Journey_ID&quot;),       <br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:Start_ID&quot;],&quot;v:Start_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:End_ID&quot;],&quot;v:End_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Bicycle&quot;,[&quot;v:Bike_Label&quot;],&quot;v:Bike_URL&quot;),    WOQL.concat(&quot;v:Start_ID - v:End_ID @ v:Start_Time&quot;,&quot;v:J_Label&quot;),<br>    WOQL.concat(&quot;Bike v:Bike from v:Start_Station to v:End_Station at v:Start_Time until v:End_Time&quot;,&quot;v:Journey_Description&quot;)<br>]; <br>//Combine with logical and<br>WOQL.and(csv, ...wrangles)</pre><p>If we hit Run Query for this section and you should see the following in the Results Viewer tab:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UQS7NyzPd5kfEmi_7EczGQ.jpeg" /></figure><p>Click on the table drop down menu in the top right of the results and switch to graph view:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pRSoGmSnilH9-QEfjQLt0w.jpeg" /></figure><p>Now you can see your results in graph form.</p><p>We’ll explain each of the steps in the query in turn. As a first step, we saved the part of the query that imports the data from CSV in a const variable named csv. Then we create a list of WOQL operators and save it in another constvariable called wrangles, we combine the two parts of the query with a WOQL.and operator.</p><p>The wrangles clause uses 3 WOQL functions to transform the data into the correct form for input. In each case, the function creates a new variable as output — the last argument in each case. The idgen function generates IDs for our three document types Journey, Station, and Bicycle. The first argument is the prefix that will be used, the second is a list of variables which combine to give a unique identity for the id. For example, in Journey we use 3 fields in the csv (Start_ID, Start_Time and Bike) to generate a unique id Journey_ID.</p><p>Besides generating IDs, we also create new fields with new data types, for example, we use typecast to cast Duration into integer and store it as Duration_Cast. We can also use concat to contract new text formatted with variables in the fields — for example, to create Journey_Label.</p><p>We’ll insert them in the graph by adding triples. Triples are the atomic data entity in the RDF data model. We add the following to the query:</p><pre>const inputs = WOQL.and(csv, ...wrangles)<br>const inserts = WOQL.and(<br>    WOQL.insert(&quot;v:Journey_ID&quot;, &quot;Journey&quot;)<br>        .label(&quot;v:J_Label&quot;)<br>        .description(&quot;v:Journey_Description&quot;)<br>        .property(&quot;start_time&quot;, &quot;v:ST_Cast&quot;)<br>        .property(&quot;end_time&quot;, &quot;v:ET_Cast&quot;)<br>        .property(&quot;duration&quot;, &quot;v:Duration_Cast&quot;)<br>        .property(&quot;start_station&quot;, &quot;v:Start_Station_URL&quot;)<br>        .property(&quot;end_station&quot;, &quot;v:End_Station_URL&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;v:Bike_URL&quot;),<br>    WOQL.insert(&quot;v:Start_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:SS_Label&quot;),<br>    WOQL.insert(&quot;v:End_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:ES_Label&quot;),<br>    WOQL.insert(&quot;v:Bike_URL&quot;, &quot;Bicycle&quot;)<br>        .label(&quot;v:Bike_Label&quot;)<br>);</pre><p>This is the clause that actually inserts the data into the structure that we defined on our schema.</p><ul><li>The insert function inserts a new node into the database with the id Journey_ID and type Journey we add properties like start_time, end_time, duration, start_station, end_station and label and put the variables produced above in their correct spots.</li><li>For Start_Station_URL, End_Station_URL and Bike_URL, we assign a type and label for each of them.</li></ul><p>Finally, we have to put all of the above together and create the query that reads the data from the csv, do the data wrangling and add them in the graph as triples. We add this to complete the query:</p><pre>WOQL.and(inputs, inserts);</pre><p>write some commit message again (e.g. “inserting data”) and click Run Query. Remember, this is getting the data in to our graph so you won’t have a query output just yet. It should look something like this when you select the ‘Documents’ tab:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kTGgQxOxUdeKkbDkJ_-PaA.jpeg" /></figure><blockquote>The full query in all it’s glory and in easy to copy format:</blockquote><pre>const csv = WOQL.get(<br>    WOQL.as(&quot;Start station&quot;,&quot;v:Start_Station&quot;)<br>        .as(&quot;End station&quot;, &quot;v:End_Station&quot;)<br>        .as(&quot;Start date&quot;, &quot;v:Start_Time&quot;)<br>        .as(&quot;End date&quot;, &quot;v:End_Time&quot;)<br>        .as(&quot;Duration&quot;, &quot;v:Duration&quot;)<br>        .as(&quot;Start station number&quot;, &quot;v:Start_ID&quot;)<br>        .as(&quot;End station number&quot;, &quot;v:End_ID&quot;)<br>        .as(&quot;Bike number&quot;, &quot;v:Bike&quot;)<br>        .as(&quot;Member type&quot;, &quot;v:Member_Type&quot;)<br>).remote(&quot;https://terminusdb.com/t/data/bike_tutorial.csv&quot;)<br>//Clean data for insert<br>const wrangles = [<br>    WOQL.typecast(&quot;v:Duration&quot;, &quot;xsd:integer&quot;, &quot;v:Duration_Cast&quot;),<br>    WOQL.typecast(&quot;v:Bike&quot;, &quot;xsd:string&quot;, &quot;v:Bike_Label&quot;),<br>    WOQL.typecast(&quot;v:Start_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ST_Cast&quot;),<br>    WOQL.typecast(&quot;v:End_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ET_Cast&quot;),<br>    WOQL.typecast(&quot;v:Start_Station&quot;, &quot;xsd:string&quot;, &quot;v:SS_Label&quot;),<br>    WOQL.typecast(&quot;v:End_Station&quot;, &quot;xsd:string&quot;, &quot;v:ES_Label&quot;),<br>    WOQL.idgen(&quot;doc:Journey&quot;,[&quot;v:Start_ID&quot;,&quot;v:Start_Time&quot;,&quot;v:Bike&quot;],&quot;v:Journey_ID&quot;),       <br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:Start_ID&quot;],&quot;v:Start_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:End_ID&quot;],&quot;v:End_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Bicycle&quot;,[&quot;v:Bike_Label&quot;],&quot;v:Bike_URL&quot;),    WOQL.concat(&quot;v:Start_ID - v:End_ID @ v:Start_Time&quot;,&quot;v:J_Label&quot;),<br>    WOQL.concat(&quot;Bike v:Bike from v:Start_Station to v:End_Station at v:Start_Time until v:End_Time&quot;,&quot;v:Journey_Description&quot;)<br>];<br>//combine inputs<br>const inputs = WOQL.and(csv, ...wrangles)<br>//generate data to be inserted<br>const inserts = WOQL.and(<br>    WOQL.insert(&quot;v:Journey_ID&quot;, &quot;Journey&quot;)<br>        .label(&quot;v:J_Label&quot;)<br>        .description(&quot;v:Journey_Description&quot;)<br>        .property(&quot;start_time&quot;, &quot;v:ST_Cast&quot;)<br>        .property(&quot;end_time&quot;, &quot;v:ET_Cast&quot;)<br>        .property(&quot;duration&quot;, &quot;v:Duration_Cast&quot;)<br>        .property(&quot;start_station&quot;, &quot;v:Start_Station_URL&quot;)<br>        .property(&quot;end_station&quot;, &quot;v:End_Station_URL&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;v:Bike_URL&quot;),<br>    WOQL.insert(&quot;v:Start_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:SS_Label&quot;),<br>    WOQL.insert(&quot;v:End_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:ES_Label&quot;),<br>    WOQL.insert(&quot;v:Bike_URL&quot;, &quot;Bicycle&quot;)<br>        .label(&quot;v:Bike_Label&quot;)<br>);<br>//Combine inputs and inserts with when clause<br>WOQL.and(inputs, inserts);</pre><p>don’t forget to put in the commit message, then click Run Query.</p><h3>Query The Data</h3><p>Back to the query page — again input into the WOQL.js query builder:</p><pre>WOQL.select(&quot;v:Start&quot;, &quot;v:Start_Label&quot;, &quot;v:End&quot;, &quot;v:End_Label&quot;).and(<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;type&quot;, &quot;scm:Journey&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;start_station&quot;, &quot;v:Start&quot;),<br>	WOQL.opt().triple(&quot;v:Start&quot;, &quot;label&quot;, &quot;v:Start_Label&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;end_station&quot;, &quot;v:End&quot;),<br>	WOQL.opt().triple(&quot;v:End&quot;, &quot;label&quot;, &quot;v:End_Label&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;journey_bicycle&quot;, &quot;v:Bike&quot;)<br>)</pre><p>Click on Run Query and you should see the query returning a table:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*PuVTuNFXpUlK6DzadJ-Qzw.jpeg" /></figure><p>Here we used select to filter out the variables (those starting with v:) that appear in our output. Then we used and to link all the conditions we want to include, as you can see there are lot’s of triples to be conditioned. The ones with opt() means that they are optional — it will be ignored if that data is missing (instead of returning an error — very handy).</p><p>The query can be translated as below:</p><ol><li>select all the Journeys</li><li>and all the start_stations of all the Journeys (let’s call them Start)</li><li>and, if any, all the labels of the start_stations (let’s call them Start_Label)</li><li>and all the end_stations of all the Journeys (let’s call them End)</li><li>and, if any, all the labels of the end_stations (let’s call them End_Label)</li><li>and all the journey_bicycles of all the Journeys (let’s call them Bike)</li></ol><p>We can click the table drop down menu in the top right and get a graph view of the results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hxHi5YPrpAL3wDblI0-wSw.jpeg" /></figure><p>You can go back to ‘DB Home’ to view all of the information about the database, including all your commit messages:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Pcsw0nBweRIkSmj-rRZ_hg.jpeg" /></figure><p>Of course, with TerminusHub there is a way to short circuit this tutorial. You can go back to the start, click on ‘Clone’ and copy the bike database in full. You can then immediately run your queries. No need for any interim steps. Life is getting easier!</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F6BK5tPHKryc%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D6BK5tPHKryc&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F6BK5tPHKryc%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/fa66d19b94d0df80f6a94ff9c194a826/href">https://medium.com/media/fa66d19b94d0df80f6a94ff9c194a826/href</a></iframe><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4a0da36bcc9f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TerminusDB: Importing, Reordering & Exporting a CSV]]></title>
            <link>https://medium.com/terminusdb/terminusdb-importing-reordering-exporting-a-csv-afe138ea66b7?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/afe138ea66b7</guid>
            <category><![CDATA[terminusdb]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[graph]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[database]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Mon, 13 Jul 2020 15:40:01 GMT</pubDate>
            <atom:updated>2020-07-16T16:38:26.055Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mH689YNxbjtVhFl6qBN5wA.png" /><figcaption>Nathan is the bald guy</figcaption></figure><p>Remember <a href="https://youtu.be/xseqLEMF7YQ">Sarah</a>? She’s a junior developer at Industrial Gizmos and Widgets. Her boss Nathan needs to figure out how to unwarble some widgets. It’s not an easy task and her senior team members don’t seem interested.</p><p>Industrial Gizmos and Widgets have some seriously antiquated IT systems and that makes everything harder.</p><p>The main gizmo and widget ERP is built on a DB2 that was first installed in 1986 when DB2 was still young. The database schema is out of control and the source code for the application has been lost. The only person that knows how to fix the DB2 was poached by IBM and they charge Industrial Gizmos and Widgets $2,500 per hour of her time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/735/0*y0PBrNz9kyjcbz64.png" /><figcaption>Database Schema</figcaption></figure><p>The system is a mess.</p><p>Sarah thinks that what they really need is some sort of abstraction layer to allow them to better understand the important things in the ERP data — leave the DB2 chugging away and extract the key information into a knowledge graph. Much easier to manipulate and query once it’s there.</p><p>If they had such a tool, solving the warbled widgets problem would be relatively easy.</p><p>Sarah saw Cheuk’s <a href="https://www.twitch.tv/videos/578697754">‘World of WOQLcraft’ stream on Twitch</a> and thought she could use some of the methods to solve the warbled widgets problem.</p><p>After watching Cheuk’s videos, Sarah ran <a href="https://github.com/terminusdb/terminusdb-quickstart">TerminusDB Quickstart</a>.</p><p>She opened up a terminal (or Git Bash for a Windows user — Sarah runs Ubuntu) and ran:</p><pre>git clone <a href="https://github.com/terminusdb/terminusdb-quickstart.git">https://github.com/terminusdb/terminusdb-quickstart.git</a><br>cd terminusdb-quickstart<br>mkdir csvs</pre><p>Sarah ran the following afterwards:</p><pre>TERMINUSDB_LOCAL=”$(pwd)/csvs” ./terminusdb-container run</pre><p>Windows users can use this:</p><pre>TERMINUSDB_LOCAL=$(cmd //c cd)\\csvs ./terminusdb-container run</pre><p>The mkdir csvs command made sure that Sarah had a local directory — Sarah will save the output of this tutorial in that folder. That is also the purpose of the TERMINUSDB_LOCAL environment variable, it sets the directory for TerminusDB to mount to the local csvs directory.</p><p>After installing TerminusDB using Quickstart, Sarah opens the Console. The default address is <a href="http://127.0.0.1:6363/console">http://127.0.0.1:6363/console</a>.</p><p>The TerminusDB console appears in Sarah’s browser:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xwODx4Vjqd2RfHjy4S6VYw.png" /><figcaption>Create A Database</figcaption></figure><p>Sarah creates a new database so she can import and manipulate her warbled widgets CSV. She calls it ProjectSpreadsheet .</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uhNuCpgxqU20aa_DBZ_ATw.png" /><figcaption>Sarah Has A Database</figcaption></figure><p>Next Sarah wants to eyeball the data to see what she is dealing with. So she <a href="http://terminusdb.github.io/terminusdb-web-assets/tutorials/episode_1/items.csv">grabs the relevant csv extract</a>. It is a mess:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*afCU7RbZqTI3mF6B_4vBYQ.png" /><figcaption>gizmos and widgets</figcaption></figure><p>Sarah quickly writes a short WOQL query to load the CSV into TerminusDB, filter the data to only contain the warbled widgets, and export a new CSV.</p><p>Sarah knows that the warbled widgets contain the date “1970–01–01T00:00:00” as the DB2 was set up in 1986 and no widgets could possibly have been created at the company at that time!</p><p>Here is the query she uses:</p><pre>WOQL.put(</pre><pre>WOQL.as(&#39;Sku&#39;, &#39;v:Sku&#39;)</pre><pre>.as(&#39;date_added&#39;, &#39;v:Date&#39;),</pre><pre>WOQL.and(</pre><pre>WOQL.get(</pre><pre>WOQL.as(&#39;sku&#39;, &#39;v:Sku&#39;)</pre><pre>.as(&#39;date_added&#39;, &#39;v:Date&#39;)</pre><pre>.as(&#39;category&#39;, &#39;v:Category&#39;)</pre><pre>).remote(&#39;http://terminusdb.github.io/terminusdb-web-assets/tutorials/episode_1/items.csv&#39;),</pre><pre>WOQL.eq(&#39;v:Date&#39;, &#39;1970-01-01T00:00:00&#39;),</pre><pre>WOQL.eq(&#39;v:Category&#39;, &#39;widgets&#39;)</pre><pre>), WOQL.file(&#39;/app/local_files/warbled_widgets.csv&#39;))</pre><p>She puts it into the query pane on her TerminusDB console and uses the commit message Importing, Ordering and Exporting CSV so she knows what happens if she has to come back to check. She then hits run query .</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X78t8J595BJ3vWtEnc5CCg.png" /><figcaption>Query Pane on the Console</figcaption></figure><p>The query returned in just under 2 seconds and Sarah found the exported CSV in the csvs directory in the terminusdb-quickstart folder. All tidied up as well:warbled_widgets.csv .</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_mlemHVUwa4s9zz5sPZqgA.png" /><figcaption>Here for Windows Users</figcaption></figure><p>When she opens the CSV, Sarah found that it had been filtered just the way she wanted it:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/903/1*SIz8rg_72p-Ud0fboiqZMQ.png" /><figcaption>No more gizmos</figcaption></figure><p>Sarah took the csv and emailed it to her boss, Nathan. He rang her the next morning to thank her for her excellent work and, while he still called her Sandra, she was happy to have solved a meaningful problem.</p><p>Sarah thought — if I could do that so easily with TerminusDB, maybe I should take a deeper look.</p><p>See you soon for Episode 2!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=afe138ea66b7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminusdb-importing-reordering-exporting-a-csv-afe138ea66b7">TerminusDB: Importing, Reordering &amp; Exporting a CSV</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Be like the mammals when the asteroid struck]]></title>
            <link>https://medium.com/terminusdb/be-like-the-mammals-when-the-asteroid-struck-f6914165629e?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/f6914165629e</guid>
            <category><![CDATA[pandemic]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Tue, 07 Jul 2020 15:52:49 GMT</pubDate>
            <atom:updated>2020-07-07T18:57:40.947Z</atom:updated>
            <content:encoded><![CDATA[<p>From DataChemist to TerminusDB</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*nDDmFBV6mVPJi1GQW_mpww.jpeg" /></figure><p>I don’t care what anyone says, global pandemics are not a good time to get a database startup off the ground. There’s talk that some of the biggest internet companies started during a recession (thanks MongoDB), but goddammit a pandemic is different. Trade shows are a leveler — they want exciting startups hanging around, so they make pricing attractive. Meet-ups are a leveler. That’s why startups use those channels so effectively. Getting from 0 to 1 means finding people that really believe in your project and that’s hard in the vast expanse of the internet. <br> <br>I take some comfort from the fact that both Google and Salesforce started just before the dot com bust and got on pretty well in the long run — but, as I say, I’m not sure that we’ve seen anything as disruptive as this pandemic/civil strife mega-crisis and it sometimes feels like it’s still getting started.</p><p>This blog is to explain to those that care what happened to DataChemist (a very limited set of people) and why you should be interested inTerminusDB.</p><p>The great thing about working in a small enterprise is agility. We were work-from-home in early March and were able to quickly transition to the new world.</p><p>But let me roll back a little (appropriate for a time-traveling database), we are a spin-out from Trinity College in Dublin. Our origins are in building data management systems for large scale research projects. We built the architecture behind <a href="http://seshatdatabank.info/">Seshat, the Global History Databank</a>. When we came out of the university system, our idea was to build large scale graph systems for the enterprise. We were focused on data quality and query — the ability to ingest complex data and then quickly return network-oriented queries (for example: give me all the entities that were connected to Goldman Sachs on January 14 in this large → 1 billion nodes — commercial intelligence dataset) — and we had some success in signing up enterprise customers. But we were drifting into a consultancy model where our engineers did most of the implementation, which is painful when working with varied complex internal systems. We were much more interested in being a technology partner. In Spring 2019, we decided to open source the database and spent the next few months getting everything ready for release (a nice euphemism for reimplementing all the code and building a custom storage layer).</p><p>We also started to think about a name for the project — very important for software as the name is going to be in all sorts of code repositories etc. — we were convinced of the ‘DB’ suffix early as it does a lot of the product positioning for you (it’s a database). Andy Pavlo of Carnegie Mellon University has <a href="http://www.cs.cmu.edu/~pavlo/blog/2020/03/on-naming-a-database-management-system.html">an interesting guide to database naming</a>; he like ‘Postgres’ and ‘Clickhouse’ and calls his new DB first ‘PoopDish’ and then ‘BusTub’ as he wants memorable two-syllable names made up of one-syllable unrelated words. We didn’t follow that at all. Instead, we picked the name of a planet in the Asimov Foundation world, which is also the name of a Roman God, and went with that. We were nearly forced to change to RegulumDB (law in Latin, but in retrospect sounds like a laxative), but late in the day the French holders of the EU wordmark for Terminus came to the rescue and allowed us to use TerminusDB. RegulumDB is available for purchase to anybody that might want to launch a new database health supplement (clears out the buffers).</p><p>After we open-sourced TerminusDB in October 2019, we pursued a dual open source and enterprise engagement strategy using both the DataChemist and TerminusDB names. This made sense as we were making some money from enterprise engagements and we wanted to grow the number of accounts under management. We also had a middle term bet on finding some SaaS customers via TerminusHub (‘GitHub for data’).</p><p>The dual strategy was difficult to pursue during the pandemic. The enterprise product was reliant on the twin horsemen of enterprise sales and residency delivery. Enterprise sales for a complex product delivered by a startup requires considerable effort and many in-person meetings (to build confidence and form allies in the enterprise). In an environment such as this, the established partner has an enormous advantage. It also feels like digital transformation allies are weakened as enterprises just try to keep their lips above the rising tide. <br> <br>In practice, the dual strategy was a strain on the project. It was difficult to maintain two separate identities with different adoption strategies — one bottom-up and the other mostly top-down. Our potential impact was probably reduced (as we are early stage, it is hard to know exactly as the data is so limited).</p><p>We made the early call to drop DataChemist and the direct to enterprise channel so we could focus all our efforts on open source and TerminusDB. To be honest, as an engineering heavy university spin-out open source is a comfortable world.</p><p>It wasn’t an easy decision. We had spent long hard hours building DataChemist software, building a public profile, building relationships with customers and potential customers. That’s a big pill to swallow — a large investment of time and effort (the two things startups lack most) disappears. In the front of house, where I work, it can feel like all your efforts were for nothing as you abandon the thing that you helped to create and with it whatever little brand equity you generated. The sunk costs problem becomes very real when you think about pivoting away from something that you spent countless hours building. Like a lot of tech startup folk, I’m well-schooled in the jargon of success after an intensive diet of business books and podcasts. ‘Low ego’ decision making seems to be almost universally considered a good. In this case, we tried to remove emotion by writing a memo that outlined the new strategy. Once it hit paper, the path became much clearer.</p><p>When we were aligned, we quickly put the DataChemist brand on ice, redirected the website, went remote-first (Dmytri <a href="https://terminusdb.com/blog/2020/03/24/reluctantly-remote/">wrote about that journey</a>) and haven’t looked back. The amazing team at TerminusDB (I know everybody says that, but you folk just don’t understand) see that the opportunity of distributed data collaboration is huge and that quickly responding to changing circumstances is essential if we’re going to succeed.</p><p>We’ll be writing much more about this over the coming weeks, but revision control is the enabler of collaboration. With revision control, you can work on different versions of the same asset at the same time in a controlled way without risk of data loss. In Software Engineering, the Git revision control system has powered a revolution. For data people there is no equivalent — DevOps, DBAs, Data Scientists and others — everything is still manual migration scripts… like software in 2005. That’s the problem we’re solving.</p><p>Since we made put all our eggs in the Open Source, TerminusDB and TerminusHub baskets at the beginning of the pandemic, we have made huge strides forward. But we are still tiny, so hopefully we can be like the mammals when the asteroid struck: small enough to not get wiped out and resilient enough to dominate the ecosystem once the dust has settled.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f6914165629e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/be-like-the-mammals-when-the-asteroid-struck-f6914165629e">Be like the mammals when the asteroid struck</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My First TerminusDB 2.0 Graph]]></title>
            <link>https://medium.com/terminusdb/my-first-terminusdb-2-0-graph-ef7f05038910?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/ef7f05038910</guid>
            <category><![CDATA[development]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[graph]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Wed, 03 Jun 2020 16:10:23 GMT</pubDate>
            <atom:updated>2020-09-01T16:01:14.530Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ZBA-5cqJbrRvQsaL.jpeg" /></figure><p><strong>There is a </strong><a href="https://medium.com/terminusdb/my-first-terminusdb-3-0-graph-4a0da36bcc9f"><strong>newer version of this blog</strong></a><strong>.</strong></p><p>In this tutorial, we will start using TerminusDB 2.0, TerminusDB Console and its query client — WOQLjs. This is a step-by-step guide to create your first knowledge graph.</p><p>We’re assuming you already have TerminusDB installed, but if you haven’t go to our <a href="https://github.com/terminusdb/terminus-quickstart">quick start installation guide</a>. It’s very easy to get started.</p><p>In this tutorial, we are going to import data from CSVs about bike journeys between stations in Washington D.C., USA, build a knowledge graph from the data, analyse and visualise it. The data is taken from Capital Bike Share (available <a href="https://www.capitalbikeshare.com/system-data">here</a> — thanks folks) and the CSV data used this tutorial is available at <a href="https://terminusdb.com/t/data/bike_tutorial.csv">https://terminusdb.com/t/data/bike_tutorial.csv</a></p><h3>Create a Database</h3><p>Open up the Terminus DB console ( default: <a href="http://localhost:6363/console">http://localhost:6363/console</a> ). Click on Create New Database to start with.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Npdm3Q71Vt3lupDR9quKsw.png" /></figure><p>Then click Create Local Database</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NCbwmytJKyKSQoIOaxvukQ.png" /></figure><p>You have to specify an id for the database, to make it memorable, let’s make it 1stdb (note that Terminus’ IDs are URLs and they cannot have spaces). As a title, enter the name you want to give your Database, something meaningful like My First Database. Then you can add a short description to your database, like It is my first database using TerminusDB 2.0.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*groaLJ3uIJHJ5nW-0h_X4A.png" /><figcaption>Click ‘Create New Database’</figcaption></figure><p>Click the Create New Database button at the top right, and you’ll automatically go to the main database page. Something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9EPD3ICI4NTfUh1Cs2rv_A.png" /><figcaption>My First Database</figcaption></figure><h3>Let’s Create a Schema</h3><p>The schema allows you to organise data into meaningful objects, and it ensures data integrity — nothing goes into your database that is not in the schema. This is a TerminusDB super power — and ensures you derive long term value from your data.</p><p>The TerminusDB Console provides a schema editor using WOQL.js. Remaining on the query page, copy this WOQL.js query into the text box:</p><pre>WOQL.and(         <br>    WOQL.doctype(&quot;Station&quot;)             <br>        .label(&quot;Bike Station&quot;)             <br>        .description(&quot;A station where bicycles are deposited&quot;),         <br>    WOQL.doctype(&quot;Bicycle&quot;)             <br>        .label(&quot;Bicycle&quot;),         <br>    WOQL.doctype(&quot;Journey&quot;)             <br>        .label(&quot;Journey&quot;)             <br>        .property(&quot;start_station&quot;, &quot;Station&quot;)<br>            .label(&quot;Start Station&quot;)             <br>        .property(&quot;end_station&quot;, &quot;Station&quot;)<br>            .label(&quot;End Station&quot;)                                                <br>        .property(&quot;duration&quot;, &quot;integer&quot;)<br>            .label(&quot;Journey Duration&quot;)         <br>        .property(&quot;start_time&quot;, &quot;dateTime&quot;)<br>            .label(&quot;Time Started&quot;)<br>        .property(&quot;end_time&quot;, &quot;dateTime&quot;)<br>            .label(&quot;Time Ended&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;Bicycle&quot;)<br>            .label(&quot;Bicycle Used&quot;)<br>)</pre><p>add a commit message in the box below (e.g. “creating schema”) and click Run Query</p><p>Let’s stop to review this schema-building WOQL query:</p><p>We perform all operations within the and.</p><p>So here’s the operations we have performed:</p><ol><li>We created three different document types (given by the doctype function): Station , Journey and Bicycle</li><li>We added label (names) or description (short descriptions) to them.</li><li>We created properties for Journey, we do that by using the property function after Journey with the first argument as the name of the property and the second argument as the type (or range) of the property.</li><li>For each property, you have to provide an id and the type of that property in property, as with class you can add a label to it as well.</li></ol><h3>Let’s Load in the Data</h3><p>Now let’s get to the good part. We’ll now load the data from the CSV. We’re going to progressively extend the query to import the data, cleaning it and matching it as we go. WOQL is a highly composable language, you can combine queries arbitrarily using logical ANDs and ORs.</p><p><strong>Let’s build the next query in steps and only hit </strong>Run Query <strong>at the end of the query (full query is available at the bottom of the section)</strong></p><p>Go back to the Query page, and copy in the following query:</p><pre>//Read data from CSV<br>const csv = WOQL.get(<br>    WOQL.as(&quot;Start station&quot;,&quot;v:Start_Station&quot;)<br>        .as(&quot;End station&quot;, &quot;v:End_Station&quot;)<br>        .as(&quot;Start date&quot;, &quot;v:Start_Time&quot;)<br>        .as(&quot;End date&quot;, &quot;v:End_Time&quot;)<br>        .as(&quot;Duration&quot;, &quot;v:Duration&quot;)<br>        .as(&quot;Start station number&quot;, &quot;v:Start_ID&quot;)<br>        .as(&quot;End station number&quot;, &quot;v:End_ID&quot;)<br>        .as(&quot;Bike number&quot;, &quot;v:Bike&quot;)<br>        .as(&quot;Member type&quot;, &quot;v:Member_Type&quot;)<br>).remote(&quot;https://terminusdb.com/t/data/bike_tutorial.csv&quot;)<br>//Transform data into correct shape for insert<br>const wrangles = [<br>    WOQL.typecast(&quot;v:Duration&quot;, &quot;xsd:integer&quot;, &quot;v:Duration_Cast&quot;),<br>    WOQL.typecast(&quot;v:Bike&quot;, &quot;xsd:string&quot;, &quot;v:Bike_Label&quot;),<br>    WOQL.typecast(&quot;v:Start_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ST_Cast&quot;),<br>    WOQL.typecast(&quot;v:End_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ET_Cast&quot;),<br>    WOQL.typecast(&quot;v:Start_Station&quot;, &quot;xsd:string&quot;, &quot;v:SS_Label&quot;),<br>    WOQL.typecast(&quot;v:End_Station&quot;, &quot;xsd:string&quot;, &quot;v:ES_Label&quot;),<br>    WOQL.idgen(&quot;doc:Journey&quot;,[&quot;v:Start_ID&quot;,&quot;v:Start_Time&quot;,&quot;v:Bike&quot;],&quot;v:Journey_ID&quot;),       <br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:Start_ID&quot;],&quot;v:Start_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:End_ID&quot;],&quot;v:End_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Bicycle&quot;,[&quot;v:Bike_Label&quot;],&quot;v:Bike_URL&quot;),    WOQL.concat(&quot;v:Start_ID - v:End_ID @ v:Start_Time&quot;,&quot;v:J_Label&quot;),<br>    WOQL.concat(&quot;Bike v:Bike from v:Start_Station to v:End_Station at v:Start_Time until v:End_Time&quot;,&quot;v:Journey_Description&quot;)<br>]; <br>//Combine with logical and<br>WOQL.and(csv, ...wrangles)</pre><p>If we hit Run Query for this section and you should see the following in the Results Viewer tab:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J90xWiEM0T1SDzPwd2_4DA.png" /></figure><p>Click on the table drop down menu in the top left of the results and switch to graph view:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*INPkkOcVBTUea88xwqbr6w.png" /></figure><p>Now you can see your results in graph form.</p><p>We’ll explain each of the steps in the query in turn. As a first step, we saved the part of the query that imports the data from CSV in a const variable named csv. Then we create a list of WOQL operators and save it in another constvariable called wrangles, we combine the two parts of the query with a WOQL.and operator.</p><p>The wrangles clause uses 3 WOQL functions to transform the data into the correct form for input. In each case, the function creates a new variable as output — the last argument in each case. The idgen function generates IDs for our three document types Journey, Station, and Bicycle. The first argument is the prefix that will be used, the second is a list of variables which combine to give a unique identity for the id. For example, in Journey we use 3 fields in the csv (Start_ID, Start_Time and Bike) to generate a unique id Journey_ID.</p><p>Besides generating IDs, we also create new fields with new data types, for example, we use typecast to cast Duration into integer and store it as Duration_Cast. We can also use concat to contract new text formatted with variables in the fields — for example, to create Journey_Label.</p><p>We’ll insert them in the graph by adding triples. Triples are the atomic data entity in the RDF data model. We add the following to the query:</p><pre>const inputs = WOQL.and(csv, ...wrangles)<br>const inserts = WOQL.and(<br>    WOQL.insert(&quot;v:Journey_ID&quot;, &quot;Journey&quot;)<br>        .label(&quot;v:J_Label&quot;)<br>        .description(&quot;v:Journey_Description&quot;)<br>        .property(&quot;start_time&quot;, &quot;v:ST_Cast&quot;)<br>        .property(&quot;end_time&quot;, &quot;v:ET_Cast&quot;)<br>        .property(&quot;duration&quot;, &quot;v:Duration_Cast&quot;)<br>        .property(&quot;start_station&quot;, &quot;v:Start_Station_URL&quot;)<br>        .property(&quot;end_station&quot;, &quot;v:End_Station_URL&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;v:Bike_URL&quot;),<br>    WOQL.insert(&quot;v:Start_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:SS_Label&quot;),<br>    WOQL.insert(&quot;v:End_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:ES_Label&quot;),<br>    WOQL.insert(&quot;v:Bike_URL&quot;, &quot;Bicycle&quot;)<br>        .label(&quot;v:Bike_Label&quot;)<br>);</pre><p>This is the clause that actually inserts the data into the structure that we defined on our schema.</p><ul><li>The insert function inserts a new node into the database with the id Journey_ID and type Journey we add properties like start_time, end_time, duration, start_station, end_station and label and put the variables produced above in their correct spots.</li><li>For Start_Station_URL, End_Station_URL and Bike_URL, we assign a type and label for each of them.</li></ul><p>Finally, we have to put all of the above together and create the query that reads the data from the csv, do the data wrangling and add them in the graph as triples. We add this to complete the query:</p><pre>WOQL.and(inputs, inserts);</pre><p>write some commit message again (e.g. “inserting data”) and click Run Query. Remember, this is getting the data in to our graph so you won’t have a query output just yet. It should look something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yAGqDFbnvi9vp2gNJyMOsw.png" /></figure><p>The full query in all it’s glory and in easy to copy format:</p><pre>const csv = WOQL.get(<br>    WOQL.as(&quot;Start station&quot;,&quot;v:Start_Station&quot;)<br>        .as(&quot;End station&quot;, &quot;v:End_Station&quot;)<br>        .as(&quot;Start date&quot;, &quot;v:Start_Time&quot;)<br>        .as(&quot;End date&quot;, &quot;v:End_Time&quot;)<br>        .as(&quot;Duration&quot;, &quot;v:Duration&quot;)<br>        .as(&quot;Start station number&quot;, &quot;v:Start_ID&quot;)<br>        .as(&quot;End station number&quot;, &quot;v:End_ID&quot;)<br>        .as(&quot;Bike number&quot;, &quot;v:Bike&quot;)<br>        .as(&quot;Member type&quot;, &quot;v:Member_Type&quot;)<br>).remote(&quot;https://terminusdb.com/t/data/bike_tutorial.csv&quot;)<br>//Clean data for insert<br>const wrangles = [<br>    WOQL.typecast(&quot;v:Duration&quot;, &quot;xsd:integer&quot;, &quot;v:Duration_Cast&quot;),<br>    WOQL.typecast(&quot;v:Bike&quot;, &quot;xsd:string&quot;, &quot;v:Bike_Label&quot;),<br>    WOQL.typecast(&quot;v:Start_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ST_Cast&quot;),<br>    WOQL.typecast(&quot;v:End_Time&quot;, &quot;xsd:dateTime&quot;, &quot;v:ET_Cast&quot;),<br>    WOQL.typecast(&quot;v:Start_Station&quot;, &quot;xsd:string&quot;, &quot;v:SS_Label&quot;),<br>    WOQL.typecast(&quot;v:End_Station&quot;, &quot;xsd:string&quot;, &quot;v:ES_Label&quot;),<br>    WOQL.idgen(&quot;doc:Journey&quot;,[&quot;v:Start_ID&quot;,&quot;v:Start_Time&quot;,&quot;v:Bike&quot;],&quot;v:Journey_ID&quot;),       <br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:Start_ID&quot;],&quot;v:Start_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Station&quot;,[&quot;v:End_ID&quot;],&quot;v:End_Station_URL&quot;),<br>    WOQL.idgen(&quot;doc:Bicycle&quot;,[&quot;v:Bike_Label&quot;],&quot;v:Bike_URL&quot;),    WOQL.concat(&quot;v:Start_ID - v:End_ID @ v:Start_Time&quot;,&quot;v:J_Label&quot;),<br>    WOQL.concat(&quot;Bike v:Bike from v:Start_Station to v:End_Station at v:Start_Time until v:End_Time&quot;,&quot;v:Journey_Description&quot;)<br>];<br>//combine inputs<br>const inputs = WOQL.and(csv, ...wrangles)<br>//generate data to be inserted<br>const inserts = WOQL.and(<br>    WOQL.insert(&quot;v:Journey_ID&quot;, &quot;Journey&quot;)<br>        .label(&quot;v:J_Label&quot;)<br>        .description(&quot;v:Journey_Description&quot;)<br>        .property(&quot;start_time&quot;, &quot;v:ST_Cast&quot;)<br>        .property(&quot;end_time&quot;, &quot;v:ET_Cast&quot;)<br>        .property(&quot;duration&quot;, &quot;v:Duration_Cast&quot;)<br>        .property(&quot;start_station&quot;, &quot;v:Start_Station_URL&quot;)<br>        .property(&quot;end_station&quot;, &quot;v:End_Station_URL&quot;)<br>        .property(&quot;journey_bicycle&quot;, &quot;v:Bike_URL&quot;),<br>    WOQL.insert(&quot;v:Start_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:SS_Label&quot;),<br>    WOQL.insert(&quot;v:End_Station_URL&quot;, &quot;Station&quot;)<br>        .label(&quot;v:ES_Label&quot;),<br>    WOQL.insert(&quot;v:Bike_URL&quot;, &quot;Bicycle&quot;)<br>        .label(&quot;v:Bike_Label&quot;)<br>);<br>//Combine inputs and inserts with when clause<br>WOQL.and(inputs, inserts);</pre><p>don’t forget to put in the commit message, then click Run Query.</p><h3>Query The Data</h3><p>Back to the query page — again input into the WOQL.js query builder:</p><pre>WOQL.select(&quot;v:Start&quot;, &quot;v:Start_Label&quot;, &quot;v:End&quot;, &quot;v:End_Label&quot;).and(<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;type&quot;, &quot;scm:Journey&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;start_station&quot;, &quot;v:Start&quot;),<br>	WOQL.opt().triple(&quot;v:Start&quot;, &quot;label&quot;, &quot;v:Start_Label&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;end_station&quot;, &quot;v:End&quot;),<br>	WOQL.opt().triple(&quot;v:End&quot;, &quot;label&quot;, &quot;v:End_Label&quot;),<br>	WOQL.triple(&quot;v:Journey&quot;, &quot;journey_bicycle&quot;, &quot;v:Bike&quot;)<br>)</pre><p>Click on Run Query and you should see the query returning a table:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9UuFcy8EHCJlm4qojSxqAw.png" /></figure><p>Here we used select to filter out the variables (those starting with v:) that appear in our output. Then we used and to link all the conditions we want to include, as you can see there are lot’s of triples to be conditioned. The ones with opt() means that they are optional — it will be ignored if that data is missing (instead of returning an error — very handy).</p><p>The query can be translated as below:</p><ol><li>select all the Journeys</li><li>and all the start_stations of all the Journeys (let’s call them Start)</li><li>and, if any, all the labels of the start_stations (let’s call them Start_Label)</li><li>and all the end_stations of all the Journeys (let’s call them End)</li><li>and, if any, all the labels of the end_stations (let’s call them End_Label)</li><li>and all the journey_bicycles of all the Journeys (let’s call them Bike)</li></ol><p>We can click the table drop down menu in the top right and get a graph view of the results:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XNMYRFv1XkXu7Tpsb1i-2Q.png" /><figcaption>Graph View</figcaption></figure><p>You can experiment with the query to find different things in the data.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ef7f05038910" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/my-first-terminusdb-2-0-graph-ef7f05038910">My First TerminusDB 2.0 Graph</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Terminalia — Our Day]]></title>
            <link>https://medium.com/terminusdb/terminalia-our-day-112f8726c88?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/112f8726c88</guid>
            <category><![CDATA[terminusdb]]></category>
            <category><![CDATA[rome]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[do-it-yourself]]></category>
            <category><![CDATA[party]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Fri, 21 Feb 2020 16:50:52 GMT</pubDate>
            <atom:updated>2020-02-21T16:50:52.019Z</atom:updated>
            <content:encoded><![CDATA[<h3>Terminalia — Our Day</h3><p>The end of the year approaches. This Sunday 23 February we celebrate Terminalia — the festival that marks the end of the Old Roman Year.</p><p>TerminusDB is named after the Roman God of boundaries, Terminus — all databases need good boundaries — and Terminalia is his festival.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/829/0*ArjA9ORibKUVEZFP.jpeg" /><figcaption><strong>We want the finest wines available to humanity</strong>!</figcaption></figure><p>Terminus is an old Roman god — probably introduced by Romulus after the founding of the city. There is a famous story told of the last king of Rome, Tarquin (so disliked &amp; proud that he was overthrown and the Republic declared) — he wanted to reshape Rome as an imperial capital and engaged in a large scale building programme. As part of his works, he set about reconstructing the Temple of Jupiter on the Capitol Hill. Tarquin thought that constructing a massive temple over the remains of a holy site consecrated by Romulus would cement his place as a great King of Rome. To facilitate the reconstruction, the older shrines to various gods had to be removed; however, when they attempted to move the shrine of Terminus, it would not budge. Terminus was the immovable rock of the Capitol. Even Jupiter, the king of the gods, had to kneel to Terminus.</p><p><a href="https://medium.com/terminusdb/terminusdb-whats-in-a-name-27b8be6b0794">More on Terminus and TerminusDB’s name</a>.</p><p>Terminus’ slogan ‘<em>Concedo Nulli</em>’ — concede to nobody — remains at the heart of the TerminusDB project. We continue to walk through walls to get the features out the door. We will not yield.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*QpWPQjM2KyznDjeD.jpg" /></figure><p>Terminalia is the official holiday for TerminusDB. As it’s on a Sunday this year, we celebrate tonight. As Ovid dictates, we will allow the pagan spirits to flow. We will pour lamb’s blood on the statue of Terminus and offer him suckling pigs and wine.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*u1Y5gT-CQ3_Fx6bz.jpg" /></figure><p>“When night has passed, let the god be celebrated With customary honour, who separates the fields with his sign. Terminus, whether a stone or a stump buried in the earth, You have been a god since ancient times. You are crowned from either side by two landowners, Who bring two garlands and two cakes in offering. An altar’s made: here the farmer’s wife herself Brings coals from the warm hearth on a broken pot. The old man cuts wood and piles the logs with skill, And works at setting branches in the solid earth. Then he nurses the first flames with dry bark, While a boy stands by and holds the wide basket. When he’s thrown grain three times into the fire The little daughter offers the sliced honeycombs. Others carry wine: part of each is offered to the flames: The crowd, dressed in white, watch silently. Terminus, at the boundary, is sprinkled with lamb’s blood, And doesn’t grumble when a sucking pig is granted him. Neighbours gather sincerely, and hold a feast, And sing your praises, sacred Terminus: `You set bounds to peoples, cities, great kingdoms: Without you every field would be disputed. You curry no favour: you aren’t bribed with gold, Guarding the land entrusted to you in good faith. If you’d once marked the bounds of Thyrean lands, Three hundred men would not have died, Nor Othryades’ name be seen on the pile of weapons. O how he made his fatherland bleed! What happened when the new Capitol was built? The whole throng of gods yielded to Jupiter and made room: But as the ancients tell, Terminus remained in the shrine Where he was found, and shares the temple with great Jupiter. Even now there’s a small hole in the temple roof, So he can see nothing above him but stars. Since then, Terminus, you’ve not been free to wander: Stay there, in the place where you’ve been put, And yield not an inch to your neighbour’s prayers, Lest you seem to set men above Jupiter: And whether they beat you with rakes, or ploughshares, Call out: “This is your field, and that is his!”’ There’s a track that takes people to the Laurentine fields, The kingdom once sought by Aeneas, the Trojan leader: The sixth milestone from the City, there, bears witness To the sacrifice of a sheep’s entrails to you, Terminus. The lands of other races have fixed boundaries: The extent of the City of Rome and the world is one.” — Ovid, Fasti II</p><p><a href="https://www.irishtimes.com/opinion/the-other-marching-season-terminalia-the-roman-festival-of-boundaries-1.3803276">Terminalia and Terminus also sneak into Irish place names and traditional events</a> (link to a great <em>Irish Times</em> article on the subject)</p><p>Terminators around the world are encouraged to celebrate this momentous day. Wear white, sacrifice a lamb, let the spirit of pre-Republican Rome flow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/0*P0qHv01fUcd8XSMU.jpg" /><figcaption>Get out there, Terminators</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=112f8726c88" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminalia-our-day-112f8726c88">Terminalia — Our Day</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TerminusDB @ PyLadies Dublin]]></title>
            <link>https://medium.com/terminusdb/terminusdb-pyladies-dublin-6fcb855959ac?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/6fcb855959ac</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[politics]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Fri, 21 Feb 2020 15:55:45 GMT</pubDate>
            <atom:updated>2020-02-21T15:55:45.036Z</atom:updated>
            <content:encoded><![CDATA[<p>Hello Terminators!</p><p>A first — for me at least — attending PyLadies, Dublin!</p><p>PyLadies helps more women become active participants and leaders in the Python open-source community. PyLadies Dublin also provides a friendly support network for women and a bridge to the larger Python world. Anyone with an interest in Python is encouraged to participate. You can see more<a href="https://www.meetup.com/PyLadiesDublin/"> here</a> — we really enjoyed, so I’d encourage people to attend.</p><p>TerminusDB Developer Advocate Extraordinaire, Cheuk, spoke at the PyLadies event earlier this week. There were about 40 PyLadies in the audience and a sprinkling of interloping PyMen.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TyLh-oQIByoTjJ37rR3HzQ.jpeg" /></figure><p>The event took place in the Dublin office of Qualtrics, which has a full replica of an Irish pub. Another good reason to attend!</p><p>Cheuk used TerminusDB to ask if we are supporting the correct politician. She used data from the Dublin City Council. It is a fascinating use case, and relevant in Ireland as we have just had an inconclusive election (and might have another in the not too distant future). There is a run through of the demo <a href="https://medium.com/terminusdb/are-you-supporting-the-right-politician-b742debbc8d9?source=collection_home---4------3-----------------------">over here</a> — give it a try.</p><p>The full video of the excellent presentation is here:</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F3XKJ9_bplUY%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D3XKJ9_bplUY&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F3XKJ9_bplUY%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/b59b8d3842d9c00645a12c1d4fb0f873/href">https://medium.com/media/b59b8d3842d9c00645a12c1d4fb0f873/href</a></iframe><p>TerminusDB is open source now and forever so it is perfect for the Python community. We are rolling out ever more features for Pythonistas — check out our <a href="https://medium.com/terminusdb/design-a-query-language-client-for-pythonistas-and-data-scientists-db5e6a76141a?source=collection_home---4------1-----------------------">python query client!</a> And watch this space for exciting news.</p><p>We’ll keep you up to date with reports from TerminusDB talks and events. A veritable Terminator news service.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6fcb855959ac" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminusdb-pyladies-dublin-6fcb855959ac">TerminusDB @ PyLadies Dublin</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TerminusDB Discussions]]></title>
            <link>https://medium.com/terminusdb/terminusdb-discussions-511c6369dcf9?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/511c6369dcf9</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[maoism]]></category>
            <category><![CDATA[dialectics]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[development]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Tue, 28 Jan 2020 16:18:32 GMT</pubDate>
            <atom:updated>2020-01-28T16:18:32.862Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1020/1*MLXi8J5vwHQheBcEo5T-EA.png" /></figure><p>One of the really nice things about TerminusDB is our weekly get together where one team member gives a short presentation on something. It doesn’t have to be work, it doesn’t have to be TerminusDB, it can be anything really.</p><p>We are a moderately distributed team and we have plenty of opportunity for detailed technical discussion (indexing strategy anyone?) so we thought that it’d be a good idea to start a less formal weekly exchange. One member of the team leads the discussion and the rest contribute with questions &amp; comments.</p><p>Over the year and a bit since we started the tradition, we’ve had everything from the innards of JavaScript to Crypto Kitties to Theories of Developer Relations to Cloud Architecture to the Dialectics of Software Development (a personal favorite — I must post the slides).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/801/1*AiOAyqAcjMq5qd8fBUdw2w.png" /><figcaption>The right superstructure is important</figcaption></figure><p>We’ve also had nuclear power, the political economy of open source software, values vs virtues and Terry Davis (TempleOS is something else — I’d encourage all to check it out).</p><p>Really whatever was of special interest to the person assigned that day. I looked back through the chat and realize that I have suggested ‘Practical Maoism for the Workplace’ twice and nobody has taken up my lead.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/654/1*qa5C7ZG-4mvCH5KKJaUiLg.png" /><figcaption>The system&#39;s features, such as its 640x480 resolution, 16-color display and single audio voice, were explicitly instructed to Terry Davis by God</figcaption></figure><p>The tradition has grown to be an important part of every TerminusDB week. If we continue to grow the core group, I’m not sure how we’ll scale — 1 week in 30 might be a little infrequent. We can cross that bridge when we get there.</p><p>So anyway, we really value these discussions and thought the the broader Terminus Community might also enjoy. We’ve recorded the latest discussion, which took place yesterday, and uploaded it to our YouTube channel.</p><p>This week we were led by our BDFL Gavin in a discussion of ‘Regular Expressions - Theory and Practice’. It covers a bunch of interesting aspects including Chomsky’s Hierarchy, Büchi automata, grammars and much more.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FmbtpAyn0n0o%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DmbtpAyn0n0o&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FmbtpAyn0n0o%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/53b0d23b2b942f18cdc61b14a173cf4b/href">https://medium.com/media/53b0d23b2b942f18cdc61b14a173cf4b/href</a></iframe><p>I apologize for the sound quality issues — we’ll work on a setup that captures the various contributors in high fidelity. At least you can hear the main presentation. Only one instance of adult language — less than normal I’d say — so be warned.</p><p>This will be a regular series, so watch this space.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=511c6369dcf9" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminusdb-discussions-511c6369dcf9">TerminusDB Discussions</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[TerminusDB 1.1 — The Big Babushka]]></title>
            <link>https://medium.com/terminusdb/terminusdb-1-1-the-big-babushka-9020aeb712b7?source=rss-5880487fd698------2</link>
            <guid isPermaLink="false">https://medium.com/p/9020aeb712b7</guid>
            <category><![CDATA[rust]]></category>
            <category><![CDATA[database]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[developer-tools]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Luke Feeney]]></dc:creator>
            <pubDate>Thu, 16 Jan 2020 17:01:38 GMT</pubDate>
            <atom:updated>2020-01-28T14:44:28.484Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*y6pmHcjylW-YGQFbtlXicw.png" /></figure><h3>TerminusDB 1.1 — The Big Babushka</h3><p>Another very big day for us — our first iteration on <a href="https://github.com/terminusdb/terminus-server">TerminusDB goes out into the wild</a>. We are calling it the <strong>Big Babushka</strong> release as it has some huge changes to the way we manage data and the operations we support — it could almost be a 2.0, but given that we are a relatively young DB, we want to save those release numbers for future exciting developments (I did see another DB on release 0.50.0 — I won’t name names, but come on folks, give us a 1.0!).</p><h3>Terminus-store</h3><p>The major development is our new storage backend, terminus-store. With the introduction of the new store, TerminusDB takes a radical departure from traditional database architectures. Our approach has three parts:</p><ol><li>We have a graph database with strong schema to retain simplicity and generality</li><li>The graph is implemented using succinct immutable data structures which enable more sparing use of main memory resources.</li><li>We adopt a delta encoding approach to updates (‘like git, but for data’) which provides the whole suite of revision control features: branch, merge, squash, rollback, blame, and time-travel facilitating CI/CD approaches on data.</li></ol><p>This final point is crucial — with TerminusDB we can now use advanced CI/CD workflows in data operations. The large impact of git on the software development world can now be envisioned in the world of data. This is made possible by synergies between an immutable layered approach and the succinct data structure approach that we’ve used for encoding. The DevOps revolution becomes the DataOps revolution.</p><p>Just as with DevOps, DataOps means a focus on collaboration, automation and integration. There is a requirement to bridge what is today a chasm between huge reams of unstructured data on the one hand, and analysts looking for access to simple, clean, structured data on the other. DataOps is that bridge.</p><p>TerminusDB is a practical tool for enabling branch, merge, rollback, and the various automated and manual testing regimes which they facilitate on a transactional database management system while providing sophisticated query support</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/597/1*ADSnYco_5vlMbLT7eiT2Rw.png" /></figure><p>The new TerminusDB store is implemented in Rust. It is low level, memory safe, has no run time and is C compatible. We think Rust will be the language of the 2020s and we are delighted to be Rustaceans!</p><p>There is a <a href="https://github.com/terminusdb/terminus-server/blob/dev/docs/whitepaper/terminusdb.pdf">TerminusDB Technical Paper</a> that goes into this in more detail for those interested.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/839/1*S9ho_dXwojark-UnLrRSeQ.png" /></figure><h3>WOQL</h3><p>The Web Object Query Language — or WOQL — has developed considerably since our first release:</p><ul><li>Extended WOQL to greater functionality for string and arithmetic processing</li><li>Better date handling in WOQL</li><li>Improved type casting in WOQL</li><li>Improvements to query processing speed</li></ul><p>The growing maturity of WOQL provides a strong platform to import, clean and query complex data. WOQL remains the best graph query language available today.</p><p>WOQL itself is specified as a JSON document format. All transmission of data uses the canonical JSON format, but this is generally annoying to write, hence WOQL.js is the preferred means by which human users can write queries manually. In developing WOQL we wanted to make querying graph data easy to learn, understand, and use, but also build on the power and functionality of other standard data access languages. As we can see from the TerminusDB tutorials, WOQL is a highly composable and flexible graph query language. Graphs are very generically applicable to a much wider range of data than table structured data. Ultimately, the flexibility of graphs with well specified schema, will show their value.</p><h3>Terminus — Console</h3><p>The new Console significantly builds on the simple JavaScript client application that released with 1.0.</p><p>In the query interface, you can now switch between WOQL.js and JSON. You can also add new queries to the original query — as inspired by the Jupyter Notebook functionality you know and love.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uScpmhNCu0UwUR30Y_RGAA.png" /></figure><p>Once you have run your query, we have a view interface to easily change the view on the output</p><p>Throughout the console you now have the ability to toggle between table and graph view. On the query page, on the document page and on the schema page, you can see as a graph. This is super helpful when you are trying to see what’s in a complex schema, or trying to derive insight from a query.</p><p>The new Console gives user a quality interface for managing and querying TerminusDB. It ships with TerminusDB.</p><h3>Python Client</h3><p>With the release of the TerminusDB python client, python users can query data and interact with TerminusDB through the API and WOQL.py. We will continue to work to make TerminusDB more accessible for the python community (firstly, by improving our documentation!). TerminusDB is the database for data people, and there are no people that are more data than Pythonistas!</p><h3>What’s Next</h3><p>We are working hard on our collaboration and DataOps features. They should be available by the end of February 2020.</p><p>We’ll be pushing out documentation as fast as we can. There is lots of new stuff available, including a <a href="https://medium.com/terminusdb/my-first-terminusdb-graph-visualisation-bike-share-data-39c59a1ab86a">couple</a> of <a href="https://medium.com/terminusdb/my-first-terminusdb-graph-visualisation-bike-share-data-39c59a1ab86a">tutorials</a>, but it is never enough!</p><p>Remember we are open source (GPLv3) now and forever.</p><p>(And for anybody wondering, that’s CowDuck, our new mascot)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9020aeb712b7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/terminusdb/terminusdb-1-1-the-big-babushka-9020aeb712b7">TerminusDB 1.1 — The Big Babushka</a> was originally published in <a href="https://medium.com/terminusdb">TerminusDB Community</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>