<?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[AsyncAPI - Medium]]></title>
        <description><![CDATA[AsyncAPI blog - Medium]]></description>
        <link>https://medium.com/asyncapi?source=rss----57fbe5a242be---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>AsyncAPI - Medium</title>
            <link>https://medium.com/asyncapi?source=rss----57fbe5a242be---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 21:39:02 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/asyncapi" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[An API Strategist Explores Event-Driven APIs]]></title>
            <link>https://medium.com/asyncapi/an-api-strategist-explores-event-driven-apis-9cc59335cfc0?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/9cc59335cfc0</guid>
            <category><![CDATA[api]]></category>
            <category><![CDATA[event-driven-architecture]]></category>
            <category><![CDATA[open-api]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[event-driven]]></category>
            <dc:creator><![CDATA[Emmelyn Wang]]></dc:creator>
            <pubDate>Fri, 14 Jun 2019 16:15:25 GMT</pubDate>
            <atom:updated>2020-05-27T13:57:12.637Z</atom:updated>
            <content:encoded><![CDATA[<p>If you’re like me and have experience with web services and the paradigm of RESTful, web services, and SOAP APIs, but curious about event-driven, message based, or streaming APIs, this article is for you.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YQtpeQz0Jk7movlxHs549A.jpeg" /><figcaption>Credit: Pixabay PIO4D Skyscraper</figcaption></figure><p>I virtually sat down with these experts:</p><ul><li><a href="https://twitter.com/fmvilas">Fran Mendez</a>, Founder of <a href="https://twitter.com/AsyncAPISpec">AsyncAPI</a></li><li><a href="https://twitter.com/JSchabowsky">Jonathan Schabowsky</a>, Chief Architect, <a href="https://twitter.com/solacedotcom">Solace</a></li><li><a href="https://www.linkedin.com/in/david-mckenna-05726b1">David McKenna</a>, SVP, Research &amp; Development, <a href="https://twitter.com/Axway">Axway</a></li><li><a href="https://twitter.com/EricHoresnyi">Eric Horesnyi</a>, General Manager, APIs &amp; Integration, Axway (CEO of Streamdata.io acquired by Axway)</li></ul><p>Question 1: <strong>What do you think is the most important value proposition of event-driven architecture and event-driven/messaging APIs?</strong></p><p>Question 2: <strong>What is the consumption experience of messaging APIs and how much does it differ from RESTful APIs?</strong></p><p>For example, at the time of writing can’t just mock up the API and run it in a Postman collection and transform it (convert the technical specification to the version I want such as YAML to RAML or JSON, for example) and consume it.</p><p>Try <a href="http://editor.asyncapi.org/">http://editor.asyncapi.org</a> to understand the sister spec to OpenAPI (OAS).</p><p>Question 3: <strong>Today, how can I test drive event-driven APIs out in the wild?</strong></p><p>My frame of reference with REST APIs: I can inspect HTTP calls in a browser and mock up the API. I can run the API calls in API Builder or Stoplight or Postman and get an InVision conceptual app to consume the API, for example. I can use Stoplight to lint or APIMatic to create starter SDKs to test drive.</p><p>Question 4: <strong>How do you measure event-driven APIs?</strong></p><p>I was thinking about how you measure the <em>effects</em> of event-driven APIs. For example, how can various types of calls live within dashboards for the sake of analytics and measurement? Future Trend: Is there a way to get wholistic data from webhooks, RESTful/JSON, SOAP/WSDL/WADL/XML, GraphQL, gRPC, etc.?</p><blockquote>POV: The future of APIs is the unified catalog having a matching data-driven view. <a href="https://twitter.com/lifewingmate">lifewingmate</a></blockquote><h3>For the greater API Community</h3><p>Recommendations are the most important piece of the event-driven architecture value proposition. Business leaders only care about event-driven architecture as much as the end result produces the business-driving experience that specifically this type of architecture can produce.</p><h4>Technology and consulting organizations need to know when and how to recommend event-driven architecture, APIs and microservices, and corresponding methods of implementation.</h4><h4>How they impact my current product engine? What business results justify the spend? — <a href="https://twitter.com/lifewingmate">lifewingmate</a></h4><p>Here we go!</p><h3>Question 1: What do you think is the most important value proposition of event-driven architecture?</h3><p>Fran: I’d say most interesting for me are:</p><ul><li>loose coupling between services (if a downstream service breaks the others just work fine),</li><li>they’re fault-tolerant since events are queued or stored somewhere, and processed once the service is up and running again, and</li><li>they allow you to build real-time products easily.</li></ul><p>Jonathan: 1) <strong>Loose Coupling </strong>— As Fran states, apps do not have to know how many different apps they are consuming data from. Rather they just care about what data they are consuming. This [scenario] is extremely powerful.</p><p>Also, data today is RARELY consumed only once, rather, it goes into big data and analytics platforms. The loose coupling described allows you to have this for free.</p><p>2) <strong>Reactive to Change</strong> — Instead of polling, consumers can register their their interest (subscribe) and react to changes in real-time. The producer just sends events, and does not care if anyone is interested… fire and forget… Meanwhile clients that are subscribed get the data in realtime and can react. This is useful where you want to service customers in real-time.</p><p>3) <strong>Scale </strong>— Want to be able to consume events, in order, across millions of devices? How about do 500k messages per second guaranteed of capital market trades?</p><blockquote>Asynchronous interactions enable enterprises to do things which are nearly impossible to do synchronously.</blockquote><blockquote>— <a href="https://twitter.com/JSchabowsky">JSchabowsky</a></blockquote><p>David, Eric: Event-driven architecture brings comfort to end users and efficiency to the web at individual component level and at system level:</p><ul><li>UX in Frontends</li><li>Network</li><li>APIs</li><li>Backends</li><li>Behind the API: Application languages, data plane, and DevOps</li><li>Entire System</li></ul><p><strong>UX in Frontends</strong> The <a href="https://hackr.io/blog/10-best-javascript-frameworks-2019">most popular</a> reactive frameworks (AngularJS, ReactJS, VueJS) react to streams of events presented to their <a href="https://apifriends.com/api-streaming/realtime-event-driven-app-with-riot-xignite-and-streamdata-io/">observables</a>, a data plane presenting events to the UI.</p><p>Why is that? Because, we -mobile and desktop human users- want to be presented data in real-time:</p><ul><li>Where is my cab? <br>What is the price of my favorite share, now?</li><li>At what time will my train arrive at destination ?</li></ul><p>Yes, our life is real-time: we do not want to waste any time and we do not want to miss anything that happens in the world (<a href="https://en.wikipedia.org/wiki/Fear_of_missing_out">FOMO</a> — Fear Of Missing Out).</p><p>Good news: Events improve latency of the UI by factor x20 according to our <a href="https://apifriends.com/api-streaming/ui-latency-json-patch/">benchmark</a>, and avoids for users to have to refresh their UI to get the latest from their newsfeed, stocks or favorite transportation, as our <a href="https://apifriends.com/api-streaming/animated-data/">reptilian brain</a> is naturally used to [understanding].</p><p><strong>Network</strong> Since 2011, with the advent of social networks and to support reactive UI, the web has RFCed protocols for low-volume bidirectional/peer-to-peer traffic (<strong>Websockets</strong>) and server to client push over HTTP (<a href="https://apifriends.com/api-streaming/push-sse-vs-websockets/"><strong>SSE</strong></a>). For lower volume of events, typically for alerts, <strong>webhooks</strong> have become popular.</p><p><strong>APIs</strong> Most API calls are useless : up to 98.5% according to <a href="https://news.ycombinator.com/item?id=6360474">Zapier</a> who created the concept of Polling Madness. That is because the API client does not know when data will change, so would keep polling an API to make sure it does not miss any update. Event-Driven Architecture in APIs reverses the paradigm: why don’t we have the component who know when data change to take charge, instead of answering to useless calls?</p><p>This [situation] leads to the API server pushing streams of events to the API client who have subscribed to a topic. Traffic hence goes north-south only when necessary, rather than south-north-south to hit updates randomly. Augmenting a REST API with a streaming API typically brings <a href="https://apifriends.com/api-streaming/the-web-environmental-footprint-in-short/">90% efficiency</a> in CPU and network use for most demanding traffic.</p><p><strong>Backends</strong> In some industries, backends have been using evented architecture for a while.</p><p><strong>Behind the API: Application Languages, data plane, and DevOps</strong> <br>Backend architectures have relied on events for a while, before REST was created. In my Flashboy days in <a href="https://www.infoq.com/podcasts/eric-horesnyi-ai-hft">High Frequency Trading</a>, we were pushing events from exchanges to hedge funds to allow them to stay in front of the market (making big money at low risk, I’m repentant), using IP-based tweaks (IP multicast and their famous storms), and proprietary middleware.</p><p>Since then, message buses have evolved to open standards adopted outside finance (thanks Linkedin for <strong>Kafka</strong>), and -as already noted- the web has made it possible to transport events over HTTP. Additionally, and almost at the same time, people -like <a href="https://apifriends.com/api-streaming/the-web-environmental-footprint-in-short/">myself-</a> having advocated microservices for years without seeing convergence of best practices are now contented: <strong>Kubernetes</strong> is here, and has instantly been adopted by all architects and DevOps to prepare their ideal microservices architecture.</p><p>And when you dig into Kubernetes and associated frameworks Istio and Envoy, what do you find? REST APIs of course, but that was expected by definition of microservices. What you also find is Async APIs for each microservice. Kaboom! Backends now have a blueprint for quite a few years: Kafka and Kubernetes to orchestrate data flows, control and scale them. And this blueprint is entirely event-driven.</p><p>To top it all and not surprisingly, languages traditionally used to create application layers linked to databases have also gone through their event revolution with reactive extensions such as <strong>RxJava</strong> widely adopted by the Java community. Now the entire backend dev, DevOps, and IT community can focus on what will make them able to compete with agility over the web.</p><p><strong>Entire System</strong> If we take a helicopter view to look at the entire system, what does the event-driven revolution?</p><p>The entire chain from data plane, app, API, network down to the frontends are event-driven: Event flow between IoT devices (eg GPS), mobile apps, and ML without any barrier, people can all align thinking in terms of data streams of events rather than databases, considering intensity and relevance of feeds rather than states and calls.</p><p>As all components have embraced the EDA revolution (I prefer evolution -revolution is always violent- as this is happening with live clients to be supported at the same time, hence slowly, but based on a Darwinian selection of design). This new <strong>end-to-end chain of events</strong> brings simplicity and efficiency to components individually and collectively.</p><h3>Question 2: What is the consumption experience of event-driven APIs? How much does it differ from RESTful APIs?</h3><p>Fran: You usually connect to a broker to start sending and/or receiving messages. Examples of brokers are RabbitMQ, Kafka, Solace, etc. The difference with synchronous patterns like REST, gRPC, and GraphQL is that you don’t ask for information. You don’t make requests.</p><p>Just connect to the broker and subscribe to a channel of your interest. Eventually (no pun intended), you’ll get this information, such as when the events occur. Similarly, you’ll send events to the broker whenever they occur to you. (replace you with your service)</p><p>Jonathan: Today, the consumption experience generally sucks! This is a gap in the market AsyncAPI is solving and Solace [and several other companies, organizations, and individuals] is looking to help solve too… help make the experience as pleasant as with RESTful APIs.</p><p>David, Eric: There are many different technologies for providing event driven APIs</p><ul><li>Server-Sent Events (SSE) for pushing data to the client to provide reactive user experience</li><li>Webhooks for making HTTP callbacks on state change</li><li>HTML5 Websockets providing full-duplex communication channels over a single TCP connection between client and server.</li><li>MQTT and AMQP for IoT use cases</li></ul><h3>Question 3: Today, how can I test drive event-driven APIs out in the wild?</h3><p>Fran: You can’t or is not easy. Part of the reason for the existence of AsyncAPI is precisely to enable that.</p><p>Jonathan: Check out cloud.solace.com… Sign up for an account (free) and play around. Create an event broker (think of that like an API gateway) and look at our runnable code pen examples. This is all more infrastructure based…. Now imagine AsyncAPI and you layer that on top, it becomes more like API management. Also, check out this blog <a href="https://solace.com/blog/api-management-event-management/">https://solace.com/blog/api-management-event-management/</a></p><p><em>(Emmelyn) I can inspect HTTP calls in a browser and mock up the API and run it in Postman and get an InVision conceptual app to consume the API, for example. </em><br> <br>You can kind of do this at Cloud.solace.com… Again its more how do you send/receive messages… with no app context, but its a start and the most useful I have ever found.</p><p><em>(Emmelyn) I can use Stoplight to lint or APIMatic to create starter SDKs to “try it”</em></p><p>See previous comment and try it out and give me feedback!</p><p>David, Eric: Many of the traditional testing tools do not natively support event- driven protocols and are built for standard HTTP request/response found in REST based services. In order to help support the testing of SSE a client SDK is provided to help build automated clients to test both functional and non-functional aspects of SSE services.</p><h3>Question 4: How do you measure event-driven APIs?</h3><p>Fran: You don’t have “calls” because you don’t ask for anything. You just show your interest in certain types of events and wait for them to happen. That’s a subscriber/consumer. If you’re building a publisher/producer then you’ll send events to the broker when certain events occur. The consumers interested in your type of event will receive them.</p><p>Jonathan: I think effectiveness is a direct correlation between consumption. An event that is consumed 0 times was actually worth 0… an event consumed 100 times is valuable. Yes, you do this via dashboards. Today, nobody deals with BOTH events and synchronous apis in one platform.</p><p>David, Eric: The enhanced user experience in the client can be tracked by NPS of the service.</p><p>Traditional monitoring tools are optimized for request-response scenario, asynchronous scenarios bring in additional complexity where by a single request could result in N number of responses been relayed to the client. Adoption of OpenTracing can help to see the spans of distributed transactions.</p><h3>Value Proposition of Event-Driven Architecture</h3><p><strong>Business leaders only care about event-driven architecture as much as the end result produces the business-driving experience that specifically this type of architecture can produce. What would you explain to this type of audience?</strong></p><p>Fran’s take</p><ul><li>Real-time experiences are built with event-driven architectures. You can’t build something real-time with REST APIs or any of the aforementioned styles.</li><li>Solace has an interesting concept called “event mesh” and <a href="http://twitter.com/Jonathan">Jonathan</a> Schabowsky can explain better and point to existing documentation.</li><li>And then you have the cool thing about event-driven microservices, which allows you to build products faster. It allows you to spend less time worrying about the systems and more about your business logic.</li></ul><p>Eric’s take</p><p>As a CEO myself, I do not invest in any feature of any technology until I can see proven track records and numbers. Well, EDA actually brings sizeable benefits in terms of topline, customer satisfaction, cost base, competitive differentiation and even HR attractiveness and retention:</p><ul><li>Topline</li><li>Customer Satisfaction</li><li>Outsmart your Competitors with Machine Learning ML</li><li>IT Cost Base Reduction</li><li>HR</li></ul><p><strong>Topline</strong> The main benefits of the EDA mindset is to optimize the time it takes between the occurrence of an event in the universe, and the reaction by the company to that event. Each industry has its own metrics for assessing such benefit. In capital markets, 1 millisecond is worth <a href="https://research.tabbgroup.com/report/v06-007-value-millisecond-finding-optimal-speed-trading-infrastructure"><strong>$8m</strong></a> (!)</p><p>In marketplace/ecommerce/retail/logistics business where players make as much money as they can reduce the time between an order is placed to them, and such order is shipped to the end user, each second count. I did a quick calculation of how much a second is worth for Amazon, by dividing revenue by the number of seconds in a year: <strong>$50m</strong> (!)</p><p><strong>Customer Satisfaction</strong> As already noted, reactive interfaces are not only the coolest but also the most natural for our human brain. Well, we can put a number on this : <a href="https://perspectives.mvdirona.com/2009/10/the-cost-of-latency/">100 ms is worth 1%</a> additional revenue on any mobile or desktop<strong> </strong>app, from mobile banking to eCommerce.</p><p>For developers using your API, when you can provide your most important customer the ability to consume your API without limit, they are happy to pay a premium for it as it means they have more data to make decision upon, and it helps streamlining their data ingestion chain.</p><p><strong>Outsmart your Competitors with ML</strong> A major way for technology to bring value to customer interactions (chatbots), optimize processes and make best-possible decisions in an ever-changing world is Machine Learning. And Machine Learning thrives on data. In the past, data architectures were based on tables and data lakes. Well, for practitioners, data lakes actually became data swamps.</p><p>With some experience, the key is to master data ingestion so that data sources are consumed upstream, as close as possible to the source, and turned into value as soon as possible by Machine Learning. Machine Learning consumes streams of data conveyed by Kafka from third party streaming APIs, such as <a href="https://www.xignite.com/product/CloudStreaming">Xignite Cloudstreaming</a>, learn from them as they come and learns from its errors against what it had predicted.</p><p>And the closer you are from data sources (acting as senses for the ML brain), the smarter your ML brain will be, faster than your competition. There is even a family of ML updating their model for each, called <a href="https://moa.cms.waikato.ac.nz/">Massive Online Analysis</a> or Streaming Algorithms. EDA has even invaded ML.</p><p><strong>IT Cost Base Reduction</strong> With the 90%+ gains in efficiency that we have mentioned earlier, an API vendor servicing clients with events can not only make them happier, but also reduce its cloud bill on CPU and network by 90% on these use cases. As cloud bills have become so high and critical to become visible to executive desks, they should understand the impact.</p><p><strong>HR</strong> I believe we would agree that developers and DevOps like new and efficient [tools]. And events are fun while bringing efficiency. Working with events makes your organization more attractive to new developers, and help keeping them happy while working on cutting-edge technology. As developers cannot meet the demand while “software is eating the world” this [circumstance] alone could be a driver for considering and EDA and API and Microservices first approach in your organization.</p><h3>Question: How and when should companies actively choose event-driven architecture?</h3><p><em>For example, how is discovery performed? Which APIM catalogs can showcase event-driven APIs so that companies can decide to consume rather than build it themselves? What kind of guidance and best practices can we provide?</em></p><p>David and Eric advise</p><p><strong>IoT</strong> — Many companies are looking to leverage IoT in their offering such as car insurance (pay as you drive), healthcare (health monitoring).</p><p>A vast variety of low-price devices (sensors, thermostats, robots, etc.) with internet connectivity is flooding the market. Their value does not lie in the hardware but in the services that are/will be attached to it. The infrastructure that support these services must have a real-time processing capability to provide real value back to the consumer.</p><p><strong>MQTT</strong> is becoming the de-facto standard to exchange data with IoT devices because it is simple and efficient on low memory devices. It is also based on a Publish/Subscribe model that will force the companies to adopt an EDA in order to process all the generated events.</p><p><strong>Cloud</strong> — Adaption of cloud-based services in Enterprise adds more integration patterns (Cloud-to-cloud, Cloud-to-ground, Ground-to-Cloud, ..) thus more complexity. The different systems needs to exchange data to keep a coherent state across the applications. Using a point-to-point integration strategy can quickly become a trap because of the exponential complexity and tight coupling between the different systems.</p><p>EDA enables a loose coupling between the components/apps in order regain agility, increases innovation pace and reduces time-to-market for new features/services.</p><blockquote>In general, I’m studying API Design Patterns and how various kinds of APIs expedite R&amp;D adoption. What business results are accelerated from these kinds of patterns?</blockquote><blockquote>Thanks for suggesting topics and connecting with feedback. You can also reach me via Twitter @lifewingmate DM or via the AsyncAPI community via our <a href="https://github.com/asyncapi">Github</a> or <a href="http://asyncapi.slack.com/">Slack channel</a>.</blockquote><p>Disclaimer: The professional opinions of those interviewed do not necessarily reflect the organizations they represent. These interviewees volunteered their time and contributions to support the AsyncAPI initiative and community.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9cc59335cfc0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/an-api-strategist-explores-event-driven-apis-9cc59335cfc0">An API Strategist Explores Event-Driven APIs</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 15, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-15-2019-e7fcf070ba15?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/e7fcf070ba15</guid>
            <category><![CDATA[event-driven-systems]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[talks]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[golang]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 11 Apr 2019 10:03:34 GMT</pubDate>
            <atom:updated>2019-04-17T13:36:16.907Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*sCdIi98tNKHTjB83" /><figcaption>Photo by <a href="https://unsplash.com/@perrygrone?utm_source=medium&amp;utm_medium=referral">Perry Grone</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p>Kaixo lagunak! This week we continued working on the parser as it’s a top priority for us. We made significant progress and plan to release a simple but functional version soon.</p><h3>Building the parser</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gbigLiu58F5yBWjcPkH-dg.png" /></figure><ol><li><a href="https://github.com/asyncapi/parser/pull/31">We’ve added support for AsyncAPI 1.x/OpenAPI schemas.</a> This is the first step before we dive into Avro and Protobuf support.</li><li><a href="https://github.com/asyncapi/parser-nodejs">Updated the Node.js to automatically test itself on Linux, Mac, and Windows.</a> We’re still struggling to debug some failures on Windows so <strong>we encourage people who work on this operating system to join and help us test</strong>.</li></ol><h3>Talks</h3><p>I’ll be speaking about AsyncAPI and event-driven architectures in a few conferences. Let’s connect!</p><ul><li><strong>API Days Madrid</strong> (April 26): <a href="http://apidaysmad.apiaddicts.org/schedule/#session-2">http://apidaysmad.apiaddicts.org/schedule/#session-2</a></li><li><strong>Gartner AADI </strong>(May 20–21): <a href="https://www.gartner.com/en/conferences/emea/applications-uk">https://www.gartner.com/en/conferences/emea/applications-uk</a></li><li><strong>KubeCon Europe</strong> (Barcelona, May 22–23). Not speaking but let’s meet there! <a href="https://events.linuxfoundation.org/events/kubecon-cloudnativecon-europe-2019/">https://events.linuxfoundation.org/events/kubecon-cloudnativecon-europe-2019/</a></li><li><strong>REST Fest Europe</strong> (Wrocław, May 31): <a href="http://2019.restfest.org/eu/schedule">http://2019.restfest.org/eu/schedule</a></li><li><strong>API Days Finland</strong> (Helsinki, June 4–5): <a href="https://www.apidays.fi/">https://www.apidays.fi/</a></li></ul><h3>Donate</h3><p>And last but not least, we’re running a <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. We’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6O84mJunpTpRKUk7J3Tk-w.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“People who think they know everything are a great annoyance to those of us who do.”</blockquote><blockquote>— Isaac Asimov</blockquote><p>😄</p><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e7fcf070ba15" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-15-2019-e7fcf070ba15">Status update (week 15, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 14, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-14-2019-f2e86c6050?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/f2e86c6050</guid>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[game-development]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[event-driven-systems]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 04 Apr 2019 16:11:03 GMT</pubDate>
            <atom:updated>2019-04-04T16:11:03.698Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*20-6DDSjgH1EvyIr1txEWQ.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/photos/vrbZVyX2k4I?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Markus Spiske</a> on <a href="https://unsplash.com/search/photos/born?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Hallo meine Freunde! This week we’ve done great progress on tooling and documentation. It will be our main focus for the next month. In the meantime, check out what we did last week. Read on!</p><h3>Improvements on documentation</h3><ol><li>Check out version 2.0.0-rc1 of the specification in our website: <a href="https://www.asyncapi.com/docs/specifications/2.0.0-rc1/">https://www.asyncapi.com/docs/specifications/2.0.0-rc1/</a>.</li><li>Added a “Hello world” article to our getting started guide: <a href="https://github.com/asyncapi/asyncapi.github.io/pull/3">https://github.com/asyncapi/asyncapi.github.io/pull/3</a>.</li><li>Added a “Servers” article to our getting started guide, explaining how the “servers” section of AsyncAPI works: <a href="https://github.com/asyncapi/asyncapi.github.io/pull/4">https://github.com/asyncapi/asyncapi.github.io/pull/4</a>.</li></ol><h3>Improvements on tooling</h3><ol><li>We managed to compile <a href="https://github.com/asyncapi/parser">our Go parser</a> to Linux, Mac, and Windows C shared objects. This sets the base for an automated building process.</li><li>As a result of the previous point, we managed to create a Node.js wrapper for the Go parser, making use of the C shared objects.</li></ol><h3>AsyncAPI SIG meeting</h3><p>We had our bi-weekly SIG meeting this week where we talked about future plans and how to improve onboarding. And it’s now uploaded to <a href="https://www.youtube.com/channel/UCIz9zGwDLbrYQcDKVXdOstQ">our Youtube channel</a>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FGMFAAyEEqus%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DGMFAAyEEqus&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FGMFAAyEEqus%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/91161e4a0b2f0dc3e828ef2e665d4971/href">https://medium.com/media/91161e4a0b2f0dc3e828ef2e665d4971/href</a></iframe><h3>Donate</h3><p>And last but not least, we’re running a <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. We’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QDzGsHV35Y0Roba48Zol5g.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“Great things in business are never done by one person. They’re done by a team of people.”</blockquote><blockquote>— Steve Jobs</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f2e86c6050" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-14-2019-f2e86c6050">Status update (week 14, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Replicating the Success of REST in Event-Driven Architecture]]></title>
            <link>https://medium.com/asyncapi/replicating-the-success-of-rest-in-event-driven-architecture-c805f83ad2c2?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/c805f83ad2c2</guid>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[event-driven-systems]]></category>
            <category><![CDATA[rest-api]]></category>
            <category><![CDATA[solace]]></category>
            <category><![CDATA[asyncapi]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 04 Apr 2019 08:34:12 GMT</pubDate>
            <atom:updated>2019-04-04T08:34:12.423Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hEdj2i93YuiJoD9GkrOE_A.png" /></figure><p><em>This post is a collaboration between </em><a href="https://solace.com/blog/author/franmendez/"><em>Fran Méndez</em></a><em> of AsyncAPI and Solace’s </em><a href="https://solace.com/blog/author/jonathan-schabowsky/"><em>Jonathan Schabowsky</em></a><em>. It was originally published at </em><a href="https://solace.com/blog/rest-event-driven-architecture/"><em>Solace’s blog</em></a><em>.</em></p><p>Jonathan explained in his <a href="https://solace.com/blog/downside-event-driven-architecture/">last blog post</a> how the loose coupling of applications associated with event-driven architecture and publish/subscribe messaging is both a strength and a weakness. As part of that, he touched on the fact that request/reply interactions using RESTful APIs are still the dominant application integration paradigm, even in hybrid cloud, machine learning and IoT use cases that benefit from event-driven interactions. There’s still tons of use cases for which RESTful request/reply interactions are perfect, but it’s important to be able to mix and match the right exchange pattern (Command, Query and Event) for the job especially where event-driven would be best suited.</p><p>In many cases, exploring why one thing has established or maintained popularity can help you understand why something else isn’t quite as hot, even though it seems like it should be. With this post I’ll investigate why the use of RESTful APIs is still so prevalent, and see if the reasons for its persistent popularity might act as a blueprint for making event-driven popular and mainstream. So, how did REST come to be the most popular way to connect applications? And why does everyone think it’s so easy?</p><h3>How did REST get to be so hot?</h3><p>REST’s popularity arose out of the need for data exchange and interactions between the web browser and backend services. In that context it became a de facto standard because it integrated so well with JavaScript and was so much easier than SOAP (a decent protocol that became bloated and complicated over time). From there, developers started using REST to connect internal enterprise applications, IoT devices and even microservices. It might not have been the best fit for all those use cases, but it got the job done.</p><p>As <a href="https://twitter.com/MattMcLartyBC">Matt McLarty</a> mentions in his blog post <a href="https://www.infoq.com/articles/overcoming-restlessness?utm_source=sumome&amp;utm_medium=twitter&amp;utm_campaign=sumome_share">Overcoming RESTlessness</a>, a complete examination about why REST started to be used in places that it’s not ideal for “would ignore the power that comes from REST’s universality.” He’s referring to the fact that REST has become universal because developers “get it” and it’s surrounded by a thriving ecosystem of complementary technology and tools. Without this ecosystem that REST inherited from the web world, that universal adoption simply would <em>not</em> have happened.</p><h3>The Building Blocks of REST’s Success</h3><p>If you look closely at this ecosystem (foreshadowing) you can see that it’s composed of some foundational components upon which the open source and vendor community have built what I’ll call “enablement tooling.” Here’s what I mean:</p><h3>Foundational Components</h3><ul><li><strong>Web servers</strong> were the workhorse of the web for years before REST came into existence. They were much simpler than the application servers of the time and optimized to deal with large numbers of lightweight request/reply communications interactions like serving up a web page that somebody requests.</li><li><strong>Development frameworks</strong> like Spring, JAX-RS, Restlet and Node.js reflect the fact that people invested time and energy to make the developer experience easy, i.e. keeping them from having to write boilerplate connection code so they could focus on the hard part of developing and refining business logic.</li><li><strong>Security frameworks</strong> like OAUTH for authentication and authorization, and TLS for encryption, established the means by which interactions and information can be made secure.</li></ul><h3>Enablement Tooling</h3><ul><li><strong>API Management</strong>: Companies like Apigee and MuleSoft built platforms that provide an API portal so developers can describe and discover APIs in design-time, API gateways to ensure security, management and API mediation, and finally usage analytics which inform which APIs are most and least used. These API management solutions are used increasingly for sophisticated API creation and design, and to act as API marketplaces.</li><li><strong>Runtime API Discovery:</strong> As APIs and applications have become increasingly dynamic and distributed due to continuous delivery, containerization, cloud-bursting, discovery tooling such as Netflix Eureka and Istio/Envoy (service mesh) have been created to reduce the complexity of API clients and enable them to connect to services anywhere.</li><li><strong>Specification for API Description:</strong> OpenAPI was created as a machine-readable metadata specification in order to document, design and consume APIs. This is incredibly valuable for use by testing tools, clients and document generation.</li><li><strong>Code Generation Tools: </strong>Swagger and its associated code generation tooling lets developers easily take an OpenAPI definition and generate either client or server code, drastically reducing the amount of work it takes development teams to use APIs.</li></ul><p>Without the foundational components, not only would the enablement tooling not have been <em>possible</em>, there wouldn’t have been any need or demand for it. This ecosystem of tools has facilitated REST’s ascension to its position as the de facto standard for application interactions today. While I lament the fact that event-driven hasn’t achieved this same level of adoption and devotion, I understand why, and know that without similar tooling it never will.</p><h3>How Event-Driven is Following in REST’s Footsteps</h3><p>There is no reason why the event-driven world can’t learn from the RESTful API world by leveraging and developing similar foundational components and enablement tools. In fact, some very exciting initiatives are underway and picking up steam in the industry and within Solace:</p><h3>Foundational Components</h3><ul><li><strong>Event Brokers:</strong> This one is easy as many simple (RabbitMQ, ActiveMQ) and advanced event brokers (Solace PubSub+, Kafka) exist today. Many of them are battle-tested and used widely in organizations that are event-driven.</li><li><strong>Development Frameworks:</strong> Spring Cloud Stream makes writing event-driven microservices easy, and Paho for MQTT makes it easy to create event-driven IoT sensors in many programming languages.</li><li><strong>Security:</strong> Frameworks like OAuth enable authentication and authorization in the event-driven world along with TLS for encryption for confidentiality/integrity.</li></ul><h3>Enablement Tooling</h3><ul><li><strong>Event Management:</strong> While <a href="https://solace.com/what-is-an-event-broker/">advanced event brokers</a> perform many functions similar to those of an API Gateway, no vendor offers a platform that does everything for events that API management platforms do for RESTful API interactions. There are no “event portals” for developers to use, for example, in order to design, document and discover events.</li><li><strong>Runtime Event Discovery:</strong> In the Eventing world, the ability to deliver events to consumers is even more complicated than with APIs because of the combination of 1-many event distribution, guaranteed and in-order quality of services along with event producers and consumers being just as dynamic and distributed as what is found with APIs. This has challenged infrastructure and operations teams for years all while client applications should not be burdened with these complexities. The <a href="https://solace.com/what-is-an-event-mesh/">event mesh</a> is an emerging architectural concept that provides similar functionality to the <a href="https://www.nginx.com/blog/what-is-a-service-mesh/">service mesh</a> but is targeted towards asynchronous interaction patterns. This removes the complexities previously described by enabling producers and consumers to exchange events regardless of where they are physically deployed all while maintaining event delivery qualities of service.</li><li><strong>API Description Specification:</strong> <a href="https://www.asyncapi.com/">AsyncAPI</a> is on a mission to standardize event-driven API interactions and support the wide variety of messaging systems available. This is a corollary to <a href="https://swagger.io/docs/specification/about/">OpenAPI</a> — a universal language for all the different messaging protocols and event schema formats. The purpose of AsyncAPI is to enable architects and developers to specify the event payload definition, channel name, application/transport headers and protocol– thus fully specifying the application’s event-driven interface. This was previously not available but, thanks to Fran Méndez and the AsyncAPI Initiative, event-driven applications will receive the same love as RESTful APIs.</li><li><strong>Code Generation Tools:</strong> AsyncAPI is also working in this direction. For instance, the ability to take an AsyncAPI definition and generate event-driven applications is underway for Spring Cloud Stream. This will drastically reduce the effort to create new applications!</li></ul><h3>Conclusion</h3><p>EDA’s popularity has started to drastically increase as many companies are realizing they MUST react in real-time to their customers, decouple their systems and transform into event-driven organizations. However, for event-driven interactions to achieve the same level of adoption as REST, the build-out of tooling for eventing must continue. Now is the time to transform and support all the patterns modern applications need for interaction, i.e. commands, queries… and events!</p><p>Solace is committed to helping organizations realize the advantages of being event-driven. We’re active on all these fronts by continuing to advance the state of the art with our PubSub+ event broker and <a href="https://solace.com/blog/event-mesh/">event mesh</a>, enthusiastically supporting <a href="https://solace.com/blog/unlock-potential-spring-cloud-stream/">Spring Cloud Streams</a>, and actively contributing expertise and financial support to <a href="https://www.asyncapi.com/">AsyncAPI</a>. Stay tuned for more information around how event management and API management are similar, how it is a key capability that organizations need, and what Solace is doing about it!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c805f83ad2c2" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/replicating-the-success-of-rest-in-event-driven-architecture-c805f83ad2c2">Replicating the Success of REST in Event-Driven Architecture</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 13, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-13-2019-b28f4165de9?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/b28f4165de9</guid>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[event-driven-systems]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[iot]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 28 Mar 2019 17:48:27 GMT</pubDate>
            <atom:updated>2019-03-28T17:48:27.585Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0ihR7oXfnxQ0dBsut5t5cA.png" /><figcaption>Photo by <a href="https://unsplash.com/photos/PAykYb-8Er8?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Ian Schneider</a> on <a href="https://unsplash.com/search/photos/celebration?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Pozdrowienia, przyjaciele! This week is a special one. We just finished implementing all the features of version 2.0.0 that affect the specification 🎉</p><p>It means we’re now entering the review period. <strong>We encourage everyone familiar with the specification to start reviewing the new version and submit your questions and suggestions to our </strong><a href="https://github.com/asyncapi/asyncapi/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc"><strong>Github repository</strong></a>. We make AsyncAPI together!</p><p>If you’re not yet that familiar with the specification, don’t worry, we’ll be publishing lots of tutorials and documentation soon, from “hello world” to real use cases. Stay tuned!</p><h3>Latest changes in the specification</h3><ol><li><a href="https://github.com/asyncapi/asyncapi/pull/186">Rename scheme as protocol</a>. It’s been a source of confusion for quite a long time now. “Scheme” is also used for security schemes and the word is very similar to “schema” (in the end most of us are not native English speakers.) No more confusions, call them protocols!</li><li><a href="https://github.com/asyncapi/asyncapi/pull/184">Remove the stream object</a>. Now that we have protocolInfo objects and channels, there’s no need to keep using the stream object. <strong>This doesn’t mean we get rid of the support for HTTP streaming APIs; on the contrary, we’re preparing the foundations to construct a bigger building</strong>. Check out the example in the pull request to learn more about the change.</li><li><a href="https://github.com/asyncapi/asyncapi/pull/183">Add support for RAML-like traits</a>. <a href="https://raml.org/">RAML</a> has been offering <a href="https://raml.org/developers/raml-200-tutorial#traits">a feature called traits</a> for a long time and they’ve proven to be very useful. Avoid repetitive blocks in your AsyncAPI documents. Use traits!</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*ncd5OAIAENfYtqlRN9orCA.png" /><figcaption>Cumulative flow chart showing the progress of AsyncAPI 2.0.0.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*Q1V5adoA6v6QrEpFdgPprw.png" /><figcaption>Release report showing the planned and predicted end dates. We’re 23 days ahead of plan! 🔥</figcaption></figure><h3>Donate</h3><p>And last but not least, we’re running a <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. We’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QDzGsHV35Y0Roba48Zol5g.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“Great things are done by a series of small things brought together.”</blockquote><blockquote>— Vincent Van Gogh</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b28f4165de9" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-13-2019-b28f4165de9">Status update (week 13, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 12, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-12-2019-e5a9c07326bc?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/e5a9c07326bc</guid>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[gamedev]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[open-source]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 21 Mar 2019 18:39:14 GMT</pubDate>
            <atom:updated>2019-03-21T18:39:14.304Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*e7O7A5rM79ng6b8Dc3EbEg.jpeg" /><figcaption>Photo by <a href="https://unsplash.com/photos/RLw-UC03Gwc?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Glenn Carstens-Peters</a> on <a href="https://unsplash.com/search/photos/planning?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Salut mes amis! While writing these lines today, I got surprised by great news: the folks at <a href="https://www.businesswire.com/news/home/20190320005892/en/">Streamdata.io and Axway joined forces to advance event-driven APIs for its customers</a>. It makes me so happy to see that <a href="https://twitter.com/StreamdataIO/status/1108760958544031744">AsyncAPI has contributed to “validate Streamdata’s focus”</a>. Congratulations!</p><h3>Welcome Jonathan Schabowsky</h3><p>I would like to welcome <a href="https://twitter.com/JSchabowsky">Jonathan Schabowsky</a>, senior architect in Solace’s Office of the CTO. Jonathan has made significant contributions to the specification in the last months, especially those making sure we establish solid foundations to grow the specification in a healthy way. And also caring about the future of the initiative, leading the Solace sponsorship. Bienvenido, amigo!</p><h3>Onboarding new members</h3><p>This week I also started an effort to document everything you need to know in order to become a regular contributor or maintainer of the project.<a href="https://github.com/asyncapi/asyncapi/pull/181"> I’m asking for your help here</a>, especially if you know nothing about AsyncAPI, because that will give us a unique vision of the documentation and what needs to be improved.</p><h3>Plans, plans, plans</h3><p>Now that we have version 2 of the specification almost finished, I wanted to take the opportunity this week to start planning and coordinating efforts for the release announcement. There’s not much to share here yet but I can only tell you it’s going to be sweet. Stay tuned!</p><h3>AsyncAPI SIG meeting</h3><p>We had our bi-weekly SIG meeting this week where we discussed the plans and the release announcement. And it’s now uploaded to <a href="https://www.youtube.com/watch?v=vfYr5Mi6Qfw">our Youtube channel</a>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FBsx3nUA0CsA%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DBsx3nUA0CsA&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FBsx3nUA0CsA%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/3014349e87f61c912d02b8e22c3d2f04/href">https://medium.com/media/3014349e87f61c912d02b8e22c3d2f04/href</a></iframe><h3>Donate</h3><p>And last but not least, we’re running a <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. We’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QDzGsHV35Y0Roba48Zol5g.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“All you need is the plan, the road map, and the courage to press on to your destination.”</blockquote><blockquote>— Earl Nightingale</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e5a9c07326bc" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-12-2019-e5a9c07326bc">Status update (week 12, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 11, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-11-2019-3a108eaab228?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/3a108eaab228</guid>
            <category><![CDATA[mqtt]]></category>
            <category><![CDATA[message-broker]]></category>
            <category><![CDATA[solace]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[open-source]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 14 Mar 2019 18:57:55 GMT</pubDate>
            <atom:updated>2019-03-14T18:57:55.478Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5qZJVwH_bw7UFVWOn20e9w.png" /></figure><p>Olà amigos! This is an exciting week for us. <a href="https://solace.com/">Solace</a> joined the AsyncAPI Initiative as a <a href="https://opencollective.com/asyncapi">Platinum Sponsor</a> 🎉. Thank you, <a href="https://twitter.com/JSchabowsky">Jonathan Schabowsky</a>, for leading the process. And if that’s not enough, we finished our “<a href="https://github.com/asyncapi/asyncapi/milestone/2?closed=1">Protocol mappings</a>” milestone. Keep reading!</p><h3>Latest changes in the specification</h3><ol><li><a href="https://github.com/asyncapi/asyncapi/issues/42">Protocol-specific information</a>. This is a long-awaited feature. It allows you to define protocol-specific information in channels, operations, and messages.</li><li><a href="https://github.com/asyncapi/asyncapi/issues/112">Separate application headers from protocol headers</a>. From now on, all the protocol-related headers must be included in a protocolInfo object, and the existing messageheaders must be used only for application headers, if any.</li><li><a href="https://github.com/asyncapi/asyncapi/issues/171">Annotate a channel parameter as multi-level</a>. Do you want to tell your code a specific parameter should span across multiple levels? Now you can annotate channel parameters to indicate that. Pretty much like the# (pound) symbol in AMQP and MQTT or the &gt; (greater than) symbol in NATS.</li><li><a href="https://github.com/asyncapi/asyncapi/issues/172">Unify </a><a href="https://github.com/asyncapi/asyncapi/issues/172">examples fields</a>. In the past, we had the example field in schemas and the examples field in some other places of the specification. They are now unified using the plural form and allowing for more than an example per object. However, we didn’t remove the support for example field in schemas, since we want to remain compatible with OpenAPI schemas.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/962/1*o9OLqaSlRft8ZhvvImfnJA.png" /><figcaption>Cumulative flow chart showing the progress of AsyncAPI 2.0.0.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/961/1*zDbF7EO_ubt0GDBjTJTSKg.png" /><figcaption>Release report showing the planned and predicted end dates. We’re 38 days ahead of plan! 🔥</figcaption></figure><h3>Donate</h3><p>And last but not least, we’ve completed our first three weeks running the <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. We greatly appreciate the support from Solace and look forward to working together. You’ll hear more exciting news in the upcoming weeks. Stay tuned!</p><p>Remember, we’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QDzGsHV35Y0Roba48Zol5g.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“Give yourself permission to slow down. You can speed up by slowing down.”</blockquote><blockquote>— Gabby Bernstein</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3a108eaab228" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-11-2019-3a108eaab228">Status update (week 11, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 10, 2009)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-10-2009-77eaf22536a0?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/77eaf22536a0</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[go]]></category>
            <category><![CDATA[event-driven-systems]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 07 Mar 2019 10:20:52 GMT</pubDate>
            <atom:updated>2019-03-14T17:44:10.949Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qQ2vx5r1VJvlnFwOaJEC5g.png" /><figcaption>Photo by <a href="https://unsplash.com/photos/uvMSarsRHzo?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Matt Antonioli</a> on <a href="https://unsplash.com/search/photos/drill?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><h3>Status update (week 10, 2019)</h3><p>Hola amigos! This week we’ve made a lot of progress on tooling planning and coding. We also started writing code for the future Go parser. Things are progressing well and ahead of time. Keep reading to learn about the latest improvements.</p><h3>Latest changes in the specification</h3><ol><li><a href="https://github.com/asyncapi/asyncapi/pull/167">Add support for correlation IDs</a>. <a href="https://medium.com/u/50756b301413">Chris Wood</a> took care of the implementation of correlation IDs. They are often used to implement request/reply patterns over messaging protocols and distributed tracing features. Check out <a href="https://github.com/asyncapi/asyncapi/pull/167/files#diff-2c9661cac5d56205e2491c6bb8e4991cR57">the examples Chris added</a> to the Pull Request because they’re really helpful. Thanks, Chris!</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/963/1*8myupMnKDgOd3Xx_Qr8SgQ.png" /><figcaption>Cumulative flow chart showing the progress of AsyncAPI 2.0.0.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*wk0W5AbRO9MILLzkhW3UlQ.png" /><figcaption>Release report showing the planned and predicted end dates. We’re 43 days ahead of plan! 🔥</figcaption></figure><h3>Tooling</h3><p>For those who are not aware of the <a href="https://github.com/asyncapi/asyncapi/blob/master/plans/TOOLING.md">tooling plans</a>, we decided to implement the parser for the next AsyncAPI version using the Go programming language. The rationale is that, given our limited resources, we need to automate as much as possible, and Go allows us to generate C modules that we can later import to create parsers in other programming languages. We know that’s far from perfect, but it’s what we can realistically afford with the number of people working on the project. And that’s why your contributions and <a href="https://opencollective.com/asyncapi">donations</a> are crucial.</p><p>Either you know Go or want to learn by doing, please join us to help develop the <a href="https://github.com/asyncapi/parser/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-desc">AsyncAPI 2.0.0 parser</a>. It’s full of fun challenges!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*WK6jqKtDhXQtQtKUGLElvQ.png" /><figcaption>Progress so far for an initial version of the AsyncAPI 2.0.0 parser.</figcaption></figure><h3>AsyncAPI SIG meeting</h3><p>We had our bi-weekly SIG meeting this week where we discussed the parser implementation and the possibilities to implement the <a href="https://github.com/knative/eventing">Knative Eventing</a> registry using <a href="http://asyncapi.org/">AsyncAPI</a>. And it’s now uploaded in <a href="https://www.youtube.com/watch?v=vfYr5Mi6Qfw">our Youtube channel</a>.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FvfYr5Mi6Qfw%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DvfYr5Mi6Qfw&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FvfYr5Mi6Qfw%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/7397b72e964458290b12e02e4536a281/href">https://medium.com/media/7397b72e964458290b12e02e4536a281/href</a></iframe><h3>Donate</h3><p>And last but not least, we’ve completed our first two weeks running the <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. So far, we have had a great response from the community, and you’ll see updates soon.</p><p>Remember, we’ve got different tiers so that everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zK7sOxmafFrUR_24wiW3_w.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“Long-term consistency trumps short-term intensity.”</blockquote><blockquote>— Bruce Lee</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77eaf22536a0" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-10-2009-77eaf22536a0">Status update (week 10, 2009)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Organizing your AsyncAPI documents]]></title>
            <link>https://medium.com/asyncapi/organizing-your-asyncapi-documents-12bccecd9ced?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/12bccecd9ced</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[api]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[programming]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Fri, 01 Mar 2019 12:29:09 GMT</pubDate>
            <atom:updated>2019-03-01T12:29:47.988Z</atom:updated>
            <content:encoded><![CDATA[<p>A recurring question that I get very often is: “how do I organize my AsyncAPI documents?”. Also, the related one: “I have two services, a publisher and a consumer, should I define both in the same AsyncAPI document?”.</p><p>Let’s break down some best practices and tips to avoid ending up in a hell of unmanageable documents.</p><h3>Organizing Microservices</h3><blockquote>I’m using the term microservices here because it’s the most common type of distributed architecture that you can find nowadays.</blockquote><p>The best practice for organizing AsyncAPI files in your microservices architecture is to have a file per microservice. This way, you end up with multiple independent files that define your application.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*DxwJA1tNl2-WHeVhOQNg4A.png" /><figcaption>One publisher and one subscriber, both sharing the UserSignUp message.</figcaption></figure><p>Microservices are meant to do a single thing and to do it well and, very importantly, they must be independently deployable. However, if you have a publisher and multiple consumers, you quickly end up having something like the following:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*b2VLCswy1jx2OX2PvUBDGA.png" /><figcaption>One publisher and various subscribers. All of them sharing the UserSignUp message.</figcaption></figure><p>It’s clear there’s a dependency between all of them: the <em>UserSignedUp</em> message. If at some point to want to change it, you’ll have to go through all of the files and change it. It’s a tedious task we want to avoid, so we can make use of the <em>$ref</em> capability of AsyncAPI to simplify things:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*S3ztpzerE3X2T8k8tuqYOw.png" /><figcaption>The value of $ref should be “common/messages.json#UserSignedUp”.</figcaption></figure><p>Now, if you have to add something to the <em>UserSignedUp</em> message, it’s just a matter of changing one file. Depending on your setup, you may have to restart your services to get the new definition. However, as simple and straightforward as it may seem, you must take care not to introduce breaking changes in the message definition. Otherwise, you’ll have inconsistent states while some services got the new message definition and some didn’t yet. Here comes the importance of versioning your messages, but that makes for another blog post alone.</p><h4>Common mistakes</h4><p>Since microservices tend to be small in scope, most probably their AsyncAPI document will not be very extensive too. And I found out this is one of the reasons people tend to re-use the same file for many services. They think the file is very small and because the publisher and the subscriber share the same message, why not putting everything there? It’s tempting at first, but the reason why you should avoid doing this is that you lose context and semantics, and it causes problems:</p><ol><li>If a single document contains publish and subscribe for the same channel (topic), how do you know which one is defining what your application does?</li><li>Since your document may contain many channels, how do you know if your application is publishing or subscribing to each channel or just a subset of them? That itself causes more problems, for instance, when you want to generate the documentation for a single service but you have a single file defining your whole architecture and thus a single documentation page describing all the services without any clue which one is doing what.</li></ol><p><strong>An AsyncAPI file is meant to define the behavior of a single application</strong>. You can obviously break the rules and use it to define the whole architecture but expect all sorts of problems to appear because you’re using a hammer to saw a piece of wood. The same way you don’t use a single OpenAPI (Swagger) file to define many of your REST APIs, you shouldn’t use a single AsyncAPI document to define many of your message-driven APIs.</p><h3>Organizing Client-Server</h3><p>Say, for instance, you have a WebSockets API and front-end application using it. The paradigm is very similar to the one we’re used to with HTTP APIs, with the subtle difference that the communication is <a href="https://en.wikipedia.org/wiki/Duplex_%28telecommunications%29#Full_duplex">full duplex</a>, i.e., the client can send and receive messages over the same channel, at any time.</p><p>This case is not very different from the microservices one. If you think about it, we can look at it as a small distributed architecture, where you only have two services: the client and the server. So the recommended best practice is to follow the same approach and have one document for each of the applications — one for the server and another for the client or front-end.</p><h3>Summary</h3><p>I want to reinforce the point that an AsyncAPI file is meant to define <strong>the behavior of a single application</strong>. Keep this always in mind, and everything will make sense to you.</p><p>Happy coding! ✌️</p><p><em>AsyncAPI is an open source project running on donations so</em><strong><em> </em>please, </strong><a href="https://opencollective.com/asyncapi"><strong>consider donating</strong></a><strong>.<em> </em></strong>🙌</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=12bccecd9ced" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/organizing-your-asyncapi-documents-12bccecd9ced">Organizing your AsyncAPI documents</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Status update (week 9, 2019)]]></title>
            <link>https://medium.com/asyncapi/status-update-week-9-2019-8080626a21ce?source=rss----57fbe5a242be---4</link>
            <guid isPermaLink="false">https://medium.com/p/8080626a21ce</guid>
            <category><![CDATA[api]]></category>
            <category><![CDATA[asyncapi]]></category>
            <category><![CDATA[iot]]></category>
            <category><![CDATA[microservices]]></category>
            <category><![CDATA[gamedev]]></category>
            <dc:creator><![CDATA[Fran Méndez]]></dc:creator>
            <pubDate>Thu, 28 Feb 2019 13:10:19 GMT</pubDate>
            <atom:updated>2019-02-28T13:10:19.404Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VZ9BObVccsno7wlgwW2QTQ.png" /><figcaption>Photo by <a href="https://unsplash.com/photos/daYZ8ixAn8g?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Nathan Anderson</a> on <a href="https://unsplash.com/search/photos/peak?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText">Unsplash</a></figcaption></figure><p>Hey folks! Exciting times here at the <a href="https://asyncapi.org/">AsyncAPI Initiative</a>! One more week and one more status update that’s full of great improvements for the Developer Experience. Have a look!</p><h3>Latest changes in the specification</h3><ol><li><a href="https://github.com/asyncapi/asyncapi/pull/158">Separate message from operation object, and change schema format definition</a>. New features require the specification to adapt. In a previous milestone, we added a way to have multiple schemas, but it didn’t work well for what was coming in this milestone. <a href="https://github.com/asyncapi/asyncapi/pull/158">Check it out!</a></li><li><a href="https://github.com/asyncapi/asyncapi/pull/156">Definitions clean-up</a>. <a href="https://twitter.com/JSchabowsky">Jonathan Schabowsky</a> from <a href="https://solace.com/">Solace</a> helped us to clean definitions on the specification. It should be much more clear now. Thanks, Jonathan!</li><li><a href="https://github.com/asyncapi/asyncapi/issues/146">Add root-level id</a>. An AsyncAPI document is meant to describe a specific application and never your whole architecture. Therefore, we’ve added a way to define the id of your application. And it’s required!</li><li><a href="https://github.com/asyncapi/asyncapi/issues/77">Add operation id</a>. From now on, you’ll be able to define an id for your operation (publish or subscribe). Code generators will use this field to know the name of the function they should call, and follows the same rules as <a href="https://github.com/OAI/OpenAPI-Specification/blob/master/versions/3.0.2.md#user-content-operationid">the OpenAPI one</a>.</li><li><a href="https://github.com/asyncapi/asyncapi/issues/79">Add title and name to message object</a>. These two new properties, allow you to assign a name and a title to your messages. Names are for machines and titles are for humans. Use them wisely!</li><li><a href="https://github.com/asyncapi/asyncapi/issues/54">Add content type to the operation object</a>. Last but not least, it is now possible to assign a content type for your message payload. Why isn’t it on the message though? For reusability purposes. The same message can be encoded/decoded in many different ways depending on the operation you’re performing. The message should remain the same independently of how you send it over the wire.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*G-TzvNq-aGul_cWxPrNP9w.png" /><figcaption>Cumulative flow chart showing the progress of AsyncAPI 2.0.0.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/964/1*MLOZkLD0gLmcyEkOVyYdgw.png" /><figcaption>Release report showing the planned and predicted end dates. We’re 54 days ahead of plan! 🔥</figcaption></figure><h3>Donate</h3><p>And last but not least, we’ve completed our first week running the <a href="https://opencollective.com/asyncapi">sponsorship campaign</a>. So far, we have had a great response from the community and you’ll see updates soon.</p><p>Remember, we’ve got different tiers, so everybody can show their love! ❤️</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*zK7sOxmafFrUR_24wiW3_w.png" /></figure><p><a href="https://opencollective.com/asyncapi">Donate here</a>. Help Open Source projects.</p><blockquote>“Discipline is the bridge between goals and accomplishment”</blockquote><blockquote>— Jim Rohn</blockquote><p>See you next week, folks! 👋</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8080626a21ce" width="1" height="1" alt=""><hr><p><a href="https://medium.com/asyncapi/status-update-week-9-2019-8080626a21ce">Status update (week 9, 2019)</a> was originally published in <a href="https://medium.com/asyncapi">AsyncAPI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>