call_end

    • chevron_right

      Erlang Solutions: Common MVP mistakes: How to build smart without overbuilding

      news.movim.eu / PlanetJabber • 5 days ago - 10:35 • 5 minutes

    A Minimum Viable Product (MVP) is your first real signal to the market, your team, and your investors that you’re solving the right problem in the right way. While it’s often mentioned alongside terms like Proof-of-Concept (PoC), prototype, or pilot, an MVP plays a distinct role: validating real value with real users.

    Avoiding common missteps early sets the stage for faster iteration, smarter growth, and long-term success. Startups are under pressure to move quickly, but speed without focus can lead to costly mistakes. Proving value fast is essential, especially with limited resources, but moving too quickly without the right foundation can stall progress just as easily as moving too slowly.

    What an MVP should be

    An MVP is the leanest version of your product that still delivers real value and helps you learn whether you’re solving the right problem.

    It’s not about perfection, but validation. Will users care enough to try, pay, or share?

    Importantly, a strong MVP also signals to investors that you can efficiently test ideas, understand your market, and move fast with limited resources.

    Focus on what matters, build with intent, and treat your MVP not as a throwaway prototype, but as the foundation of everything to come.

    Small by design, smart by strategy

    Popularised by Eric Ries in The Lean Startup , the MVP is designed to reduce wasted time, money, and effort. By building only what’s needed to test your core assumptions, you can learn quickly and adjust early, before burning through too much time, money, or energy.

    A good MVP doesn’t just mean “basic”

    A strong MVP isn’t just a stripped-down prototype. It’s the foundation of your product. Lightweight, but also reliable, secure, and built for change. If it can’t be used, demoed, or trusted, it’s not doing its job.

    Focus on what matters, build with intent, and treat your MVP not as a throwaway prototype, but as the foundation of everything to come.

    Minimise risk, maximise learning

    An MVP helps you move fast and stay focused. It’s not about trial and error. It’s about proving your idea works and showing investors that you’re building something ready to grow.

    Common MVP mistakes (and how to avoid them)

    Building an MVP is about speed and learning. But moving fast shouldn’t mean skipping the fundamentals. Many startups fall into familiar traps: doing too much too soon, choosing the wrong tools, or cutting corners that cause problems later.

    By spotting these mistakes early, you can build smarter, avoid rework, and give your product a better chance of success.

    Overbuilding Before Validation

    Adding too many features at the start slows you down, increases costs, and weakens your core value. A bloated MVP is harder to test, more expensive to maintain, and often confusing for users.

    Why it happens:

    • Unclear priorities
    • Perfectionism
    • Fear of missing out

    How to avoid it:

    Focus on solving one clear problem well. Use low-code or no-code tools to test ideas quickly without overcommitting time or budget.

    Choosing the wrong tech stack

    Selecting technology based on trends instead of fit creates long-term issues. The wrong stack can lead to expensive rebuilds, poor stability, development slowdowns, and scaling challenges.

    Why it matters:

    Your tech choices affect how fast you can iterate, how well you scale, and how easy it is to adapt later.

    How to avoid it:

    Choose a simple, flexible stack that fits your domain. Use tools that support rapid development and long-term growth. Involve technical partners or advisors with experience to help avoid common mistakes.

    Ignoring security and code quality

    When speed trumps structure, the result is often messy, unreliable code.

    A growing trend, vibe coding , uses AI (especially large language models) to quickly generate code from natural language. While this accelerates initial progress, it often skips testing, documentation, and consistency, leading to hidden risks and technical debt.

    Though fast at first, vibe coding can leave fragile code that’s hard to debug, extend, or transfer, with teams diverging in approach and progress stalling over time.

    Why does it happen?

    • Misunderstanding MVP as “low quality” rather than “focused and efficient”
    • Overreliance on AI-generated code without review or standards
    • Lack of experienced engineering oversight

    Risks include:

    • System instability and hidden failures
    • Security vulnerabilities and compliance breaches
    • Technical debt and poor maintainability
    • Loss of trust from investors and partners

    How to avoid it:

    Prioritise quality from day one:

    • Review AI code for security, clarity, and maintainability
    • Apply secure authentication and data encryption
    • Set shared coding standards and style guides
    • Require basic tests and documentation, even for MVPs
    • Limit LLM use in critical paths unless thoroughly validated
    • Track shortcuts and log them as technical debt to resolve later

    A little rigour early on prevents major setbacks down the line.

    What smart MVP development looks like

    A smart MVP is fast, focused, and built for flexibility. It doesn’t aim to include everything, just enough to test your core idea with real users.

    Here’s what that looks like in practice:

    Built fast, not rushed

    Speed should serve as validation. The best MVPs reach users quickly without creating confusion or technical debt.

    Focus on:

    • Delivering one clear value
    • Releasing early to gather feedback
    • Improving in tight, focused cycles

    Easy to change, because feedback is coming

    A smart MVP is flexible by design. Once feedback starts coming in, you need to be ready to adjust quickly without overhauling everything.

    Make this easier with:

    • Modular code
    • Clear documentation
    • A prioritised backlog for fast iteration

    Safe and secure – even if it’s lean

    Even a small MVP needs to be stable and secure. If users are testing it, they’re trusting it.

    Trust depends on:

    • Data security and privacy (including GDPR compliance)
    • A clear, usable interface
    • Consistent, reliable performance

    A strong MVP is :

    • Right-sized: Solves one problem well
    • Stable: Works reliably in demos and tests
    • Scalable: Built on a foundation that can grow
    • Trustworthy: Respects and protects user data

    Smart MVP development means building fast, but building right. When you combine speed with strategy, you don’t just ship faster, you learn faster, improve faster, and grow stronger.

    Build fast. Build smart. Build for growth.

    A strong MVP helps you validate your idea, attract early users or investors, and gather feedback, without overbuilding or overspending. The goal is not just to launch quickly, but to launch with clarity, purpose, and scalability in mind.

    Many teams fall into the same traps: bloated feature sets, the wrong technology choices, or neglecting long-term costs. These missteps waste time, burn cash, and kill momentum. The most effective MVPs are built with focus, tested against the right assumptions, and grounded in a foundation that supports growth from day one.

    At Erlang Solutions, we can help your startup launch MVPs that are resilient under pressure and built for the future. If you’re ready to build something that works, let’s talk .

    The post Common MVP mistakes: How to build smart without overbuilding appeared first on Erlang Solutions .

    • wifi_tethering open_in_new

      This post is public

      www.erlang-solutions.com /blog/common-mvp-mistakes-how-to-build-smart-without-overbuilding/

    • chevron_right

      The XMPP Standards Foundation: The XMPP Newsletter April 2025

      news.movim.eu / PlanetJabber • 5 May • 8 minutes

    XMPP Newsletter Banner

    XMPP Newsletter Banner

    Welcome to the XMPP Newsletter, great to have you here again! This issue covers the month of April 2025.

    Like this newsletter, many projects and their efforts in the XMPP community are a result of people’s voluntary work. If you are happy with the services and software you may be using, please consider saying thanks or help these projects! Interested in supporting the Newsletter team? Read more at the bottom .

    XSF Announcements

    XSF Membership

    If you are interested in joining the XMPP Standards Foundation as a member, submissions are open until May 18th, 2025, 00:00 UTC! .

    XSF Fiscal Hosting Projects

    The XSF offers fiscal hosting for XMPP projects. Please apply via Open Collective . For more information, see the announcement blog post . Current projects you can support:

    XMPP Events

    • Berlin XMPP Meetup [DE / EN]: monthly meeting of XMPP enthusiasts in Berlin, every 2nd Wednesday of the month at 6pm local time.
    • XMPP Italian happy hour [IT]: monthly Italian XMPP web meeting, every third Monday of the month at 7:00 PM local time (online event, with web meeting mode and live streaming).
    • XMPP Sprint in Berlin : On Friday, 23rd, Saturday, 24th, and Sunday, 25th of May 2025.

    XMPP Articles

    XMPP Software News

    XMPP Clients and Applications

    • Cheogram has released version 2.17.10-1 for Android. This version introduces an initial implementation of Spaces ( XEP-503 ), among other improvements, bugfixes and more!
    • Conversations has released versions 2.18.0 , 2.18.1 and 2.18.2 for Android. Notable changes include the ability to pick a custom backup location, a prominent backup restore option for Quicksy , and improved support for more kinds of URIs. The latter includes tel phone numbers, mailto email addresses, and more interestingly the web+ap scheme for ActivityPub proposed by Fedi Links .
    • Dino has released version 0.5 featuring OMEMO encryption by default, improved file transfers, image preview and other file details before downloading, and two completely reworked dialogs. See the release blog post for all the details.
      • At the same time, Dino has also received funding from NLnet to begin development on a slew of new features. This includes message moderation in group chats, local message deletion, modern connection handling with FAST and SASL2, more formatting options with Message Markup, and more! Visit the project page for all the details.
    • Gajim has released versions 2.1.0 and 2.1.1 with a new ‘Activity feed’ page, layout improvements for its ‘Start Chat’ dialog and support for ‘Message Display Synchronisation’ ( XEP-0490 ) across group chats among other improvements and bugfixes. Head over to their News section for all the details.
    Activity feed in Gajim 2.1

    Activity feed in Gajim 2.1

    Account and status selection in Gajim 2.1

    Account and status selection in Gajim 2.1

    • Kaidan has received NLnet funding for various improvement across the board , most notably multi-user chat and support for legacy OMEMO. The second point is significant because while Kaidan is using a newer version of the OMEMO end-to-end encryption protocol, other popular clients including Conversations, Monal, and Dino are still using an older version. Since the two are not compatible, this meant Kaidan users were unable to use OMEMO encryption with users of most other clients. By implementing the older spec as well, Kaidan will help bridge that gap.

    • Monocles Chat 2.0.6 has been released for Android. This version brings initial support for file captions, the option to pin an unencrypted message to the top of a conversation, providers list support, and the option to register on your own XMPP server, among many other new features and improvements.

    Monocles Chat 2.0.6: Initial captions to files and pin message to the top

    Monocles Chat 2.0.6: Initial captions to files and pin message to the top

    Monocles Chat 2.0.6: Register on your own XMPP server or pick one from the providers list

    Monocles Chat 2.0.6: Register on your own XMPP server or pick one from the providers list

    • Movim has released version 0.30 (code named “ Encke ”), the biggest Movim evolution in many years! This release brings multi-participant calls , reactions being displayed in the detailed message view, support for Unicode 15.1 with plenty of new emojis to use, and avatars that change when a contact adds to their Story .
    Movim 0.30 (Encke): Multi Participant Calls. Bob Cat looking disgruntled by the presence of the ‘Hooman’ on the lower right of the screen!

    Movim 0.30 (Encke): Multi Participant Calls. Bob Cat looking disgruntled by the presence of the ‘Hooman’ on the lower right of the screen!

    Movim 0.30 (Encke): Meow OwO bedazzled by the looks of Multi Participant Calls on his mobile device!

    Movim 0.30 (Encke): Meow OwO bedazzled by the looks of Multi Participant Calls on his mobile device!

    • and following right on its heels, Movim also published its first bug-fix release: version 0.30.1 , adding animated pictures support in the image proxy and a new Avatar and Banner Configuration Panel, as well as implementing ( XEP-0392 ) Consistent Color Generation, among many other improvements and bugfixes. Make sure to check out the official announcements at the Movim Blog for all the details!
    Movim 0.30.1: Avatar and banner configuration panel

    Movim 0.30.1: Avatar and banner configuration panel

    XMPP Servers

    • MongooseIM has released version 6.3.3 of its Enterprise Instant Messaging Solution. This minor update includes various fixes and improvements. For more information, check out the documentation .
    • ProcessOne has published ejabberd 25.04 . This release brings an important security fix, several bug fixes and a new API command.
    • Prosody IM is pleased to announce version 13.0.1 , a new minor release from the latest stable branch. It fixes some important bugs that were discovered after the latest release. Read all the details on the release changelog . As always, detailed download and install instructions are available on the download page for your convenience.
    • The Prosody app for YunoHost has been updated to provide a bunch of supported XEPs by default, configured for all YunoHost users in just one click. YunoHost is a set of tools to easily manage your own selfhosted services, and while it used to come bundled with the Prosody fork Metronome by default, it has recently bundled its XMPP functionality into a separate “app” so that people can swap in any other XMPP server of their choice.

    XMPP Libraries & Tools

    Extensions and specifications

    The XMPP Standards Foundation develops extensions to XMPP in its XEP series in addition to XMPP RFCs . Developers and other standards experts from around the world collaborate on these extensions, developing new specifications for emerging practices, and refining existing ways of doing things. Proposed by anybody, the particularly successful ones end up as Final or Active - depending on their type - while others are carefully archived as Deferred. This life cycle is described in XEP-0001 , which contains the formal and canonical definitions for the types, states, and processes. Read more about the standards process . Communication around Standards and Extensions happens in the Standards Mailing List ( online archive ).

    Proposed

    The XEP development process starts by writing up an idea and submitting it to the XMPP Editor . Within two weeks, the Council decides whether to accept this proposal as an Experimental XEP.

    • No XEPs proposed this month.

    New

    • No New XEPs this month.

    Deferred

    If an experimental XEP is not updated for more than twelve months, it will be moved off Experimental to Deferred. If there is another update, it will put the XEP back onto Experimental.

    • No XEPs deferred this month.

    Updated

    • Version 1.1.3 of XEP-0313 (Message Archive Management)
      • Fixed typo (XEP Editor (dg))
    • Version 0.4.0 of XEP-0377 (Spam Reporting)
      • Add spam report processing opt-in.
      • Add Guus der Kinderen as co-author. (gdk)
    • Version 1.0.1 of XEP-0421 (Occupant identifiers for semi-anonymous MUCs)
      • Fixed typo (XEP Editor (dg))
    • Version 0.3.0 of XEP-0455 (Service Outage Status)
      • Remove all in-band event signaling. (mp)

    Last Call

    Last calls are issued once everyone seems satisfied with the current XEP status. After the Council decides whether the XEP seems ready, the XMPP Editor issues a Last Call for comments. The feedback gathered during the Last Call can help improve the XEP before returning it to the Council for advancement to Stable.

    • No Last Call this month.

    Stable

    • No XEPs moved to Stable this month.

    Deprecated

    • No XEPs deprecated this month.

    Rejected

    • No XEPs rejected this month.

    Spread the news

    Please share the news on other networks:

    Subscribe to the monthly XMPP newsletter
    Subscribe

    Also check out our RSS Feed !

    Looking for job offers or want to hire a professional consultant for your XMPP project? Visit our XMPP job board .

    Newsletter Contributors & Translations

    This is a community effort, and we would like to thank translators for their contributions. Volunteers and more languages are welcome! Translations of the XMPP Newsletter will be released here (with some delay):

    • English (original): xmpp.org
      • General contributors: Adrien Bourmault (neox), Alexander “PapaTutuWawa”, Arne, Badri Sunderarajan, Benson Muite, cal0pteryx, emus, Federico, Gonzalo Raúl Nemmi, Jonas Stein, Kris “poVoq”, Licaon_Kter, Ludovic Bocquet, Mario Sabatino, melvo, MSavoritias (fae,ve), nicola, Schimon Zachary, Simone Canaletti, singpolyma, XSF iTeam
    • French: jabberfr.org and linuxfr.org
      • Translators: Adrien Bourmault (neox), alkino, anubis, Arkem, Benoît Sibaud, mathieui, nyco, Pierre Jarillon, Ppjet6, Ysabeau
    • Italian: notes.nicfab.eu
      • Translators: nicola
    • Spanish: xmpp.org
      • Translators: Gonzalo Raúl Nemmi
    • German: xmpp.org
      • Translators: Millesimus
    • Português (BR): xmpp.org
      • Translators: Paulo

    Help us to build the newsletter

    This XMPP Newsletter is produced collaboratively by the XMPP community. Each month’s newsletter issue is drafted in this simple pad . At the end of each month, the pad’s content is merged into the XSF GitHub repository . We are always happy to welcome contributors. Do not hesitate to join the discussion in our Comm-Team group chat (MUC) and thereby help us sustain this as a community effort. You have a project and want to spread the news? Please consider sharing your news or events here, and promote it to a large audience.

    Tasks we do on a regular basis:

    • gathering news in the XMPP universe
    • short summaries of news and events
    • summary of the monthly communication on extensions (XEPs)
    • review of the newsletter draft
    • preparation of media images
    • translations
    • communication via media accounts

    Unsubscribe from the XMPP Newsletter

    To unsubscribe from this list, please log in first . If you have not previously logged in, you may need to set up an account with the appropriate email address.

    License

    This newsletter is published under CC BY-SA license .

    • wifi_tethering open_in_new

      This post is public

      xmpp.org /2025/05/the-xmpp-newsletter-april-2025/

    • chevron_right

      Erlang Solutions: Looking Forward to ElixirConf EU 2025

      news.movim.eu / PlanetJabber • 1 May • 6 minutes

    ElixirConf EU 2025 is only two weeks away! Soon, it will once again be time for the Elixir community from Europe and beyond to come together, connect, and learn from each other about Elixir and its ecosystem. There will be dozens of talks from speakers across the community, training opportunities, and more.

    This year’s event is being held in Kraków, Poland from May 15-16, with optional training available on May 14. Both in-person and virtual tickets are available.

    Here’s a preview of some of the speakers on the schedule, and what to expect from their talks:

    Keynotes

    José Valim: Type System and Elixir Updates

    No ElixirConf would be complete without an appearance from Dashbit’s José Valim , the creator of Elixir.


    Since announcing it at ElixirConf EU 2022 , José and the Elixir team have been hard at work on a type system for the language based on set-theoretic types. Implementation of this system is still underway , and José will offer an update on its progress, as well as other recent activities from the Elixir team and what to expect in the next release. Guillaume Duboc , a PhD student at Université Paris Cité who is working on the type system for his thesis, will provide a more in-depth look at it in a separate talk, “What’s New with Elixir Types”.

    James Arthur: Introducing Phoenix.Sync

    James Arthur is a TED Prize and Microsoft Prize-winning co-founder of Opendesk , Hazy , and Post Urban , with prior software development experience for Apple and IDEO .

    He is currently the CEO of ElectricSQL , developers of the Elixir sync engine Electric , and will be bringing that knowledge to his keynote introducing Phoenix.Sync , a new library that adds real-time sync to Phoenix. Expect a tour through the basics of the library, how to use it to sync data into both backend and frontend applications, and its potential uses in AI systems.

    Davide Bettio and Mateusz Front: The AtomVM and New Horizons for Elixir

    SECO Mind’s Davide Bettio has years of embedded systems and IoT experience, and began work on AtomVM , a lightweight BEAM implementation tailored for IoT devices, in 2017.

    He’ll be joined by Mateusz Front , an Elixir developer at Software Mansion , to explain the results of a collaboration that has proved Elixir can be run on a whole new platform. They’re keeping quiet about the specifics, so you’ll have to check out their keynote to find out what they’ve come up with, along with plenty more information about AtomVM and Elixir runtimes.


    Industry Stories

    Anita Ludermann: Turn Old Into New: Moving to Elixir Feature by Feature

    Anita Ludermann is a software engineer currently at Modell Aachen GmbH , where she has spent the last few years working with a team whose goal is to modernise a legacy codebase by transitioning it to Elixir.

    She’ll be offering a first-hand account of what adoption of Elixir and the functional paradigm in general looks like in a real-world industry context, with all the challenges of shifting an entire system to a new language bit by bit while maintaining consistent functionality for the user.

    Ettore Berardi: How Elixir Powers the BBC: From PoC to Production at Scale

    After a stint at MacMillan Science and Education , Ettore Berardi joined the BBC , where he has now been working for eight years as a tech lead.

    He’ll be relaying another story of Elixir adoption in action, namely the four-year process of shifting to Elixir at the BBC, starting with an initial proof of concept and culminating with nearly all BBC app and web traffic being served by Elixir. On top of the technical side of things, he’ll also discuss the realities of building and maintaining Elixir expertise in an environment largely unfamiliar with the language.

    Libraries and Frameworks

    Gus Workman: Introducing Trellis: Open Source Hardware for Nerves

    You may already be familiar with Nerves , a framework for using Elixir to build embedded systems. Gus Workman is the founder of Protolux Electronics , a company specialising in Nerves-based hardware and software.

    His talk will introduce Trellis, an open-source hardware platform reminiscent of Raspberry Pi , but specifically built with the purpose of producing custom circuit boards for Nerves and Elixir.

    Parker and Shannon Selbert: The Architecture of Oban Web

    Oban Web , the live dashboard for the prominent Elixir background job library Oban , recently became fully open source.

    Now, Oban’s creator Parker Selbert and Oban core team member Shannon Selbert, co-founders of Soren , will be diving into the internals of the project, touring its key elements and interesting features. They promise to reveal how to structure a complex LiveView application with extensive real-world use, backed by Oban Met , a purpose-built distributed database.

    Ridwan Otun and Sola-Aremu ‘Pelumi: Whisperer: An Elixir-Based Multi-Agent Framework

    Ridwan Otun and Sola-Aremu ‘Pelumi are two self-described Elixir evangelists working as engineering vice presidents at Juicyway and Bank of America , respectively.
    Their talk will introduce Whisperer , a new and recently open-sourced framework that allows users to define, sequence, and orchestrate multi-agent AI systems. It’s another talk that’s well worth checking out if you’re interested in learning how Elixir developers are taking advantage of the new possibilities afforded by agents and LLMs.

    Deep Dives

    Conor Sinclair: Meta-Programming in Elixir: Dynamic Function Calling with LLMs

    Multiverse’s Conor Sinclair is a senior software engineer specialising in frontend and with a passion for Elixir.

    He’ll be joining the conference to address the now-ubiquitous topic of LLMs. Conor will be addressing Elixir specifically, walking the audience through the process of creating generic interfaces and using the Langchain library to expose them to LLMs. It’s sure to be a fascinating talk for anyone curious as to how these models interface with an existing application.

    Robert Virding: Unveiling the Magic of Erlang/OTP Behaviours: A Deep Dive in the Codebase & Björn Gustavsson: My Beam History

    We’ll also be getting talks from two developers whose experience with Erlang and the BEAM goes back to the very beginning. Robert Virding was one of the co-creators of Erlang at Ericsson , and now works for Erlang Solutions as Principle Language Expert.
    He’ll be offering a deep dive into the Erlang/OTP codebase, as well as a BEAM VM training session the day before the conference proper. Björn Gustavsson , meanwhile, joined the OTP team in 1996 and later became the main developer for the BEAM, which is partially named for him. His talk will explore his three decades of history with the BEAM.

    And More!

    Of course, that’s only a fraction of the talks at this year’s event – there are plenty of others, and it’s worth checking out the full list of speakers to see if there’s anything else that catches your eye. If you’ve already committed to attending, the schedule is now available, so you can start deciding which talks to attend.

    There’s sure to be something for everyone with an interest in Elixir and the BEAM, and Erlang Solutions will be present at the event as always, so be sure to say hi to our team if you do attend. We hope to see you in Kraków in a few weeks for ElixirConf EU 2025!

    The post Looking Forward to ElixirConf EU 2025 appeared first on Erlang Solutions .

    • chevron_right

      Erlang Solutions: Reduce, Reuse… Refactor: Clearer Elixir with the Enum Module

      news.movim.eu / PlanetJabber • 24 April • 4 minutes

    “When an operation cannot be expressed by any of the functions in the Enum module, developers will most likely resort to reduce/3.”

    From the docs for Enum.reduce/3

    In many Elixir applications, I find Enum.reduce is used frequently. Enum.reduce can do anything, but that doesn’t mean it should. In many cases, other Enum functions are more readable, practically as fast, and easier to refactor.

    I would also like to discuss situations that are a good fit for Enum.reduce and also introduce you to a custom credo check I’ve created, which can help you identify places where Enum.reduce could be replaced with a simpler option.

    Readability

    Here are a few common reduce patterns—and their simpler alternatives.  For example, here’s something I see quite often:

    Enum.reduce(numbers, [], fn i, result -> [i * 10 | result] end)
    |> Enum.reverse()
    
    

    This is a situation that the Enum.map function was designed for:

    Enum.map(numbers, & &1 * 10)
    

    Perhaps you know about Enum.map , but you might see a call to reduce like this:

    Enum.reduce(numbers, 0, fn number, result -> (number * 2) + result end)
    
    

    Let me introduce you to Enum.sum_by !


    Enum.sum_by(numbers, & &1 * 2)
    
    

    Let’s look at something a bit more complex:

    Enum.reduce(numbers, [], fn item, acc ->
      if rem(item, 2) == 0 do
        [item * 2 | acc]
      else
        acc
      end
    end)
    |> Enum.reverse()
    
    

    This is a perfect case for piping together two Enum functions:

    numbers
    |> Enum.filter(& rem(&1, 2) == 0)
    |> Enum.map(& &1 * 2)
    

    Another option for this case could even be to use Enum.flat_map :


    Enum.flat_map(numbers, fn number ->
      if rem(number, 2) == 0 do
        [number * 2]
      else
        []
      end
    end)
    
    

    This is a decent option, but while this achieves the purpose of both filtering and mapping in a single pass, it may not be as intuitive for everybody.

    Lastly, say you see something like this and think that it would be difficult to improve:

    Enum.reduce(invoices, {[], []}, fn invoice, result ->
      Enum.reduce(invoice.items, result, fn item, {no_tax, with_tax} ->
        if Invoices.Items.taxable?(item) do
          tax = tax_for_value(item.amount, item.product_type)
          item = Map.put(item, :tax, tax)
    
          if Decimal.equal?(tax, 0) do
            {no_tax ++ [item], with_tax}
          else
            {no_tax, with_tax ++ [item]}
          end
        else
          {no_tax, with_tax}
        end
      end)
    end)
    
    

    But this is just the same:

    invoices
    |> Enum.flat_map(& &1.items)
    |> Enum.filter(&Invoices.Items.taxable?/1)
    |> Enum.map(& Map.put(&1, :tax, tax_for_value(&1.amount, &1.product_type)))
    |> Enum.split_with(& Decimal.equal?(&1.tax, 0))
    
    
    

    Aside from improving readability, splitting code out into pipes like this can make it easier to see the different parts of your logic.  Especially once you’ve created more than a few lines of pipes, it becomes easier to see how I can pull out different pieces when refactoring.  In the above, for example, you might decide to create a calculate_item_taxes function which takes a list of items and performs the logic of the Enum.map line.

    Performance

    You may have already thought of a counterpoint: when you pipe functions together, you end up creating new lists, which means more work to be done as well as more memory usage (which means more garbage collection).  This is absolutely true, and you should be thinking about this!

    But I find that 99% of the time, the data I’m working with makes the performance difference negligible.  If you find that your code is slow because of the amount of data that you need to process, you might try using the Stream module — it has many of the same functions as Enum , but works lazily.  If that doesn’t work, then by all means, create a reduce (and maybe put it into a well-named function)!

    As Joe Armstrong said:

    “Make it work, then make it beautiful, then if you really, really have to, make it fast.”

    For some information about benchmarks that I’ve run to understand this better, see this analysis and discussion .

    Good Opportunities for Enum.reduce

    Aside from occasional performance reasons, Enum.reduce can often be the simplest solution when you want to transform a data structure over a series of steps.  For example:

    Find Cases in Your Own Code with credo_unnecessary_reduce

    Remember that no one pattern works in all cases, so know what tools you have available! If you’d like to quickly find instances for potential improvements in readability, I built a Credo check to help spot where reduce can be swapped for something simpler.

    You can drop it into your project and start catching these anti-patterns automatically.

    https://github.com/cheerfulstoic/credo_unnecessary_reduce

    Simply add it to your mix.exs file:


    {:credo_unnecessary_reduce, "~> 0.1.0"}
    

    …and then enable it in your .credo.exs file:


    {CredounnecessaryReduce.Check, []}
    

    The post Reduce, Reuse… Refactor: Clearer Elixir with the Enum Module appeared first on Erlang Solutions .

    • wifi_tethering open_in_new

      This post is public

      www.erlang-solutions.com /blog/reduce-reuse-refactor-clearer-elixir-with-the-enum-module/

    • chevron_right

      Erlang Solutions: Erlang Solutions’ Blog round-up

      news.movim.eu / PlanetJabber • 17 April • 2 minutes

    The tech world doesn’t slow down, and neither do we. From the power of big data in healthcare to keeping you up-to-date about fintech compliance, our latest blog posts explore the important topics shaping today’s digital world.

    Whether you’re leading a business, building software, or just curious about the future of tech, check out what the Erlang Solutions team has been talking about.

    Understanding Big Data in Healthcare


    From smartwatches to digital patient records, healthcare is generating more data than ever before. In our post, Understanding Big Data in Healthcare , we explore how big data is transforming patient care—think predictive health trends, precision medicine, and smarter decision-making.

    We also dig into the challenges of handling sensitive information and how tech like Erlang, Elixir, and SAFE can help keep that data secure. It’s a must-read if you’re curious about how data is shaping the future of healthcare.

    Understanding Digital Wallets


    Digital wallets aren’t just convenient, but they’re quickly becoming essential. In Understanding Digital Wallets , we break down how they work, why they’re gaining momentum, and what they mean for businesses today.

    From improved security and cost savings to global access and smoother customer experiences, this blog gives a clear look at the benefits (and a few of the challenges) of going digital.

    Women in BEAM

    Lorena Mireles takes us through the key takeaways from her Women in BEAM survey, while also sharing her personal journey within the Elixir community.

    From representation gaps to the importance of strong role models, this piece highlights real experiences from developers in the BEAM ecosystem and why inclusion still matters in tech.

    Top 5 IoT Business Security Basics

    Billions of connected devices = billions of entry points for hackers. In Top 5 IoT Business Security Basics , we outline five practical tips to boost your IoT security:

    • Use strong, unique passwords
    • Encrypt your data
    • Run regular security audits
    • Train your team
    • Disable features you don’t use

    Simple, effective, and easy to implement—these tips will help you keep your systems (and customers) safe.

    DORA Compliance: What Fintech Businesses Need to Know


    Digital Operational Resilience Act (DORA) for fintech

    As of January 2025, the Digital Operational Resilience Act (DORA) is live, and if you’re in fintech, you need to be ready. Our post, DORA Compliance: What Fintech Businesses Need to Know , breaks down what DORA is, who it affects, and what steps you need to take to stay compliant (if you haven’t already).

    We explore its five key pillars and how to build a more resilient, disruption-proof business. With over 22,000 companies affected, it’s a timely read for anyone in the financial space.

    To conclude

    Staying ahead in tech doesn’t have to be overwhelming. Our goal? To make it easier for you to understand what’s happening and what it means for your business.

    Got questions? Let’s chat.

    The post Erlang Solutions’ Blog round-up appeared first on Erlang Solutions .

    • chevron_right

      Gajim: Gajim 2.1.1

      news.movim.eu / PlanetJabber • 17 April • 1 minute

    This release brings layout improvements to Gajim’s Start Chat dialog, an improved message search, and includes fixes for some issues with Message Displayed Synchronization. Thank you for all your contributions!

    What’s New

    Gajim 2.1 comes with a new ‘Activity feed’ which displays events around group chat invitations, contact requests, and updates. This will be the central feed for all kinds of activities in the future (e.g. reactions, replies, mentions, message reminders).

    Activity feed in Gajim 2.1

    Activity feed in Gajim 2.1

    Gajim 2.1.1 brings layout improvements for its Start Chat dialog, which now renders more compact and shows more information. This release also improves message search by displaying the last correction of a message and by not showing moderated messages. Last but not least, some issues with XEP-0490: Message Displayed Synchronization have been fixed.

    A note for Windows users: At the time of writing, there are some issues with emoji rendering on Windows. That’s why there is no release of Gajim 2.1 for Windows yet. This issue should soon be resolved and we will post an update once Gajim 2.1 is released on Windows.

    More Changes

    • Group chats now show voice request errors, if they happen
    • Setting your status for multiple accounts though the account sidebar now works properly

    And much more! Have a look at the changelog for a complete list.

    Gajim

    As always, don’t hesitate to contact us at gajim@conference.gajim.org or open an issue on our Gitlab .

    Support Gajim

    Gajim is free software developed by volunteers.
    If you like to support Gajim, please consider making a donation.

    Donate via Liberapay:

    liberapay-donate.svg

    • wifi_tethering open_in_new

      This post is public

      gajim.org /post/2025-04-17-gajim-2.1.1-released/

    • chevron_right

      ProcessOne: ejabberd 25.04

      news.movim.eu / PlanetJabber • 16 April • 2 minutes

    ejabberd 25.04

    Just a few weeks after previous release, ejabberd 25.04 is published with an important security fix, several bug fixes and a new API command.

    Release Highlights:

    If you are upgrading from a previous version, there are no changes in SQL schemas, configuration, API commands or hooks.

    Other contents:

    Below is a detailed breakdown of the improvements and enhancements:

    mod_muc_occupantid: Fix handling multiple occupant-id

    Fixed issue with handling of user provided occupant-id in messages and presences sent to muc room. Server was replacing just first instance of occupant-id with its own version, leaving other ones untouched. That would mean that depending on order in which clients send occupant-id, they could see value provided by sender, and that could be used to spoof as different sender.

    New kick_users API command

    There is a new API command kick_users that disconnects all the client sessions in a given virtual host.

    Acknowledgments

    We would like to thank the contributions to the source code, documentation, and translation provided for this release by:

    And also to all the people contributing in the ejabberd chatroom, issue tracker...

    Improvements in ejabberd Business Edition

    For customers of the ejabberd Business Edition , in addition to all those improvements and bugfixes:

    • Bugfix on max_concurrent_connections for mod_gcm , mod_webhook and mod_webpush

    ChangeLog

    This is a more complete list of changes in this ejabberd release:

    Security fixes

    • mod_muc_occupantid : Fix handling multiple occupant-id

    Commands API

    • kick_users : New command to kick all logged users for a given host

    Bugfixes

    • Fix issue with sql schema auto upgrade when using sqlite database
    • Fix problem with container update, that could ignore previous data stored in mnesia database
    • Revert limit of allowed characters in shared roster group names, that will again allow using symbols like :
    • Binary installers and ejabberd container image: Updated to Erlang/OTP 27.3.2

    Full Changelog

    https://github.com/processone/ejabberd/compare/25.03...25.04

    ejabberd 25.04 download & feedback

    As usual, the release is tagged in the Git source code repository on GitHub .

    The source package and installers are available in ejabberd Downloads page. To check the *.asc signature files, see How to verify ProcessOne downloads integrity .

    For convenience, there are alternative download locations like the ejabberd DEB/RPM Packages Repository and the GitHub Release / Tags .

    The ecs container image is available in docker.io/ejabberd/ecs and ghcr.io/processone/ecs . The alternative ejabberd container image is available in ghcr.io/processone/ejabberd .

    If you consider that you&aposve found a bug, please search or fill a bug report on GitHub Issues .

    • chevron_right

      ProcessOne: Hello from the other side: Matrix LEFT RIGHT ARROW XMPP via ejabberd 25.03

      news.movim.eu / PlanetJabber • 15 April • 2 minutes

    Hello from the other side: Matrix ↔ XMPP via ejabberd 25.03


    With ejabberd 25.03 , the Matrix gateway ( mod_matrix_gw ) now supports not only one-to-one chats, but also joining Matrix rooms via XMPP. That’s right — your favorite XMPP client can now talk to Matrix users or hop into Matrix rooms just like regular MUCs. ✨

    In this guide, we’ll show a quick demo of:

    1. One-to-one chat between an XMPP and a Matrix user.
    2. Joining a Matrix room and chatting from your XMPP client.

    And the best part? You don’t need to install a Matrix client at all. Keep your favorite XMPP client near, welcome to the magic of federation & interoperability.


    🛠 Setup Summary

    We won’t repeat the full configuration steps here — they’re already covered in this earlier blogpost and the 25.03 release note .

    In short, you’ll need:

    • A properly configured ejabberd server with mod_matrix_gw
    • Block outgoing connections to lethe.matrix.org to avoid conflicts with their XMPP instance

    Here’s a minimal config snippet:

    listen:
      -
        port: 8448
        module: ejabberd_http
        tls: true
        request_handlers:
          "/_matrix": mod_matrix_gw
    
    modules:
      mod_matrix_gw:
        key_name: "xxxxxx"
        key: "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
        matrix_id_as_jid: true
    

    Wondering what is the Matrix signing key? Please refer to previous blog post: Matrix gateway setup with ejabberd

    Still not sure if your setup is correct? Try Matrix Federation Tester to check if everything is wired up properly.

    🧪 One-to-One Chat (XMPP ↔ Matrix)

    In your XMPP client (e.g. Psi), add a Matrix contact using this format:
    matrixUser%theirMatrixServer.tld@yourXMPPserver.tld

    In my case, that means:
    adrien-p1%matrix.org@matrix.mickael.m.in-app.io

    On the Matrix side (e.g. Element Web), your contact will get a request. Once accepted, you’re good to chat!

    ✅ Demo: XMPP user (Psi) chats with Matrix user (Element Web)

    🧪 Join a Matrix Room from your XMPP Client

    Alright, let’s join a public Matrix room now.

    From your XMPP client, you can join any Matrix public room using this format:
    #matrixRoom%theirMatrixServer.tld@yourXMPPserver.tld

    We’ll use the following room: #ejabberd-demo:matrix.org , so in my case, that means joining:
    #ejabberd-demo%matrix.org@matrix.mickael.m.in-app.io

    Once connected, you’ll be able to send and receive messages from any other participant, just like in a regular MUC. :)

    ✅ Demo: Join and chat in a Matrix room from XMPP

    🐞 Known Caveats

    There&aposs still a lot of work to do to make it seamless, here&aposs a short list of currently known caveats:

    • Room presence can be overwhelming in large rooms (thousands of Matrix users may appear "online").
    • No E2EE support between Matrix and XMPP — encryption must be disabled for now.
    • If the server restarts, 1-to-1 conversations must be restarted (re-added), as persistence is not implemented yet.
    • Only Matrix room protocol versions 9, 10, and 11 are supported.

    🧵 Wrapping Up

    With this, ejabberd makes another step into being a powerful bridge into the Matrix federation, for both private and public communication.

    It’s a great way to keep using your favorite XMPP tools while staying connected to Matrix communities.

    Got feedback? Drop us a line in the comments or open a PR .

    Happy bridging! 🙌

    • wifi_tethering open_in_new

      This post is public

      www.process-one.net /blog/hello-from-the-other-side-matrix-xmpp-via-ejabberd-25-03/

    • chevron_right

      Dino: Dino 0.5 Release

      news.movim.eu / PlanetJabber • 11 April • 1 minute

    Dino is a secure and open-source messaging application. It uses the XMPP (Jabber) protocol for decentralized communication. We aim to provide an intuitive and enjoyable user interface.

    The 0.5 release improves the user experience around file transfers and includes two completely reworked dialogs.

    Improved file sharing

    image_preview_loading.png

    The way file transfers are currently done in the XMPP ecosystem is limited in functionality and files can sometimes be received out-of-order. Dino now supports a new method for announcing file transfers ( XEP-0447 ), which solves this issue. Additionally, users can now see preview images or other file details before downloading the file. Dino currently only uses the new method for unencrypted file transfers, for example in public channels. Encrypted file transfers will also support the new protocol once Dino supports full-stanza encryption. All file transfers now also display the upload or download progress.

    Screenshots of three dialogs for account settings, encryption and contact details

    Reworked dialogs

    The account and preferences dialogs have been combined into a single, new dialog. This dialog lets you manage accounts and adjust encryption and other settings. It now also includes some new settings like an option for OMEMO encryption by default, which is enabled by default.

    Additionally, the conversation details dialog has been completely redesigned. Both dialogs are now fully compatible with mobile devices.

    Colors and more

    Dino now uses the same fallback avatar colors as other clients ( XEP-0392 ), creating a more consistent experience across applications.

    A new unread line has been added, indicating the point up to which you’ve already read the messages.

    Dino has also switched from CMake to Meson, which allows for an easier development process.

    Alentejo

    We named this Dino release “Alentejo” after a region in Portugal.

    monsaraz.jpg

    Alentejo is a region in southern Portugal that is known for its wide plains that are dotted with wineyards and cork trees. The region has a Mediterranean climate with summers regularly reaching temperatures above 40 degrees.

    Currently, about 3.6% of all deaths in the region are caused by heat. Heatwaves, in particular, pose a serious health risk and are expected to become more frequent and severe due to global warming [ 1 ]. If CO₂ emissions keep increasing, heat-related deaths could make up 15.8% of all deaths in the region by 2100. However, if action is taken to combat climate change, this number could be limited to 6.6% [ 2 ].

    • wifi_tethering open_in_new

      This post is public

      dino.im /blog/2025/04/dino-0.5-release/