<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>Partech Systems — Journal</title>
    <link>https://partechsystems.com/blog/</link>
    <description>An independent studio making thoughtful games, useful apps, desktop tools and original typefaces, built on decades of software and hardware experience.</description>
    <language>en-GB</language>
    <lastBuildDate>Tue, 17 Nov 2026 16:25:03 +0000</lastBuildDate>
    <atom:link href="https://partechsystems.com/feed.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>A box closed, and a beam followed</title>
      <link>https://partechsystems.com/blog/a-box-closed-and-a-beam-followed/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/a-box-closed-and-a-beam-followed/</guid>
      <pubDate>Mon, 09 Nov 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bhaskar Paratey</dc:creator>
      <category>Product stories</category>
      <description>Two of our newest games began long before they were games, with a packing problem at university and childhood afternoons spent making rainbows.</description>
      <content:encoded><![CDATA[&lt;p&gt;Some games start with a mechanic or a gap in a shop. These two started much earlier, with things I could never quite leave alone.&lt;/p&gt;
&lt;h2 id=&#34;bento-closing-the-box&#34;&gt;Bento: closing the box&lt;/h2&gt;
&lt;p&gt;At university, fitting things into a box was one of the first problems that showed me how hard a simple question can be. You have some shapes and a box. Can they all go in, with nothing left over?&lt;/p&gt;
&lt;p&gt;It sounds like something you would do with a suitcase. Computer scientists call it a packing problem, and it is NP-hard: as the pieces add up, no known method can find the answer quickly every time. Fitting shaped pieces into a box exactly is one of the puzzles that has been &lt;a href=&#34;https://erikdemaine.org/papers/Jigsaw_GC/&#34;&gt;proved to be that hard&lt;/a&gt;.&lt;/p&gt;
&lt;p&gt;It stayed with me. Over the years I kept coming back to smaller versions of it: a knapsack problem here, a box-packing routine there, each one a corner of the same question.&lt;/p&gt;
&lt;p&gt;&lt;a href=&#34;/games/bento/&#34;&gt;Bento&lt;/a&gt; is the first time I have built the whole thing as something you can hold. You drag rice, salmon, tamago and the rest into a lacquer lunchbox, tap a piece to turn it, and fill every compartment with nothing left over. The hard problem is still there underneath. It has just become forty-five boxes, and one more each day, every one of them checked so that it can be packed.&lt;/p&gt;
&lt;p&gt;&lt;img alt=&#34;A Bento lunchbox half packed, with onigiri, salmon and broccoli in their compartments and three pieces still on the mat.&#34; src=&#34;/static/images/games/bento/packing.webp&#34; /&gt;&lt;/p&gt;
&lt;p&gt;Making it felt like finishing something. Not solving the problem, because nobody has done that, but answering it the way I always wanted to: with a puzzle a person can enjoy, rather than an algorithm a machine has to grind through.&lt;/p&gt;
&lt;h2 id=&#34;lamplighter-following-the-beam&#34;&gt;Lamplighter: following the beam&lt;/h2&gt;
&lt;p&gt;As a child, I was fascinated by mirrors and prisms. I spent many afternoons catching the light, making rainbow patterns, moving things a little to see what changed, and trying to understand what the light was actually doing.&lt;/p&gt;
&lt;p&gt;The fascination grew up with me. Later it was interference patterns and the two-slit experiment: light behaving in ways that still feel slightly impossible, however many times they are explained.&lt;/p&gt;
&lt;p&gt;&lt;a href=&#34;/games/lamplighter/&#34;&gt;Lamplighter&lt;/a&gt; is where that has finally gone. You turn mirrors to send a beam down a night street until every lamp is lit. Further along the road, prisms split the beam in two, and ruby and jade glass colour it, so a coloured lamp only lights in its own colour.&lt;/p&gt;
&lt;p&gt;&lt;img alt=&#34;A Lamplighter puzzle with a square of ruby glass in the beam’s path and a ruby lamp waiting to be lit.&#34; src=&#34;/static/images/games/lamplighter/glass.webp&#34; /&gt;&lt;/p&gt;
&lt;p&gt;It is a small game, but it is made from those same afternoons: point the light, watch where it goes, change one thing and try again.&lt;/p&gt;
&lt;h2 id=&#34;why-these-two&#34;&gt;Why these two&lt;/h2&gt;
&lt;p&gt;A small studio could make almost anything small. It helps when the thing is something you have wanted to make for years. You notice which details matter, because you have been turning them over for a long time.&lt;/p&gt;
&lt;p&gt;Bento is on Google Play and in your browser. &lt;a href=&#34;/games/bento/&#34;&gt;Pack a box&lt;/a&gt;, or &lt;a href=&#34;/games/lamplighter/&#34;&gt;light a street&lt;/a&gt;.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>A price is part of the product</title>
      <link>https://partechsystems.com/blog/a-price-is-part-of-the-product/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/a-price-is-part-of-the-product/</guid>
      <pubDate>Fri, 06 Nov 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Kamal Paratey</dc:creator>
      <category>Stewardship</category>
      <description>Free tools, paid software and the practical value of making the terms clear before somebody starts.</description>
      <content:encoded><![CDATA[&lt;p&gt;A price makes a promise before somebody has used the product.&lt;/p&gt;
&lt;p&gt;It tells them something about what the relationship will be: a purchase, a recurring service, a free tool, or a game supported in another way. If the terms are hard to understand, the uncertainty becomes part of the experience.&lt;/p&gt;
&lt;p&gt;Working in operations and finance makes me attentive to both sides of that relationship. The person needs to know what they are agreeing to. The company needs to understand the work and cost that the agreement will leave it with.&lt;/p&gt;
&lt;h2 id=&#34;explain-the-arrangement&#34;&gt;Explain the arrangement&lt;/h2&gt;
&lt;p&gt;A product page should make the practical terms easy to find. What can somebody try? What costs money? Is the payment one-off or recurring? Are there ads? Is there a limit that matters to the task they have in mind?&lt;/p&gt;
&lt;p&gt;These are useful questions even for a small application. “Free” should have enough detail behind it that a person is not surprised at the point where they want to finish their work. A purchase should describe what is included without relying on an optimistic interpretation of a headline.&lt;/p&gt;
&lt;p&gt;Different products can reasonably have different arrangements. &lt;a href=&#34;/apps/roomy/&#34;&gt;Roomy&lt;/a&gt; is offered as a free tool. &lt;a href=&#34;/apps/stetpdf/&#34;&gt;StetPDF&lt;/a&gt;, still in development, is being presented around a one-off purchase rather than a subscription. The individual product page is where the current terms belong.&lt;/p&gt;
&lt;h2 id=&#34;time-belongs-in-the-calculation&#34;&gt;Time belongs in the calculation&lt;/h2&gt;
&lt;p&gt;The price on a page is visible. Other costs arrive as effort. Creating an account, understanding a licence, getting a file out, asking for support: each can take time from the person who wanted to do a straightforward job.&lt;/p&gt;
&lt;p&gt;That does not mean every step can be removed. It means a step ought to have a reason proportionate to the product. A small tool should be able to explain why it needs what it asks for.&lt;/p&gt;
&lt;p&gt;The same attention helps the company. Complicated terms generate questions. Unclear limitations turn into disappointment. An arrangement that is easy to explain can be easier to administer and easier to honour.&lt;/p&gt;
&lt;h2 id=&#34;leave-room-to-keep-the-promise&#34;&gt;Leave room to keep the promise&lt;/h2&gt;
&lt;p&gt;There is no universal price that makes a product responsible. The relevant question is whether the arrangement supports the work behind it and makes sense for the people using it.&lt;/p&gt;
&lt;p&gt;For an independent studio, this is closely connected to scope. A modest product with clear boundaries can be something we are able to maintain over time. Expanding what is included creates obligations as surely as expanding the code.&lt;/p&gt;
&lt;p&gt;The terms are therefore a design decision as well as an operating decision. We should be able to describe them plainly, keep them consistent with the product and explain a change before it becomes a surprise.&lt;/p&gt;
&lt;p&gt;A good relationship with a customer has room for ordinary questions. What does it do? What will it cost me? What can I expect after that? Those deserve ordinary, understandable answers.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>The promise in a small sentence</title>
      <link>https://partechsystems.com/blog/a-product-description-is-a-promise/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/a-product-description-is-a-promise/</guid>
      <pubDate>Thu, 05 Nov 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Dimple Paratey</dc:creator>
      <category>Product language</category>
      <description>The words around a product should help you decide whether it belongs in your day.</description>
      <content:encoded><![CDATA[&lt;p&gt;Lens could be a camera app. Kept could be a journal. Plain could be a text editor.&lt;/p&gt;
&lt;p&gt;A name can give a product character, but it cannot carry the whole explanation. The small sentence beside it has a practical job: help somebody understand what they would be opening.&lt;/p&gt;
&lt;p&gt;&lt;a href=&#34;/games/lens/&#34;&gt;Lens&lt;/a&gt; is a hidden-object game set in a linocut market. &lt;a href=&#34;/apps/kept/&#34;&gt;Kept&lt;/a&gt;, which is still in Labs, is being built to track receipt return dates and warranties. &lt;a href=&#34;/apps/plain/&#34;&gt;Plain&lt;/a&gt;, also in development, is intended to explain what an official letter asks you to do.&lt;/p&gt;
&lt;p&gt;Once those sentences are there, the names have something to attach to. Before that, they are invitations to guess.&lt;/p&gt;
&lt;h2 id=&#34;start-with-what-the-person-can-do&#34;&gt;Start with what the person can do&lt;/h2&gt;
&lt;p&gt;It is easy to describe a product from the inside. We know the technique, the architecture and the effort behind it. We want the description to reflect that work.&lt;/p&gt;
&lt;p&gt;The person arriving at the page usually has a different question. Can I read this document on my phone? Is this a game I can play here? Will this help me find the files taking up space on my computer?&lt;/p&gt;
&lt;p&gt;A useful description meets them there. The engineering can have its own explanation further down the page, for someone who wants to understand it. It does not have to stand in the doorway.&lt;/p&gt;
&lt;p&gt;This matters especially now that AI can become a label for almost any sort of software. Saying that a product uses AI explains very little about the job it will do. It also leaves the reader to guess where their information goes and how much trust to place in the result.&lt;/p&gt;
&lt;h2 id=&#34;small-claims-are-easier-to-keep&#34;&gt;Small claims are easier to keep&lt;/h2&gt;
&lt;p&gt;Every adjective adds to the promise. “Instant” is a claim about time. “Private” is a claim about information. “Free” is a claim about what happens when someone starts using the product.&lt;/p&gt;
&lt;p&gt;Those words need an explanation behind them. If an application works on the device, say which work happens there. If it contains ads, say when they appear. If a product is still being built, say that where a download button would otherwise be.&lt;/p&gt;
&lt;p&gt;There is no benefit in attracting somebody through an expectation the software will immediately disappoint.&lt;/p&gt;
&lt;h2 id=&#34;a-shared-voice-with-room-for-character&#34;&gt;A shared voice, with room for character&lt;/h2&gt;
&lt;p&gt;Our games can sound playful. A prayer app calls for a more respectful register. A disk-cleaning tool needs to be especially clear about what an action will affect.&lt;/p&gt;
&lt;p&gt;They can all sound as if they come from Partech. We can use ordinary words, name the action and keep the explanation close to the control. We can prefer a direct sentence to a claim about transformation. We can make the price and the product’s stage easy to find.&lt;/p&gt;
&lt;p&gt;That common language is a form of hospitality. It lets a person decide whether to stay, with enough information and without a sales pitch getting in the way.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>A game you can settle into</title>
      <link>https://partechsystems.com/blog/sunroom-made-to-be-read/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/sunroom-made-to-be-read/</guid>
      <pubDate>Tue, 03 Nov 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Dimple Paratey</dc:creator>
      <category>Product stories</category>
      <description>Sunroom brings familiar games together through clear type, comfortable space and an unhurried invitation to play.</description>
      <content:encoded><![CDATA[&lt;p&gt;You can tell a lot about a game by where it expects you to look.&lt;/p&gt;
&lt;p&gt;Sometimes the board is competing with a timer, a special offer, a score multiplier and a row of small symbols whose meaning you are meant to remember. You have opened a familiar game and been given another interface to learn.&lt;/p&gt;
&lt;p&gt;With Sunroom, the starting point is the familiar part. Solitaire should look like solitaire. A sudoku should give the numbers room. Word Search should make the letters easy to follow.&lt;/p&gt;
&lt;p&gt;The family now includes &lt;a href=&#34;/games/sunroom/&#34;&gt;Solitaire&lt;/a&gt;, &lt;a href=&#34;/games/sunroom-sudoku/&#34;&gt;Sudoku&lt;/a&gt;, &lt;a href=&#34;/games/sunroom-freecell/&#34;&gt;FreeCell&lt;/a&gt;, &lt;a href=&#34;/games/sunroom-words/&#34;&gt;Word Search&lt;/a&gt; and &lt;a href=&#34;/games/sunroom-mahjong/&#34;&gt;Mahjong&lt;/a&gt;. Their resemblance comes from a shared intention: make the game comfortable to read and easy to find your way around.&lt;/p&gt;
&lt;h2 id=&#34;clear-without-being-childish&#34;&gt;Clear without being childish&lt;/h2&gt;
&lt;p&gt;There is a particular design mistake we want to avoid. Increasing the size of a button can turn into treating the person pressing it as if they need everything simplified for them.&lt;/p&gt;
&lt;p&gt;Those are different decisions. A person can know a game very well and still prefer larger numbers. They can enjoy a difficult puzzle while wanting the controls to be straightforward. Clarity should leave the interesting part intact.&lt;/p&gt;
&lt;p&gt;In Sunroom Solitaire, large corner numbers help you read the cards. In Sudoku, hints explain a reason. In Word Search, selecting the first and last letter gives the gesture a clear shape. Each game has its own details; the family should make those details easier to approach.&lt;/p&gt;
&lt;p&gt;&lt;img alt=&#34;Sunroom Solitaire on its green Garden table, with large card numbers.&#34; src=&#34;/static/images/games/sunroom/garden.webp&#34; /&gt;&lt;/p&gt;
&lt;h2 id=&#34;a-family-resemblance-that-earns-its-name&#34;&gt;A family resemblance that earns its name&lt;/h2&gt;
&lt;p&gt;Sunroom is a useful name because it describes a shared feeling without trying to describe every mechanic. We can give the games a related visual language and still let a card game and a number puzzle be different things.&lt;/p&gt;
&lt;p&gt;The important resemblance is in the experience. You should be able to find the board, understand the controls and make your next move. If you need a hint, the explanation should help you return to the game. Settings should be where you expect them.&lt;/p&gt;
&lt;p&gt;That is also why we keep the familiar game name in each title. People looking for FreeCell should not have to decipher a marketing name before they know what is inside.&lt;/p&gt;
&lt;h2 id=&#34;room-in-the-day&#34;&gt;Room in the day&lt;/h2&gt;
&lt;p&gt;A game does not have to be a major event to deserve good design. There are the minutes between two other things, the evening when you would rather play a little than scroll a lot, the puzzle you leave and come back to.&lt;/p&gt;
&lt;p&gt;Sunroom is made for that sort of place in a day. We want the board to be inviting, the lettering to be generous, and the person playing to remain in charge of the pace.&lt;/p&gt;
&lt;p&gt;&lt;a href=&#34;/collections/sunroom/&#34;&gt;Find a Sunroom game to try.&lt;/a&gt;&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>The work that keeps a product alive</title>
      <link>https://partechsystems.com/blog/the-work-that-keeps-a-product-alive/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/the-work-that-keeps-a-product-alive/</guid>
      <pubDate>Fri, 21 Aug 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Kamal Paratey</dc:creator>
      <category>Stewardship</category>
      <description>Release day creates an obligation. Operations and finance help us understand whether we can keep meeting it.</description>
      <content:encoded><![CDATA[&lt;p&gt;Release day adds a new obligation.&lt;/p&gt;
&lt;p&gt;There is now a product somebody may rely on, a support address they may write to and a set of expectations attached to its name. The work has acquired a future.&lt;/p&gt;
&lt;p&gt;I have been a director of Partech since 1997, working in operations and finance. Those functions follow a product beyond the moment when its first version is ready. They ask what it takes to keep the thing available and whether the company can continue to provide it with care.&lt;/p&gt;
&lt;h2 id=&#34;count-the-work-around-the-feature&#34;&gt;Count the work around the feature&lt;/h2&gt;
&lt;p&gt;An implementation is only one part of a product’s cost. There is time spent explaining it, testing changes, preparing releases, answering questions and keeping up with the platforms it runs on. There may be services to pay for and tools to replace. There are also small administrative tasks that become important when they are missed.&lt;/p&gt;
&lt;p&gt;A feature can appear inexpensive while adding a lasting complication to several of those jobs. A setting needs documentation. A new platform needs testing. Another way to pay creates another set of questions to answer.&lt;/p&gt;
&lt;p&gt;It helps to see these as part of the decision while there is still a choice about what to build.&lt;/p&gt;
&lt;h2 id=&#34;a-catalogue-is-a-set-of-commitments&#34;&gt;A catalogue is a set of commitments&lt;/h2&gt;
&lt;p&gt;An independent studio can make a wide range of things. That is one of its pleasures. Each additional product also asks for a share of attention.&lt;/p&gt;
&lt;p&gt;A small app may need very little intervention for a while, then require work because the world around it has changed. A store changes a requirement. An operating system behaves differently. Somebody finds a problem that had not appeared in testing.&lt;/p&gt;
&lt;p&gt;The catalogue therefore needs more than enthusiasm for the next release. It needs a practical view of the products already in people’s hands. Which ones need attention? Which questions keep returning? Is the explanation on the website still accurate?&lt;/p&gt;
&lt;p&gt;None of those questions looks particularly striking in a launch announcement. Their answers are part of whether a person can keep trusting the product.&lt;/p&gt;
&lt;h2 id=&#34;keeping-independence-practical&#34;&gt;Keeping independence practical&lt;/h2&gt;
&lt;p&gt;Finance can make this conversation more useful by making the obligations visible. Time, recurring costs and the capacity to respond all belong in the picture. So does the value of retaining a product that serves a clear purpose.&lt;/p&gt;
&lt;p&gt;The decision is not always to expand. Sometimes it is to improve an existing application before starting another. Sometimes it is to keep the scope small enough that the product remains straightforward to support. Sometimes a service needs to be retired, with a clear explanation and a reasonable route for the people using it.&lt;/p&gt;
&lt;p&gt;These are ordinary operating decisions. Taken together, they shape the sort of company we can continue to be.&lt;/p&gt;
&lt;p&gt;Partech’s work has changed over the years, from solutions built for clients to a growing collection of our own software. The responsibility has stayed recognisable. If our name is on a product, we should understand what it takes to keep looking after it.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>The person after the handover</title>
      <link>https://partechsystems.com/blog/the-work-after-the-handover/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/the-work-after-the-handover/</guid>
      <pubDate>Fri, 12 Jun 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bhavna Ate</dc:creator>
      <category>People</category>
      <description>A product continues into somebody else’s day. Good handovers help that person understand what has changed and what to do next.</description>
      <content:encoded><![CDATA[&lt;p&gt;Handover sounds like an ending when you are looking at a project plan.&lt;/p&gt;
&lt;p&gt;For the person receiving the work, it can be the point at which the important questions begin. What has changed? What am I responsible for now? Where do I go when something does not behave as expected?&lt;/p&gt;
&lt;p&gt;Partech spent more than twenty years building solutions for other companies. That kind of work puts the handover in plain sight. A system has to become part of somebody’s working day. It has to be explained, supported and understood by people who were not in every design conversation.&lt;/p&gt;
&lt;p&gt;Our own products cross the same boundary, often without a conversation at all. Someone sees a page, decides to try an app and meets the interface on their own.&lt;/p&gt;
&lt;h2 id=&#34;the-knowledge-that-did-not-travel&#34;&gt;The knowledge that did not travel&lt;/h2&gt;
&lt;p&gt;A team knows why a permission is requested. The person using the product sees a question about access to their files. A developer understands why an operation takes time. The person waiting sees a screen that may have stopped responding.&lt;/p&gt;
&lt;p&gt;The missing information can be quite small. It still matters. A clear explanation beside the permission, an honest indication of progress, or a message about where a result has been saved can make the next step understandable.&lt;/p&gt;
&lt;p&gt;We can ask a useful question while reviewing a product: what does the person need to know here that the team has forgotten it knows?&lt;/p&gt;
&lt;h2 id=&#34;help-without-taking-over&#34;&gt;Help without taking over&lt;/h2&gt;
&lt;p&gt;Offering help should leave somebody with more confidence in their own next action. A hint in a game can explain a move. A reading tool can make text easier to see or hear. An application handling a document can keep the original available for checking.&lt;/p&gt;
&lt;p&gt;The help is specific to the task. It should not require the person to accept a broader judgement about their ability.&lt;/p&gt;
&lt;p&gt;That is one reason I value familiar language. “Text size” tells you what a setting changes. “Read aloud” tells you what will happen when you press it. An unfamiliar name adds a small piece of work at exactly the point where the interface is meant to remove one.&lt;/p&gt;
&lt;h2 id=&#34;make-the-route-back-visible&#34;&gt;Make the route back visible&lt;/h2&gt;
&lt;p&gt;People are interrupted. They change their minds. They try a control to discover what it does. An interface can allow for those ordinary behaviours through understandable navigation, clear consequences and recovery where it is possible.&lt;/p&gt;
&lt;p&gt;Support is another route back. The person writing for help has already spent time on the problem. They should not have to explain it repeatedly or discover that the advertised contact leads nowhere.&lt;/p&gt;
&lt;p&gt;A handover is successful when the person receiving the work can continue with it. That is a useful standard for a large client system, and an equally useful one for a small app opened for the first time.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>A small team with a long memory</title>
      <link>https://partechsystems.com/blog/small-team-long-memory/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/small-team-long-memory/</guid>
      <pubDate>Thu, 23 Apr 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bhavna Ate</dc:creator>
      <category>People</category>
      <description>A century of combined leadership experience is useful when it helps a team notice more, listen better and make room for new judgement.</description>
      <content:encoded><![CDATA[&lt;p&gt;A small team can hold a great deal of working memory.&lt;/p&gt;
&lt;p&gt;There are ways of approaching a problem that have been tested in different organisations. There is an instinct for the decision that looks minor now and becomes expensive later. There are also assumptions that once served us well and need to be examined again.&lt;/p&gt;
&lt;p&gt;Across Partech’s leadership there is more than a hundred years of combined experience. I find that useful as context for the work. It tells you that the team brings several long working lives to the table. It does not settle the question of whether a particular decision is right.&lt;/p&gt;
&lt;h2 id=&#34;make-experience-available&#34;&gt;Make experience available&lt;/h2&gt;
&lt;p&gt;Experience helps a team when it can be passed around. A concern needs to be explained well enough that someone else can examine it. A decision needs a reason that survives the person who made it leaving the room.&lt;/p&gt;
&lt;p&gt;“I have seen this before” can be the start of a useful conversation. The next part is to describe what seems familiar and what might be different this time.&lt;/p&gt;
&lt;p&gt;That distinction matters in a company with a long history. Work for clients, work involving physical equipment, and work on independent apps can teach related lessons. They also take place under different constraints. The resemblance deserves attention; so do the differences.&lt;/p&gt;
&lt;h2 id=&#34;leave-room-for-a-fresh-pair-of-eyes&#34;&gt;Leave room for a fresh pair of eyes&lt;/h2&gt;
&lt;p&gt;A product team can become fluent in its own interface. The labels stop looking ambiguous. A sequence of screens feels natural because everybody knows what is coming next.&lt;/p&gt;
&lt;p&gt;Someone new to the product does not have that shared memory. Their question may expose a problem the team has learnt to step around. It is worth preserving the question before explaining the design.&lt;/p&gt;
&lt;p&gt;The same is true inside a team. A quiet contribution, a written note after a meeting or a question from somebody outside the speciality can change how a problem is understood. Speed and confidence are only two ways that an idea arrives.&lt;/p&gt;
&lt;h2 id=&#34;keep-the-organisation-small-enough-to-understand&#34;&gt;Keep the organisation small enough to understand&lt;/h2&gt;
&lt;p&gt;One attraction of a small independent studio is that the distance between a decision and its consequences can stay short. The people writing the description can speak to the people building the feature. A support question can reach somebody able to change the product.&lt;/p&gt;
&lt;p&gt;That closeness takes work to preserve. Unclear responsibilities and decisions left in private conversations can make even a small team difficult to navigate. A brief written decision, an explicit owner and a useful explanation can save considerable effort later.&lt;/p&gt;
&lt;p&gt;For me, this is where the long memory matters most. It can help us recognise the habits worth keeping while leaving enough room to learn something new.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>Software has to live somewhere</title>
      <link>https://partechsystems.com/blog/software-has-to-live-somewhere/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/software-has-to-live-somewhere/</guid>
      <pubDate>Thu, 05 Mar 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bhaskar Paratey</dc:creator>
      <category>Engineering</category>
      <description>What building hardware and software together teaches you about the person at the other end.</description>
      <content:encoded><![CDATA[&lt;p&gt;A cable is a useful corrective to a software engineer’s optimism.&lt;/p&gt;
&lt;p&gt;It can be plugged into the wrong socket. It can be too short. It can work perfectly on a desk and become a nuisance once the equipment is installed where somebody actually needs it.&lt;/p&gt;
&lt;p&gt;Hardware makes context difficult to ignore. There is light on the screen, dust around the connector, somebody wearing gloves, somebody reaching from the wrong side. The design has to survive those conditions as well as the ones in the specification.&lt;/p&gt;
&lt;p&gt;Building software and hardware solutions for clients was part of Partech’s work for more than twenty years. The equipment and applications varied. The underlying obligation was steady: the parts had to work together in a place where people had other things to do.&lt;/p&gt;
&lt;h2 id=&#34;beyond-the-successful-test&#34;&gt;Beyond the successful test&lt;/h2&gt;
&lt;p&gt;It is possible to test a component carefully and still misunderstand the system it belongs to. A device may produce the right reading while its display is hard to see. A program may save a record correctly while giving the operator no confidence that the save happened.&lt;/p&gt;
&lt;p&gt;The gap is often at a boundary. Between the physical action and the software’s response. Between what the engineer meant and what the person understood. Between a failure and the explanation of how to recover.&lt;/p&gt;
&lt;p&gt;That is why I like asking what happens immediately before and immediately after a feature is used. Where did the information come from? What does the person need to do with the result? What happens if they are interrupted halfway through?&lt;/p&gt;
&lt;p&gt;The answers bring the work back into view.&lt;/p&gt;
&lt;h2 id=&#34;a-phone-is-a-place-too&#34;&gt;A phone is a place, too&lt;/h2&gt;
&lt;p&gt;Our own apps live in physical circumstances just as surely as a piece of equipment does. A phone can be used in poor light, with one hand, at the end of a tiring day. A document can have small print. The person opening a game may be looking for a few undemanding minutes.&lt;/p&gt;
&lt;p&gt;Larger corner numbers can make a card easier to distinguish. Reflowing a page can make a document fit a smaller screen. Those are decisions about the circumstances in which the task happens, and they can be as important as the capability underneath.&lt;/p&gt;
&lt;p&gt;Neither example needs an elaborate theory. It needs attention to the part of the experience that is easy to overlook while building it on a large monitor.&lt;/p&gt;
&lt;h2 id=&#34;leave-a-useful-way-back&#34;&gt;Leave a useful way back&lt;/h2&gt;
&lt;p&gt;Physical systems also make recovery very concrete. If something stops working, somebody has to notice, understand enough of the problem to act, and get back to useful work.&lt;/p&gt;
&lt;p&gt;The same standard belongs in an app. An error message should tell you what could not be done. An undo should reverse the action you think it reverses. Saving should give you a clear result. A temporary interruption should not become an unexplained loss of work.&lt;/p&gt;
&lt;p&gt;These are engineering decisions with a human shape. We learnt to care about them in client systems. They remain worth caring about when the entire product fits in somebody’s pocket.&lt;/p&gt;]]></content:encoded>
    </item>
    <item>
      <title>Building to our own brief</title>
      <link>https://partechsystems.com/blog/from-client-work-to-our-own-products/</link>
      <guid isPermaLink="true">https://partechsystems.com/blog/from-client-work-to-our-own-products/</guid>
      <pubDate>Fri, 16 Jan 2026 00:00:00 +0000</pubDate>
      <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Bhaskar Paratey</dc:creator>
      <category>Our story</category>
      <description>The client years, the product years, and the engineering habits that connect them.</description>
      <content:encoded><![CDATA[&lt;p&gt;For a long time, our work began with another company’s problem.&lt;/p&gt;
&lt;p&gt;There would be a process that needed to work differently, a piece of equipment that needed software, or an application that had to fit into an organisation already full of applications. The brief belonged to somebody else. Our job was to understand it well enough to build something they could use.&lt;/p&gt;
&lt;p&gt;Partech’s roots go back to 1997. Bhaskar and Kamal Paratey have been directors since those early years. Over more than two decades, the business built software and hardware solutions for other companies. Over the last decade, making our own software for the wider world has become another substantial part of the work.&lt;/p&gt;
&lt;p&gt;Those periods overlap. There was no single morning when we stopped being one kind of company and became another. The change was in who got to ask the first question.&lt;/p&gt;
&lt;h2 id=&#34;learning-to-listen-to-a-brief&#34;&gt;Learning to listen to a brief&lt;/h2&gt;
&lt;p&gt;Client work teaches you to distinguish a request from the problem behind it. Someone asks for a report; what they need is to decide what to do next. Someone asks for a new screen; the difficulty may be the three screens they already have to use first.&lt;/p&gt;
&lt;p&gt;It also teaches respect for the things around the software. The equipment it runs on. The person who supports it. The work that happens before a record reaches the system and after it leaves. A technically sound application can still make somebody’s day harder.&lt;/p&gt;
&lt;p&gt;We carry that understanding into products that look much smaller. A medicine label, a receipt and a game of solitaire have very different consequences, but each deserves to be understood on its own terms. The useful starting point is a person trying to do something.&lt;/p&gt;
&lt;h2 id=&#34;a-different-kind-of-brief&#34;&gt;A different kind of brief&lt;/h2&gt;
&lt;p&gt;With your own product, there is nobody across the table to settle the scope for you. You choose the problem, the audience and the point at which you are willing to put your name on it.&lt;/p&gt;
&lt;p&gt;You also inherit the work after release. A download is the beginning of a relationship. The application still needs to open after an operating-system update. A setting still needs to make sense six months later. A support address needs to lead somewhere.&lt;/p&gt;
&lt;p&gt;That changes what a good idea looks like. A feature can be easy to build and expensive to explain. A small product can be worth keeping because it does one job particularly well. Choosing what to leave out becomes part of looking after what remains.&lt;/p&gt;
&lt;p&gt;A familiar game with a legible board and a document that can be read comfortably on a small screen are useful expressions of that thinking. So is the type in which an instruction is set. Different products can share a practical interest in how people read and use things.&lt;/p&gt;
&lt;h2 id=&#34;the-work-we-want-our-name-on&#34;&gt;The work we want our name on&lt;/h2&gt;
&lt;p&gt;The tools available to a small team have changed enormously. We can make and try things faster. That makes judgement more useful, because producing an implementation is no longer the only difficult part of making a product.&lt;/p&gt;
&lt;p&gt;We still have to decide whether the thing deserves somebody’s attention, whether it behaves as its description suggests, and whether we can keep looking after it.&lt;/p&gt;
&lt;p&gt;This journal is where we will write about those decisions. Some pieces look back at the experience that shaped Partech. Others look closely at something we are making now. The connection is the work itself: understanding a problem, making a useful thing, and staying responsible for it.&lt;/p&gt;]]></content:encoded>
    </item>
  </channel>
</rss>