<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.10.0">Jekyll</generator><link href="https://samueleamato.xyz/feed.xml" rel="self" type="application/atom+xml" /><link href="https://samueleamato.xyz/" rel="alternate" type="text/html" /><updated>2026-07-15T22:08:26+00:00</updated><id>https://samueleamato.xyz/feed.xml</id><title type="html">Samuele Amato</title><subtitle>Low-level dev building tools in C, Rust, and more.</subtitle><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><entry xml:lang="it"><title type="html">Non hai bisogno di Nextcloud</title><link href="https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud-it" rel="alternate" type="text/html" title="Non hai bisogno di Nextcloud" /><published>2026-05-03T00:00:00+00:00</published><updated>2026-05-03T00:00:00+00:00</updated><id>https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud-it</id><content type="html" xml:base="https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud-it"><![CDATA[<p>Nextcloud è una suite open source self-hosted che permette di creare un cloud privato per storage, sincronizzazione file e molto altro.</p>

<p>Per anni l’ho utilizzato sul mio server di casa principalmente per due motivi:</p>

<ul>
  <li>Sincronizzare i file tra tutti i miei dispositivi, mantenendo una directory sempre aggiornata.</li>
  <li>Avere una copia centralizzata dei file sul server, utile anche come backup in caso di problemi alle macchine locali.</li>
</ul>

<p>In pratica, il server manteneva sempre la versione più aggiornata dei file e la distribuiva agli altri dispositivi.</p>

<h2 id="il-problema">Il problema</h2>

<p>Col tempo mi sono reso conto che, per questo specifico utilizzo, Nextcloud è <strong>overkill</strong>.</p>

<p>Oltre alla sincronizzazione file, include moltissime funzionalità (calendar, note, condivisioni, web UI, ecc.) che nel mio caso non utilizzo. Questo si traduce in:</p>

<ul>
  <li>maggiore complessità</li>
  <li>più risorse consumate</li>
  <li>più manutenzione</li>
</ul>

<h2 id="la-soluzione">La soluzione</h2>

<p>Cercando un’alternativa più semplice, ho scoperto Syncthing.</p>

<p>Syncthing è un software open source che sincronizza file in modo <strong>peer-to-peer</strong>, senza bisogno di un server centrale o servizi terzi.</p>

<p>Ogni dispositivo comunica direttamente con gli altri e mantiene i file aggiornati in automatico.</p>

<p>Alcuni vantaggi:</p>

<ul>
  <li>configurazione semplice</li>
  <li>niente dipendenze da cloud esterni</li>
  <li>sincronizzazione diretta tra dispositivi</li>
  <li>gestione automatica della connessione (anche senza IP pubblico)</li>
</ul>

<p>Il setup è rapido e ben documentato nella <a href="https://docs.syncthing.net/intro/getting-started.html">guida ufficiale</a>.</p>

<h2 id="serve-ancora-un-server">Serve ancora un server?</h2>

<p>Sì, se vuoi replicare esattamente il comportamento che avevi con Nextcloud.</p>

<p>Syncthing è P2P: questo significa che i file vengono sincronizzati solo tra dispositivi online nello stesso momento.</p>

<p>Esempio:</p>

<ul>
  <li>Modifichi un file sul PC 1</li>
  <li>Il PC 2 è spento</li>
  <li>Se spegni il PC 1 prima che il PC 2 si accenda, la modifica non viene trasferita</li>
</ul>

<p>Per risolvere questo problema, è sufficiente avere un terzo dispositivo <strong>sempre acceso</strong> (ad esempio un server):</p>

<ul>
  <li>riceve tutte le modifiche</li>
  <li>le conserva</li>
  <li>le distribuisce ai dispositivi quando si connettono</li>
</ul>

<p>In questo modo ottieni:</p>

<ul>
  <li>sincronizzazione continua</li>
  <li>copia centralizzata dei file</li>
</ul>

<h2 id="limiti-di-syncthing">Limiti di Syncthing</h2>

<p>Syncthing non è una sostituzione completa di Nextcloud.</p>

<p>Mancano alcune funzionalità:</p>

<ul>
  <li>interfaccia web avanzata per la gestione dei file</li>
  <li>condivisione tramite link pubblici</li>
  <li>integrazione con app (calendar, note, ecc.)</li>
</ul>

<p>Se ti servono queste cose, Nextcloud resta una scelta valida.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Nextcloud è una suite open source self-hosted che permette di creare un cloud privato per storage, sincronizzazione file e molto altro.]]></summary></entry><entry xml:lang="en"><title type="html">When is Nextcloud Overkill?</title><link href="https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud" rel="alternate" type="text/html" title="When is Nextcloud Overkill?" /><published>2026-05-03T00:00:00+00:00</published><updated>2026-05-03T00:00:00+00:00</updated><id>https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud</id><content type="html" xml:base="https://samueleamato.xyz/2026/05/03/you-dont-need-nextcloud"><![CDATA[<p>Nextcloud is an open-source self-hosted suite that lets you build a private cloud for storage, file synchronization, and much more.</p>

<p>For years, I used it on my home server mainly for two reasons:</p>

<ul>
  <li>Synchronizing files across all my devices, keeping a directory always up to date.</li>
  <li>Having a centralized copy of my files on the server, useful as a backup in case something goes wrong with my local machines.</li>
</ul>

<p>In practice, the server always held the most up-to-date version of the files and distributed them to other devices.</p>

<h2 id="the-problem">The problem</h2>

<p>Over time, I realized that for this specific use case, Nextcloud is <strong>overkill</strong>.</p>

<p>Beyond file synchronization, it includes many additional features (calendar, notes, sharing, web UI, etc.) that I don’t actually use. This results in:</p>

<ul>
  <li>more complexity</li>
  <li>higher resource usage</li>
  <li>more maintenance</li>
</ul>

<h2 id="the-solution">The solution</h2>

<p>While looking for a simpler alternative, I came across Syncthing.</p>

<p>Syncthing is an open-source tool that synchronizes files in a <strong>peer-to-peer</strong> way, without requiring a central server or third-party services.</p>

<p>Each device communicates directly with the others and keeps files in sync automatically.</p>

<p>Some advantages:</p>

<ul>
  <li>simple setup</li>
  <li>no dependency on external cloud services</li>
  <li>direct synchronization between devices</li>
  <li>automatic connection management (even without a public IP)</li>
</ul>

<p>The setup is quick and well documented in the <a href="https://docs.syncthing.net/intro/getting-started.html">official guide</a>.</p>

<h2 id="do-you-still-need-a-server">Do you still need a server?</h2>

<p>Yes—if you want to replicate exactly the same behavior you had with Nextcloud.</p>

<p>Syncthing is P2P, which means files are only synchronized between devices that are online at the same time.</p>

<p>Example:</p>

<ul>
  <li>You modify a file on PC 1</li>
  <li>PC 2 is turned off</li>
  <li>If you shut down PC 1 before PC 2 turns on, the change won’t be transferred</li>
</ul>

<p>To solve this, you can use a third device that is <strong>always on</strong> (for example, a server):</p>

<ul>
  <li>it receives all changes</li>
  <li>stores them</li>
  <li>distributes them when other devices connect</li>
</ul>

<p>This way you get:</p>

<ul>
  <li>continuous synchronization</li>
  <li>a centralized copy of your files</li>
</ul>

<h2 id="syncthing-limitations">Syncthing limitations</h2>

<p>Syncthing is not a full replacement for Nextcloud.</p>

<p>Some features are missing:</p>

<ul>
  <li>advanced web interface for file management</li>
  <li>public link sharing</li>
  <li>integration with apps (calendar, notes, etc.)</li>
</ul>

<p>If you need those, Nextcloud is still a solid choice.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Nextcloud is an open-source self-hosted suite that lets you build a private cloud for storage, file synchronization, and much more.]]></summary></entry><entry xml:lang="it"><title type="html">Smetti di scrivere if in Bash</title><link href="https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash-it" rel="alternate" type="text/html" title="Smetti di scrivere if in Bash" /><published>2026-02-24T00:00:00+00:00</published><updated>2026-02-24T00:00:00+00:00</updated><id>https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash-it</id><content type="html" xml:base="https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash-it"><![CDATA[<p>Non comportarti come se Bash fosse Python o C. Il problema più comune negli script shell è l’uso eccessivo di <code class="language-plaintext highlighter-rouge">if</code> per controlli semplici. Anche se la logica condizionale serve, Bash ha un modo migliore e più idiomatico per farlo: i codici di uscita e gli operatori logici.</p>

<h2 id="le-basi">Le basi</h2>

<p>Ogni comando in Bash restituisce un codice di uscita. <code class="language-plaintext highlighter-rouge">0</code> significa successo, qualsiasi altro valore indica errore. Puoi concatenare comandi usando questi codici con operatori logici, evitando così lunghi blocchi <code class="language-plaintext highlighter-rouge">if</code>.</p>

<ul>
  <li><code class="language-plaintext highlighter-rouge">;</code> (punto e virgola): esegue il secondo comando indipendentemente dal risultato del primo.</li>
  <li><code class="language-plaintext highlighter-rouge">&amp;&amp;</code> (AND): esegue il secondo comando solo se il primo ha avuto successo.</li>
  <li><code class="language-plaintext highlighter-rouge">||</code> (OR): esegue il secondo comando solo se il primo ha fallito.</li>
</ul>

<h2 id="assegnazioni-condizionali-migliori">Assegnazioni condizionali migliori</h2>

<p>È comune controllare se una variabile è vuota e assegnarle un valore di default.</p>

<h3 id="metodo-lungo">Metodo lungo:</h3>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="k">if</span> <span class="o">[</span> <span class="nt">-z</span> <span class="s2">"</span><span class="nv">$EDITOR</span><span class="s2">"</span> <span class="o">]</span><span class="p">;</span> <span class="k">then
    </span><span class="nv">EDITOR</span><span class="o">=</span><span class="s2">"vim"</span>
<span class="k">fi</span>
</code></pre></div></div>

<p>Qui <code class="language-plaintext highlighter-rouge">[ -z "$EDITOR" ]</code> restituisce vero (0) se la variabile è vuota. L’operatore &amp;&amp; fa sì che l’assegnazione avvenga solo in quel caso.</p>

<p>Concatenare il flusso dei comandi</p>

<p>Questo approccio funziona bene anche per workflow più complessi. Invece di annidare if, usa catene logiche.</p>

<p>Problema con annidamenti:</p>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code>func1
<span class="k">if</span> <span class="o">[</span> <span class="nv">$?</span> <span class="nt">-eq</span> 0 <span class="o">]</span><span class="p">;</span> <span class="k">then
    </span>func2
    <span class="k">if</span> <span class="o">[</span> <span class="nv">$?</span> <span class="nt">-eq</span> 0 <span class="o">]</span><span class="p">;</span> <span class="k">then
        </span>func3
    <span class="k">else
        </span>handle_error
    <span class="k">fi
fi</span>
</code></pre></div></div>

<h2 id="catena-logica">Catena logica</h2>

<p><code class="language-plaintext highlighter-rouge">func1 &amp;&amp; func2 &amp;&amp; func3 || handle_error</code></p>

<table>
  <tbody>
    <tr>
      <td>Qui l’esecuzione è lineare: ogni comando parte solo se il precedente ha avuto successo. Se uno fallisce, la catena si interrompe e viene eseguito il blocco</td>
      <td> </td>
      <td>.</td>
    </tr>
  </tbody>
</table>

<p>Conclusione</p>

<p>L’eleganza negli script nasce dallo sfruttare il comportamento nativo della shell.</p>

<p>Usa if quando migliora davvero la leggibilità.</p>

<p>Per tutto il resto, le catene logiche sono spesso la scelta più pulita e idiomatica.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Non comportarti come se Bash fosse Python o C. Il problema più comune negli script shell è l’uso eccessivo di if per controlli semplici. Anche se la logica condizionale serve, Bash ha un modo migliore e più idiomatico per farlo: i codici di uscita e gli operatori logici.]]></summary></entry><entry><title type="html">Stop Writing if in Bash</title><link href="https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash" rel="alternate" type="text/html" title="Stop Writing if in Bash" /><published>2026-02-24T00:00:00+00:00</published><updated>2026-02-24T00:00:00+00:00</updated><id>https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash</id><content type="html" xml:base="https://samueleamato.xyz/2026/02/24/stop-writing-if-in-bash"><![CDATA[<p>Don’t act as though Bash is Python, or C. The most frequent trouble
with shell scripts is using too many <code class="language-plaintext highlighter-rouge">if</code> statements for simple control.
Although you’ll need conditional logic at times, Bash has its own,
better way – right in the command line – of doing this: exit codes and
logical operators.</p>

<h2 id="the-basics">The Basics</h2>

<p>Each command in Bash gives back an exit code when it’s done. <code class="language-plaintext highlighter-rouge">0</code> means
it worked, and anything else means it didn’t. You can link command
running by using these codes with logical operators, and so get rid of
long <code class="language-plaintext highlighter-rouge">if</code> blocks.</p>

<ul>
  <li>
    <p><code class="language-plaintext highlighter-rouge">;</code> (Semicolon): Runs the second command no matter what the first
command did. It is the same as a new line.</p>
  </li>
  <li>
    <p><code class="language-plaintext highlighter-rouge">&amp;&amp;</code> (AND): Runs the second command only if the first command worked
(exit code <code class="language-plaintext highlighter-rouge">0</code>).</p>
  </li>
  <li>
    <p><code class="language-plaintext highlighter-rouge">||</code> (OR): Runs the second command only if the first command failed
(non-zero exit code).</p>
  </li>
</ul>

<h2 id="making-conditional-assignments-better">Making Conditional Assignments Better</h2>

<p>It’s usual to see if a variable has been set, and if not, give it a
default.</p>

<h3 id="the-long-way">The Long Way:</h3>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="k">if</span> <span class="o">[</span> <span class="nt">-z</span> <span class="s2">"</span><span class="nv">$EDITOR</span><span class="s2">"</span> <span class="o">]</span><span class="p">;</span> <span class="k">then
    </span><span class="nv">EDITOR</span><span class="o">=</span><span class="s2">"vim"</span>
<span class="k">fi</span>
</code></pre></div></div>

<h3 id="the-bash-way">The Bash Way:</h3>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code><span class="o">[</span> <span class="nt">-z</span> <span class="s2">"</span><span class="nv">$EDITOR</span><span class="s2">"</span> <span class="o">]</span> <span class="o">&amp;&amp;</span> <span class="nv">EDITOR</span><span class="o">=</span><span class="s2">"vim"</span>
</code></pre></div></div>

<p>In this, <code class="language-plaintext highlighter-rouge">[ -z "$EDITOR" ]</code> is a command that returns true (<code class="language-plaintext highlighter-rouge">0</code>) if the
variable is empty. The <code class="language-plaintext highlighter-rouge">&amp;&amp;</code> operator makes sure the assignment happens
only when that’s the case. This turns three lines of standard code into
one line that’s easy to read.</p>

<h2 id="linking-command-flow">Linking Command Flow</h2>

<p>This works for involved workflows with several things that have to be
done. Rather than putting <code class="language-plaintext highlighter-rouge">if</code> statements inside each other to see if
each step worked, use running chains.</p>

<h3 id="the-nested-problem">The Nested Problem:</h3>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code>func1
<span class="k">if</span> <span class="o">[</span> <span class="nv">$?</span> <span class="nt">-eq</span> 0 <span class="o">]</span><span class="p">;</span> <span class="k">then
    </span>func2
    <span class="k">if</span> <span class="o">[</span> <span class="nv">$?</span> <span class="nt">-eq</span> 0 <span class="o">]</span><span class="p">;</span> <span class="k">then
        </span>func3
    <span class="k">else
        </span>handle_error
    <span class="k">fi
fi</span>
</code></pre></div></div>

<h3 id="the-logical-chain">The Logical Chain:</h3>

<div class="language-bash highlighter-rouge"><div class="highlight"><pre class="highlight"><code>func1 <span class="o">&amp;&amp;</span> func2 <span class="o">&amp;&amp;</span> func3 <span class="o">||</span> handle_error
</code></pre></div></div>

<p>In this structure, execution flows linearly. If <code class="language-plaintext highlighter-rouge">func1</code> succeeds,
<code class="language-plaintext highlighter-rouge">func2</code> runs. If <code class="language-plaintext highlighter-rouge">func2</code> succeeds, <code class="language-plaintext highlighter-rouge">func3</code> runs. If any command in the
<code class="language-plaintext highlighter-rouge">&amp;&amp;</code> sequence fails, the chain breaks immediately, skipping the
remaining <code class="language-plaintext highlighter-rouge">&amp;&amp;</code> steps and triggering the <code class="language-plaintext highlighter-rouge">||</code> block to handle the failure
(e.g., cleanup or error logging).</p>

<h2 id="conclusion">Conclusion</h2>

<p>Script elegance lies in leveraging the shell’s native behavior.</p>

<p>Use <code class="language-plaintext highlighter-rouge">if</code> when it genuinely improves clarity.</p>

<p>For everything else, logic chains are often the cleaner, more idiomatic
choice.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Don’t act as though Bash is Python, or C. The most frequent trouble with shell scripts is using too many if statements for simple control. Although you’ll need conditional logic at times, Bash has its own, better way – right in the command line – of doing this: exit codes and logical operators.]]></summary></entry><entry xml:lang="it"><title type="html">Leggere non ha nulla a che fare con la carta</title><link href="https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper-it" rel="alternate" type="text/html" title="Leggere non ha nulla a che fare con la carta" /><published>2025-12-27T00:00:00+00:00</published><updated>2025-12-27T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper-it</id><content type="html" xml:base="https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper-it"><![CDATA[<p>Ho spesso sentito dire che chi ama davvero leggere non può accontentarsi di un e-reader, ha bisogno della carta. In questo breve post voglio spiegare perché trovo questa idea sia limitata che completamente infondata.</p>

<p>Prima di tutto, definire l’uso di un e-reader come un “accontentarsi” non ha senso. A differenza di un libro fisico, un e-reader non limita ciò che puoi leggere. Anzi, può contenere migliaia di libri, mentre un libro cartaceo solo uno. Quindi dire che usare un e-reader significhi “accontentarsi” è semplicemente sbagliato.</p>

<p>Prima di confrontare e-reader e libri, però, bisogna chiarire cosa significa davvero “leggere”.</p>

<p>Leggere significa assorbire informazioni attraverso parole scritte. Nei secoli, le persone hanno scritto su vari supporti: pietra, argilla, pergamena, rotoli. Il libro, così come lo conosciamo, è relativamente recente ed è nato per comodità. Proprio per questo ha sostituito i supporti precedenti, e allo stesso modo gli e-reader sono destinati a sostituire i libri cartacei perché ancora più pratici. In sostanza, leggere significa elaborare informazioni scritte, indipendentemente dal supporto.</p>

<p>Detto questo, ecco un rapido confronto:</p>

<table>
  <thead>
    <tr>
      <th>Caratteristica</th>
      <th>E-Reader</th>
      <th>Libro cartaceo</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>Portabilità</td>
      <td>Migliaia di libri in un unico dispositivo</td>
      <td>Un libro alla volta, molto più ingombrante</td>
    </tr>
    <tr>
      <td>Batteria</td>
      <td>Dura settimane o mesi</td>
      <td>Non necessaria</td>
    </tr>
    <tr>
      <td>Personalizzazione</td>
      <td>Dimensione font regolabile, utile per ADHD o problemi visivi</td>
      <td>Nessuna personalizzazione</td>
    </tr>
    <tr>
      <td>Altri vantaggi</td>
      <td>Dizionario integrato, ricerca, evidenziazioni e note digitali</td>
      <td>Esperienza tattile, nessuna elettronica</td>
    </tr>
  </tbody>
</table>

<h2 id="quindi-perché-alcuni-preferiscono-ancora-la-carta"><em>Quindi perché alcuni preferiscono ancora la carta?</em></h2>

<p>Preferire la carta è legittimo: l’esperienza tattile può piacere, e capita anche a me, soprattutto se devo leggere un solo libro. Ma questa è una preferenza, non ha nulla a che vedere con la lettura in sé. Scegliere la carta può avere senso per piacere personale, ma dire che sia “migliore” di un e-reader è semplicemente falso.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Ho spesso sentito dire che chi ama davvero leggere non può accontentarsi di un e-reader, ha bisogno della carta. In questo breve post voglio spiegare perché trovo questa idea sia limitata che completamente infondata.]]></summary><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://samueleamato.xyz/assets/images/bookvsereader.jpg" /><media:content medium="image" url="https://samueleamato.xyz/assets/images/bookvsereader.jpg" xmlns:media="http://search.yahoo.com/mrss/" /></entry><entry><title type="html">Reading has nothing to do with paper</title><link href="https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper" rel="alternate" type="text/html" title="Reading has nothing to do with paper" /><published>2025-12-27T00:00:00+00:00</published><updated>2025-12-27T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper</id><content type="html" xml:base="https://samueleamato.xyz/2025/12/27/reading-has-nothing-to-do-with-paper"><![CDATA[<p>I’ve often heard people say that those who truly love reading can’t be satisfied with an e-reader, they need the feel of paper. In this short post, I want to explain why I find that idea both narrow-minded and completely unfounded.</p>

<p>First of all, calling the use of an e-reader “settling” makes no sense. Unlike a physical book, an e-reader doesn’t limit what you can read. In fact, it can hold thousands of books, while a single paper book can only hold one. So labeling the use of an e-reader as “settling” is just wrong.</p>

<p>But before comparing e-readers to paper books, I want to clarify what it actually means to “read” or to “want to read” a book.</p>

<p>Reading simply means absorbing information through written words. Over the centuries, people have written on various mediums, from stone and clay tablets to parchment and scrolls. Books, as we know them, are historically recent, and their rise was largely due to convenience. Exactly for this reason, books replaced older mediums, and in the same way, e-readers are poised to replace physical books because they are even more convenient. Ultimately, reading is about processing information through written words, whether those words are on paper, carved into stone, or displayed on a digital screen.</p>

<p>With that said, here’s a quick comparison between e-readers and paper books:</p>

<table>
  <thead>
    <tr>
      <th>Feature</th>
      <th>E-Reader</th>
      <th>Paper Book</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>Portability</td>
      <td>Can carry thousands of books in a tiny device</td>
      <td>One book at a time; takes up far more space</td>
    </tr>
    <tr>
      <td>Battery life</td>
      <td>Lasts weeks or even a month, so rarely an issue</td>
      <td>N/A</td>
    </tr>
    <tr>
      <td>Customization</td>
      <td>Adjustable font size, helpful for people with ADHD or vision challenges</td>
      <td>Fixed font, no customization</td>
    </tr>
    <tr>
      <td>Other advantages</td>
      <td>Built-in dictionary, search functions, highlights and notes saved digitally</td>
      <td>Tangible experience, no electronics required</td>
    </tr>
  </tbody>
</table>

<h2 id="so-why-do-some-people-still-prefer-paper"><em>So why do some people still prefer paper?</em></h2>

<p>Preferring paper is fine, it’s not wrong to enjoy the tactile experience of a physical book. I do it myself sometimes, especially if I know I’ll only read a single book and want the experience of holding it. But this preference has nothing to do with reading itself. Choosing paper makes sense for enjoyment, but claiming it’s inherently “better” than an e-reader is simply incorrect.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[I’ve often heard people say that those who truly love reading can’t be satisfied with an e-reader, they need the feel of paper. In this short post, I want to explain why I find that idea both narrow-minded and completely unfounded.]]></summary><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://samueleamato.xyz/assets/images/bookvsereader.jpg" /><media:content medium="image" url="https://samueleamato.xyz/assets/images/bookvsereader.jpg" xmlns:media="http://search.yahoo.com/mrss/" /></entry><entry xml:lang="it"><title type="html">Il mio setup</title><link href="https://samueleamato.xyz/2025/07/30/hardware-i-use-it" rel="alternate" type="text/html" title="Il mio setup" /><published>2025-07-30T00:00:00+00:00</published><updated>2025-07-30T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/07/30/hardware-i-use-it</id><content type="html" xml:base="https://samueleamato.xyz/2025/07/30/hardware-i-use-it"><![CDATA[<p><em>Nessun link affiliato: solo roba che uso davvero.</em></p>

<ul>
  <li>Setup scrivania
    <ul>
      <li>Mouse: <a href="https://it.aliexpress.com/item/1005007556974236.html">Seenda Trackball</a></li>
      <li>Tastiera: <a href="https://tex.com.tw/products/shura-diy-type">Shura (DIY)</a></li>
      <li>Microfono: <a href="https://rode.com/en-us/products/podmic">Rode PodMic</a></li>
      <li>Monitor: <a href="https://www.amazon.com/KOORUI-FreeSyncTM-Compatible-Ultra-Thin-24E4/dp/B09TTDRXNS?th=1">KOORUI 24” 1920×1080</a></li>
      <li>Braccio monitor: <a href="https://www.amazon.com/VIVO-Monitor-Adjustable-Screens-STAND-V002/dp/B009S750LA?dib=eyJ2IjoiMSJ9.52rzfU0jr0ATjhu5aeeFX5kFpW0OOVRwGyyQmW84zkOsp4piJWNjs51Q4aw-syY5Sw2qNk5kCkuieYBY4_tb9zziqj0-gvGMFbyQ6hWMjY5HM_sjufUFjAP2BGJnkN7aQaOXj-7HtiaToOLLD9gTika-_eiwz7Re8-WQq5ImcmSctogFlkSmMXJbSVuOAwgB8gp-ZYCxpyoOUQQl1-evv5vVReoO7-igFaZZcWLy9-w.2vdF9XlL_DVjTag0mnrZJVmXAV5sE1ccl35kTSyXuGM&amp;dib_tag=se&amp;keywords=monitor%2Bstand&amp;qid=1753906636&amp;sr=8-8&amp;th=1">VIVO Dual Monitor Desk Mount</a></li>
      <li>Auricolari: <a href="https://www.amazon.it/dp/B07QKYTGH9?th=1&amp;keywords=kz%2Bearphones&amp;linkCode=gs2&amp;tag=sixtytrend02-21">Linsoul KZ ZS10</a></li>
      <li>Tablet E-Ink: <a href="https://remarkable.com/products/remarkable-2">reMarkable 2</a></li>
      <li>Leggio: <a href="https://www.amazon.com/Readaeer-Cookbook-Kitchen-Foldable-Adjustable/dp/B06XQMVGMJ?crid=GTYZIK7E2MEU&amp;dib=eyJ2IjoiMSJ9.UZgyIfZ81lZCsi3qmUf4sxtYGBW8aI87AAUaD6eQCndCK_xj5Sy8_n_Uq1_R_2_IlqBGHKYYa7Lp_J-mFGwvwOK7RAwfKMELrMqY565-j8r5G9XwhJiezj73oHlHLsTbI5-V0Cn9PS-NVRwonwUH8fjBK6E4RTe_QnFJ9MPCoRKQnitjHbCJ_rsEA2Y6bT8mvPK23fP2SehnIyMio4qkXRpN7Owuc9CPCOjOiVaEn9KTzALAuSlw3esWHU9QoRklEQRYY0CVLZPEj3trSQ3tRpX2coMlxgiMCBPBtCwDB4Q.Q6mKgORsbLUl-m97Y3LHLwAi1SsXWjz85cSiAGoXEFA&amp;dib_tag=se&amp;keywords=leggio&amp;qid=1753906667&amp;sprefix=leggi%2Caps%2C233&amp;sr=8-9">Readaeer Bamboo Book Stand</a></li>
    </ul>
  </li>
  <li>Computer
    <ul>
      <li>laptop 1: Lenovo Ideapad Chromebook 15 (alpine linux)</li>
      <li>laptop 2: Samsung Galaxy Chromebook (arch linux)</li>
      <li><del>laptop 3: ThinkPad X220 (arch linux)</del> -&gt; <a href="https://www.ebay.it/itm/236431703086">in vendita qui per 160$</a></li>
      <li>computer 1: i7 10700 gtx 1660s 32gb ram (arch linux)</li>
    </ul>
  </li>
</ul>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Nessun link affiliato: solo roba che uso davvero.]]></summary></entry><entry><title type="html">My Gear</title><link href="https://samueleamato.xyz/2025/07/30/hardware-i-use" rel="alternate" type="text/html" title="My Gear" /><published>2025-07-30T00:00:00+00:00</published><updated>2025-07-30T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/07/30/hardware-i-use</id><content type="html" xml:base="https://samueleamato.xyz/2025/07/30/hardware-i-use"><![CDATA[<p><em>No affiliate links: just stuff I use.</em></p>

<ul>
  <li>Desk Gear
    <ul>
      <li>Mouse: <a href="https://it.aliexpress.com/item/1005007556974236.html">Seenda Trackball</a></li>
      <li>Keyboard: <a href="https://tex.com.tw/products/shura-diy-type">Shura (DIY)</a></li>
      <li>Microphone: <a href="https://rode.com/en-us/products/podmic">Rode PodMic</a></li>
      <li>Monitors: <a href="https://www.amazon.com/KOORUI-FreeSyncTM-Compatible-Ultra-Thin-24E4/dp/B09TTDRXNS?th=1">KOORUI 24” 1920×1080</a></li>
      <li>Monitor Arm: <a href="https://www.amazon.com/VIVO-Monitor-Adjustable-Screens-STAND-V002/dp/B009S750LA?dib=eyJ2IjoiMSJ9.52rzfU0jr0ATjhu5aeeFX5kFpW0OOVRwGyyQmW84zkOsp4piJWNjs51Q4aw-syY5Sw2qNk5kCkuieYBY4_tb9zziqj0-gvGMFbyQ6hWMjY5HM_sjufUFjAP2BGJnkN7aQaOXj-7HtiaToOLLD9gTika-_eiwz7Re8-WQq5ImcmSctogFlkSmMXJbSVuOAwgB8gp-ZYCxpyoOUQQl1-evv5vVReoO7-igFaZZcWLy9-w.2vdF9XlL_DVjTag0mnrZJVmXAV5sE1ccl35kTSyXuGM&amp;dib_tag=se&amp;keywords=monitor%2Bstand&amp;qid=1753906636&amp;sr=8-8&amp;th=1">VIVO Dual Monitor Desk Mount</a></li>
      <li>Earphones: <a href="https://www.amazon.it/dp/B07QKYTGH9?th=1&amp;keywords=kz%2Bearphones&amp;linkCode=gs2&amp;tag=sixtytrend02-21">Linsoul KZ ZS10</a></li>
      <li>E-Ink Tablet: <a href="https://remarkable.com/products/remarkable-2">reMarkable 2</a></li>
      <li>Book Stand: <a href="https://www.amazon.com/Readaeer-Cookbook-Kitchen-Foldable-Adjustable/dp/B06XQMVGMJ?crid=GTYZIK7E2MEU&amp;dib=eyJ2IjoiMSJ9.UZgyIfZ81lZCsi3qmUf4sxtYGBW8aI87AAUaD6eQCndCK_xj5Sy8_n_Uq1_R_2_IlqBGHKYYa7Lp_J-mFGwvwOK7RAwfKMELrMqY565-j8r5G9XwhJiezj73oHlHLsTbI5-V0Cn9PS-NVRwonwUH8fjBK6E4RTe_QnFJ9MPCoRKQnitjHbCJ_rsEA2Y6bT8mvPK23fP2SehnIyMio4qkXRpN7Owuc9CPCOjOiVaEn9KTzALAuSlw3esWHU9QoRklEQRYY0CVLZPEj3trSQ3tRpX2coMlxgiMCBPBtCwDB4Q.Q6mKgORsbLUl-m97Y3LHLwAi1SsXWjz85cSiAGoXEFA&amp;dib_tag=se&amp;keywords=leggio&amp;qid=1753906667&amp;sprefix=leggi%2Caps%2C233&amp;sr=8-9">Readaeer Bamboo Book Stand</a></li>
    </ul>
  </li>
  <li>Computers
    <ul>
      <li>laptop 1: Lenovo Ideapad Chromebook 15 (alpine linux)</li>
      <li>laptop 2: Samsung Galaxy Chromebook (arch linux)</li>
      <li><del>laptop 3: ThinkPad X220 (arch linux)</del> -&gt; <a href="https://www.ebay.it/itm/236431703086">selling here for 160$</a></li>
      <li>computer 1: i7 10700 gtx 1660s 32gb ram (arch linux)</li>
    </ul>
  </li>
</ul>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[No affiliate links: just stuff I use.]]></summary></entry><entry xml:lang="it"><title type="html">Il futuro dell’autenticità</title><link href="https://samueleamato.xyz/2025/07/30/the-future-of-authenticity-it" rel="alternate" type="text/html" title="Il futuro dell’autenticità" /><published>2025-07-30T00:00:00+00:00</published><updated>2025-07-30T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/07/30/the-future-of-authenticity-it</id><content type="html" xml:base="https://samueleamato.xyz/2025/07/30/the-future-of-authenticity-it"><![CDATA[<p>Qualche giorno fa stavo ascoltando una playlist lofi, una di quelle infinite usate come sottofondo per studiare. Dopo pochi minuti, qualcosa mi è sembrato strano. Per un attimo, il brano ha perso direzione: ritmo e melodia vagavano in modo forzato, come se stessero cercando di seguire uno schema senza riuscire davvero a farlo.</p>

<p>Ed è lì che ho avuto l’intuizione: probabilmente era <em>composto da un’IA</em>. All’inizio non ne ero sicuro, ma leggendo i commenti ho avuto conferma: anche altri ascoltatori avevano notato la stessa cosa.</p>

<p>Oggettivamente, non c’era “nulla che non andasse” nel suono. Ma sapere che <em>nessuno lo aveva realmente composto</em> ha cambiato completamente la mia esperienza. La musica non era sbagliata, ma risultava vuota, come se mancasse la sua funzione.</p>

<h2 id="perché-sapere-chi-crea-conta-ancora">Perché sapere chi crea conta ancora</h2>
<p>Non credo sia fondamentale associare un’opera al suo autore. Personalmente non lo faccio: un artista può essere antipatico, moralmente discutibile o semplicemente una “persona di merda”, ma se mi piace ciò che crea riesco comunque ad apprezzarlo. L’opera basta.</p>

<p>Quello che conta davvero è sapere che è stata creata da un <strong>essere umano</strong>, qualcuno capace di pensiero, intenzione ed emozione. Questo le dà un livello di significato diverso, anche se non ci interessa l’individuo in sé. C’è una differenza reale tra consumare qualcosa nato da un processo cosciente ed emotivo e qualcosa prodotto da un algoritmo.</p>

<p>Attribuiamo valore non solo alla perfezione, ma anche allo sforzo e all’intenzione. Uno schizzo fatto a mano, una melodia costruita in settimane, o un breve film realizzato con cura portano tracce di decisioni umane, tentativi ed investimento emotivo. Sapere questo ci fa connettere di più con l’opera.</p>

<p>Qualcuno potrebbe dire: “Se non sai che è IA, puoi comunque godertelo.”
È facile dirlo <strong>oggi</strong>, quando diamo per scontato che ogni canzone sia fatta da un umano. Ma in un futuro molto vicino, in cui i contenuti generati da IA saranno ovunque, il dubbio “sto ascoltando un artista vero?” emergerà molto più spesso.</p>

<h2 id="saturazione-dellia">Saturazione dell’IA</h2>

<p>Anche quando l’IA riesce a essere precisa quanto o più degli umani, emerge un pattern comune: i contenuti generati sono troppo perfetti. Nessun vero errore, nessuna deviazione, nessuna imperfezione o scelta inaspettata — ed è normale, perché sono algoritmi a operare.</p>

<p>L’IA può replicare stili, approcci e perfino segnali emotivi, ma quelle imperfezioni che caratterizzano il lavoro umano si perdono.</p>

<p>È simile a quanto successo con attori e modelli: un tempo avevano tratti distintivi, imperfezioni che li rendevano unici. Oggi, con standard estetici omologati, volti e performance sono uniformi. L’IA porta questo fenomeno all’estremo.</p>

<p>Si può dire che l’IA possa imitare gli errori umani — ed è vero. Ma non può crearne di nuovi: non si stanca, non dubita, non intuisce. Può solo combinare ciò che ha appreso. La perfezione algoritmica può imitare i difetti umani, ma non la loro origine.</p>

<p>Molte innovazioni artistiche nascono proprio da errori: il suono distorto della chitarra nel rock è nato da un amplificatore danneggiato, il jump cut nel cinema da un errore di montaggio, interi movimenti artistici da limiti o incidenti.</p>

<h2 id="il-marchio-creato-da-umani">Il marchio “creato da umani”</h2>

<p>Dopo tutto questo, non è difficile immaginare un futuro in cui nasceranno etichette — e forse anche istituzioni — per certificare l’autenticità umana delle opere.</p>

<p>Un badge “Human-Made Certified” potrebbe diventare un nuovo segno di qualità, indicando che nessuna IA è stata coinvolta nel processo creativo. Permetterebbe a chiunque di sapere immediatamente se ciò che sta guardando, leggendo o ascoltando è nato da una mente umana o da un sistema algoritmico.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[Qualche giorno fa stavo ascoltando una playlist lofi, una di quelle infinite usate come sottofondo per studiare. Dopo pochi minuti, qualcosa mi è sembrato strano. Per un attimo, il brano ha perso direzione: ritmo e melodia vagavano in modo forzato, come se stessero cercando di seguire uno schema senza riuscire davvero a farlo.]]></summary><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://samueleamato.xyz/assets/images/human-made.jpg" /><media:content medium="image" url="https://samueleamato.xyz/assets/images/human-made.jpg" xmlns:media="http://search.yahoo.com/mrss/" /></entry><entry><title type="html">The Future of Authenticity</title><link href="https://samueleamato.xyz/2025/07/30/the-future-of-authenticity" rel="alternate" type="text/html" title="The Future of Authenticity" /><published>2025-07-30T00:00:00+00:00</published><updated>2025-07-30T00:00:00+00:00</updated><id>https://samueleamato.xyz/2025/07/30/the-future-of-authenticity</id><content type="html" xml:base="https://samueleamato.xyz/2025/07/30/the-future-of-authenticity"><![CDATA[<p>A few days ago, I had been listening to a lofi playlist, one of those endless background mixes people use to study. In a few minutes, something felt off. Briefly, the track lost its point, the rhythm and the melody wandered in a way that felt forced, as though it was trying to follow a pattern but couldn’t commit to one.</p>

<p>And that was when it hit me: this likely was <em>AI-composed</em>. I wasn’t sure at first, but reading the comments put my mind at ease, some of the other listeners had noticed the same thing.</p>

<p>Objectively, there was ““nothing wrong”” with the sound. But having discovered that <em>no one had actually composed it</em> made all the difference to my experience. The music wasn’t “wrong”, but it felt hollow, like the function of it was missing.</p>

<h2 id="why-knowing-the-creator-still-matters">Why Knowing the Creator Still Matters</h2>
<p>I do not believe that it’s important to associate a piece of art with its artist. Personally, i dont do: an artist may be obnoxious, morally dubious, or simply a “piece of shit,” yet if I like his work I am able to appreciate it completely. The art itself is sufficient.</p>

<p>What does matter is that we know that a <strong>human being</strong>, someone capable of thought, intention and feeling, made it. Knowing that gives it a level of meaning, even if we don’t particularly care about the individual who created it. There is a small but real distinction between consuming something that emerged from a conscious, emotional process and something produced by an algorithm.</p>

<p>We attribute value not only to perfection, but to effort and intent. A sketch drawn by hand, a tune crafted over a series of weeks, or a brief film made with attention all carry traces of human decision-making, testing, and emotional investment. Having this knowledge inspires us to connect more deeply with the work and to appreciate it.</p>

<p>Some might say: “If you don’t know it’s AI, you can still enjoy it.”
That’s easy to say <strong>now</strong>, when we automatically assume every song we hear was made by a human. But in a not-so-distant future, when AI-generated content becomes everyday consumption, the doubt “am I listening to a real artist or not?” will come up much more often.</p>

<h2 id="ai-saturation">AI Saturation</h2>

<p>Even when AI succeeds at being as accurate as humans, or even more accurate, there’s a common pattern: AI-generated content is too perfect. No real “mistakes,” no small deviations, impefection or unexpected choices, and that makes sense, because it’s algorithms that are doing the job. AI can replicate styles, approaches, and even emotional signals, but those imperfections that mark human work simply get lost.</p>

<p>It’s similar to how actors or models used to have unique features, flaws that gave them character and individuality. Today, with beauty standards homogenized, faces and performances are polished to near uniformity. AI takes this phenomenon to the extreme.</p>

<p>One may argue that AI can mimic human mistakes, and to a certain extent, it is possible. AI, however, cannot make genuinely new human mistakes — it cannot get tired, doubt, or intuit. It can only mix and match within the algorithms it has learned. That is, algorithmic perfection can imitate human failings, but never their origin.</p>

<p>Some may argue that AI can imitate human mistakes, that’s true. But precisely because it can only <em>imitate</em>, it can never create new ones. Many artistic breakthroughs were born out of human error: the distorted guitar tone that defined rock music was the result of a <a href="https://en.wikipedia.org/wiki/Rocket_88">damaged amplifier</a>, the jump cut in film editing emerged from an <a href="https://en.wikipedia.org/wiki/Jump_cut">editing mistake</a> that turned into a stylistic revolution; entire art movements, from abstract expressionism to glitch aesthetics, were sparked by accidents, limitations, or missteps.</p>

<h2 id="the-human-created-label">The “Human-Created” Label</h2>

<p>After all this, it’s not hard to imagine a future where new labels, and perhaps even dedicated institutions, emerge to certify the human authenticity of creative works.</p>

<p>A “Human-Made Certified” badge could become the new mark of integrity, signaling that no AI was involved in the creative process. It would allow any person to immediately know if what they’re watching, reading, or listening to was born from a human mind or an algorithmic system.</p>]]></content><author><name>Samuele Amato</name><email>samuele@samueleamato.xyz</email></author><summary type="html"><![CDATA[A few days ago, I had been listening to a lofi playlist, one of those endless background mixes people use to study. In a few minutes, something felt off. Briefly, the track lost its point, the rhythm and the melody wandered in a way that felt forced, as though it was trying to follow a pattern but couldn’t commit to one.]]></summary><media:thumbnail xmlns:media="http://search.yahoo.com/mrss/" url="https://samueleamato.xyz/assets/images/human-made.jpg" /><media:content medium="image" url="https://samueleamato.xyz/assets/images/human-made.jpg" xmlns:media="http://search.yahoo.com/mrss/" /></entry></feed>