<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: aliosn bekar thiago silba</title>
    <description>The latest articles on DEV Community by aliosn bekar thiago silba (@aliosn_bekarthiagosilba).</description>
    <link>https://dev.to/aliosn_bekarthiagosilba</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4245958%2F04c70a7e-4716-43b7-8355-8a6823ff2450.png</url>
      <title>DEV Community: aliosn bekar thiago silba</title>
      <link>https://dev.to/aliosn_bekarthiagosilba</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/aliosn_bekarthiagosilba"/>
    <language>en</language>
    <item>
      <title>Three states your network client must show, and one it should hide</title>
      <dc:creator>aliosn bekar thiago silba</dc:creator>
      <pubDate>Mon, 16 Nov 2026 12:53:07 +0000</pubDate>
      <link>https://dev.to/aliosn_bekarthiagosilba/three-states-your-network-client-must-show-and-one-it-should-hide-1ihd</link>
      <guid>https://dev.to/aliosn_bekarthiagosilba/three-states-your-network-client-must-show-and-one-it-should-hide-1ihd</guid>
      <description>&lt;p&gt;Most network clients have a single button and a lot of hidden state. That works until something goes wrong, and then the user has no vocabulary to describe what happened. Three states solve most of it.&lt;/p&gt;

&lt;h2&gt;
  
  
  1. Connecting is not the same as connected
&lt;/h2&gt;

&lt;p&gt;The moment between pressing connect and traffic actually flowing is where users get lost. If the button keeps saying "Connect", people press it again, and again, and now you have three connection attempts and a support ticket.&lt;/p&gt;

&lt;p&gt;Show the state explicitly: idle, connecting, connected, reconnecting, error. A spinner alone is not a state, it is an anxiety indicator.&lt;/p&gt;

&lt;h2&gt;
  
  
  2. The route is part of the state
&lt;/h2&gt;

&lt;p&gt;"Connected" on a route with 400 ms latency is technically true and practically useless. Users cannot tell the difference between a slow route and a broken client unless you surface both facts together: which route, and how slow.&lt;/p&gt;

&lt;p&gt;This one change replaced an entire diagnostics screen that nobody opened. The status line became the diagnostics screen.&lt;/p&gt;

&lt;h2&gt;
  
  
  3. The device count is invisible until it breaks
&lt;/h2&gt;

&lt;p&gt;A per-account device limit is the most common "it stopped working" cause in consumer network clients, and almost nobody shows the count in the UI. One line under the status, &lt;code&gt;2 of 3 devices online&lt;/code&gt;, prevents a whole category of support requests.&lt;/p&gt;

&lt;h2&gt;
  
  
  What to hide: the route list
&lt;/h2&gt;

&lt;p&gt;The opposite mistake is showing everything. A list of twenty endpoints with latency numbers is not transparency, it is a decision the user never asked to make. Probe, pick, remember, and expose a single override control for the people who actually want it.&lt;/p&gt;




&lt;p&gt;The client I have been writing about is QuickQ: one-click connect, automatic route selection per network, native clients for Windows, macOS, iOS, Android and Ubuntu Linux, and one account with up to 3 devices online at once. I keep notes on the flow, and the product page is here if you want to see the result: &lt;a href="https://quickq-check.com.cn/" rel="noopener noreferrer"&gt;https://quickq-check.com.cn/&lt;/a&gt; — it is an independent write-up, not the vendor site, so verify current details before relying on it.&lt;/p&gt;

</description>
      <category>networking</category>
      <category>ux</category>
      <category>webdev</category>
    </item>
    <item>
      <title>Keeping five platform clients in sync: notes from a cross-platform network app</title>
      <dc:creator>aliosn bekar thiago silba</dc:creator>
      <pubDate>Mon, 16 Nov 2026 06:07:39 +0000</pubDate>
      <link>https://dev.to/aliosn_bekarthiagosilba/keeping-five-platform-clients-in-sync-notes-from-a-cross-platform-network-app-30ci</link>
      <guid>https://dev.to/aliosn_bekarthiagosilba/keeping-five-platform-clients-in-sync-notes-from-a-cross-platform-network-app-30ci</guid>
      <description>&lt;p&gt;Shipping one client is a UI problem. Shipping five that behave identically is a process problem. These are the practices that kept parity manageable on a network client that runs on Windows, macOS, iOS, Android and Ubuntu Linux.&lt;/p&gt;

&lt;h2&gt;
  
  
  1. Write the connect flow down once
&lt;/h2&gt;

&lt;p&gt;Before any code, we wrote the flow as a numbered list: install, sign in, connect, show status, switch route, disconnect. Every platform implements that list, in that order. When a platform wanted a different order, the default answer was no, and the reason had to be a platform constraint rather than a designer preference.&lt;/p&gt;

&lt;h2&gt;
  
  
  2. One status vocabulary
&lt;/h2&gt;

&lt;p&gt;States are named identically everywhere: idle, connecting, connected, reconnecting, error. The words on screen differ by language, never by platform. This sounds trivial until a support ticket says "the Mac version is stuck" and the actual state is reconnecting on all platforms.&lt;/p&gt;

&lt;h2&gt;
  
  
  3. Latency is data, not decoration
&lt;/h2&gt;

&lt;p&gt;The status line shows the live route and its latency. It is fed by the same measurement on every client, so a screenshot from a phone can be compared with one from a laptop. That single decision replaced a diagnostics screen that nobody opened.&lt;/p&gt;

&lt;h2&gt;
  
  
  4. Ship the boring platform first
&lt;/h2&gt;

&lt;p&gt;We stabilised on desktop before mobile. Mobile adds background restrictions, battery rules and store review, and doing that before the core flow is settled means you debug two problems at once.&lt;/p&gt;

&lt;h2&gt;
  
  
  5. Keep the account model boring
&lt;/h2&gt;

&lt;p&gt;One account, up to 3 devices online at the same time. No per-platform pricing, no separate mobile subscription. Users move between devices; the account should not care which one is in front of them.&lt;/p&gt;

&lt;h2&gt;
  
  
  What parity is not
&lt;/h2&gt;

&lt;p&gt;Parity does not mean identical code. It means identical behaviour. The Linux build uses a different networking stack than iOS, and that is fine, because the user-visible contract is the same.&lt;/p&gt;

&lt;p&gt;If you want to see the result, the flow is documented here: &lt;a href="https://quickq-check.com.cn/" rel="noopener noreferrer"&gt;https://quickq-check.com.cn/&lt;/a&gt; — it is an independent write-up, not the vendor site, and platform support or pricing can change, so check before you rely on it.&lt;/p&gt;

</description>
      <category>productivity</category>
    </item>
    <item>
      <title>A five-minute triage for remote-desktop lag</title>
      <dc:creator>aliosn bekar thiago silba</dc:creator>
      <pubDate>Sun, 15 Nov 2026 14:14:50 +0000</pubDate>
      <link>https://dev.to/aliosn_bekarthiagosilba/a-five-minute-triage-for-remote-desktop-lag-3ngg</link>
      <guid>https://dev.to/aliosn_bekarthiagosilba/a-five-minute-triage-for-remote-desktop-lag-3ngg</guid>
      <description>&lt;p&gt;Remote desktop sessions that "feel slow" usually fail on one of three things, and each has a different fix. Before opening a ticket about bandwidth, spend five minutes on this order.&lt;/p&gt;

&lt;h2&gt;
  
  
  1. Is it input latency or render latency?
&lt;/h2&gt;

&lt;p&gt;Type in a local text field, then type in the remote one. If the characters appear late but arrive in a burst, the problem is the path (latency + jitter). If they appear smoothly but the screen redraws slowly, the problem is encoding or the remote GPU. These look similar on a video call and are completely different problems.&lt;/p&gt;

&lt;h2&gt;
  
  
  2. Measure the path, not the speed
&lt;/h2&gt;

&lt;p&gt;A speed test measures throughput to a nearby server and tells you almost nothing about a cross-region session. You want round-trip time, loss and jitter to the actual destination:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;mtr -rwzbc 100 &lt;span class="o"&gt;&amp;lt;&lt;/span&gt;remote-host&lt;span class="o"&gt;&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Read it top to bottom. The first hop where loss stays flat and latency jumps is the segment to avoid. Anything under ~0.5% loss is usually absorbed by TCP; above 1% you feel it as stalls.&lt;/p&gt;

&lt;h2&gt;
  
  
  3. Check whether the session survives a route change
&lt;/h2&gt;

&lt;p&gt;If a route switch resets the TCP connection, every switch is a visible freeze, and the client will look worse than a static path. A session-preserving switch - moving packets to a better path without tearing down the socket - is what turns an unstable link into a usable one. This is the part a &lt;a href="https://kawlian.com.cn/" rel="noopener noreferrer"&gt;cross-border network acceleration&lt;/a&gt; client is responsible for; the measurement is the easy half.&lt;/p&gt;

&lt;h2&gt;
  
  
  4. Only then look at bandwidth
&lt;/h2&gt;

&lt;p&gt;Bandwidth matters for file transfer and video, not for typing. If your session is unusable at 5 Mbps but the link shows 100 Mbps idle, bandwidth was never the bottleneck.&lt;/p&gt;

&lt;h2&gt;
  
  
  The short version
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Input lag in bursts -&amp;gt; path (latency/jitter/loss) -&amp;gt; measure with mtr.&lt;/li&gt;
&lt;li&gt;Smooth input, slow redraw -&amp;gt; encoding/GPU -&amp;gt; check codec and resolution.&lt;/li&gt;
&lt;li&gt;Freeze on every route change -&amp;gt; the client, not the network.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Most "slow remote desktop" reports I see resolve at step 2, and the ones that do not are usually step 3 wearing a disguise.&lt;/p&gt;

</description>
      <category>devops</category>
    </item>
    <item>
      <title>Latency, loss or jitter? Three numbers that explain a slow cross-border link</title>
      <dc:creator>aliosn bekar thiago silba</dc:creator>
      <pubDate>Sun, 15 Nov 2026 13:12:05 +0000</pubDate>
      <link>https://dev.to/aliosn_bekarthiagosilba/latency-loss-or-jitter-three-numbers-that-explain-a-slow-cross-border-link-5fin</link>
      <guid>https://dev.to/aliosn_bekarthiagosilba/latency-loss-or-jitter-three-numbers-that-explain-a-slow-cross-border-link-5fin</guid>
      <description>&lt;p&gt;When a connection to another continent feels slow, the reflex is to blame bandwidth. Bandwidth is rarely the limit. Three other numbers decide what a link actually feels like: round-trip latency, packet loss and jitter. They show up as different symptoms, need different measurements, and only some of them can be fixed by choosing a better route.&lt;/p&gt;

&lt;h2&gt;
  
  
  Latency is a floor, not a fault
&lt;/h2&gt;

&lt;p&gt;Round-trip time is the shortest time a packet needs to reach the far end and come back. It is set by physics plus the path: a detour through another region adds 50-150 ms before any congestion is involved. A link at 180 ms RTT will feel sluggish for keystrokes and remote desktop, no matter how much bandwidth you throw at it.&lt;/p&gt;

&lt;p&gt;The useful question is not "how fast is it" but "how close is it to the floor". Measure the RTT to several nearby targets and to the actual destination. If the destination is 80 ms worse than a peer in the same city, the route is the problem, not the capacity.&lt;/p&gt;

&lt;h2&gt;
  
  
  Loss is what users call "stalling"
&lt;/h2&gt;

&lt;p&gt;Packet loss below ~0.5% is mostly absorbed by TCP. Above 1% it stops being invisible: transfers stall, video buffers, SSH sessions freeze for a second and then catch up. Loss on a cross-border path is usually congestion at a peering point or a saturated transit link, and it is rarely symmetrical - the return path is often the worse one.&lt;/p&gt;

&lt;p&gt;Loss is also the number most likely to be misread. A single ping timeout is not loss; you need a few hundred probes before the percentage means anything.&lt;/p&gt;

&lt;h2&gt;
  
  
  Jitter is what makes a call unusable
&lt;/h2&gt;

&lt;p&gt;Jitter is the variation in latency between packets. Audio and video calls buffer a small window to hide it, so high jitter shows up as robotic voice and frozen frames even when average latency looks fine. A route with 120 ms average and 5 ms jitter is far more usable than one with 90 ms average and 60 ms of jitter.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;&lt;span class="c"&gt;# one pass, three numbers: loss per hop, average RTT, jitter&lt;/span&gt;
mtr -rwzbc 100 203.0.113.10
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Read the report top to bottom. The first hop where loss stays constant and latency jumps is the segment you actually need to avoid.&lt;/p&gt;

&lt;h2&gt;
  
  
  What route selection can and cannot fix
&lt;/h2&gt;

&lt;p&gt;Nothing fixes distance. If the destination is 9,000 km away, no client will bring the RTT under the physical minimum. What a routing layer can do is avoid the parts of the path that are congested or unstable:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;probe several entry points continuously instead of once at connect time;&lt;/li&gt;
&lt;li&gt;score each path on latency, loss and jitter, not on latency alone;&lt;/li&gt;
&lt;li&gt;move an established session to a better path without tearing the connection down;&lt;/li&gt;
&lt;li&gt;pin a route when a user asks for it, so a "randomly better" switch does not surprise them mid-call.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;That last point is where most implementations get it wrong: a switch that resets the TCP connection is worse than the jitter it was trying to fix. The switch has to be invisible to the socket.&lt;/p&gt;

&lt;p&gt;This is the part a &lt;a href="https://kawlian.com.cn/" rel="noopener noreferrer"&gt;cross-border network acceleration&lt;/a&gt; client is responsible for - measuring the three numbers above on more than one path, and moving the session before the user notices. The measurement side is the easy half; doing the switch without dropping the connection is the hard one.&lt;/p&gt;

&lt;h2&gt;
  
  
  A short checklist
&lt;/h2&gt;

&lt;ol&gt;
&lt;li&gt;Measure RTT against a nearby peer, not against a speed-test server.&lt;/li&gt;
&lt;li&gt;Use &lt;code&gt;mtr&lt;/code&gt; for at least 100 probes before believing a loss number.&lt;/li&gt;
&lt;li&gt;Treat jitter as its own metric - it predicts call quality better than average latency.&lt;/li&gt;
&lt;li&gt;When a path is bad, ask whether the client can switch session-preserving, or whether it needs a reconnect.&lt;/li&gt;
&lt;li&gt;Re-check after a fix: a route that improved at 10:00 can degrade again at 20:00 peak.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;None of this needs new tooling. &lt;code&gt;ping&lt;/code&gt;, &lt;code&gt;mtr&lt;/code&gt; and a timestamped log of the three numbers will tell you more than any single "speed" figure, and it will tell you which of the three is actually costing your users time.&lt;/p&gt;

</description>
      <category>devops</category>
    </item>
  </channel>
</rss>
