<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>skrot</title>
    <link>https://scrolls.anduin.net/ltning/</link>
    <description>Skrot[NO-NB] = junk; these are my retrocomputing notes</description>
    <pubDate>Tue, 17 Nov 2026 19:08:01 +0000</pubDate>
    <item>
      <title>OS/2: Connecting ezIRC/2 to ZNC</title>
      <link>https://scrolls.anduin.net/ltning/os-2-connecting-ezirc-2-to-znc</link>
      <description>&lt;![CDATA[For anyone who has followed my escapades in retrocomputing at all, me &#34;doing silly stuff with old stuff&#34; should not surprise anyone. But on the list of weird problems I&#39;ve decided to tackle, this one might be among the silliest.&#xA;&#xA;Wait, what?&#xA;So I&#39;m a huge OS/2 fan, as I have been since the early 90&#39;s when the first beta versions became availablea id=&#34;n1&#34;[1/a]. The first versions were incredibly painful to run on the hardware I had: My dad&#39;s 25MHz 486 - can&#39;t remember if it was an SX or a DX - had 4MB of RAM. While this did meet the minimum requirements, it was really, really painful. My own 16MHz 386SX at home had all of 5MB, and despite the dog slow CPU (compared to the 486), it practically flew.&#xA;&#xA;Once I got past that initial pain, the wonders of this operating system did not cease to impress me. I remained a fan, and avid user, until around 2003, when I finally switched to FreeBSD even on my desktop and laptop machines.&#xA;&#xA;It should therefore surprise exactly nobody that I&#39;m now running OS/2, 24/7 or at least part-time, on \checks notes\ ...8? machinesa id=&#34;n2&#34;[2/a].&#xA;&#xA;And .. IRC?&#xA;Back in 1996 when I started studying was when I met my first UNIX machines. The computer room at uni was full of Ultrix, HP-UX, Solaris and other things I had heard of but never actually used. I was still using OS/2 privately, but I could not help being fascinated by this new world that opened up to me.&#xA;&#xA;Among the first applications I truly fell in love with was IRC. I don&#39;t even know what the first IRC client I used was, probably ircII or some variant thereof. I soon learned of scripts and such to improve the &#34;user experience&#34;, and before long I was running OpenChat/2 on my OS/2 machines. But since 1996 was already the year of OS/2 Warp 4, I was late to the game: Others had been on IRC for years already, and there were several older IRC applications out there that - at the time - I didn&#39;t know of, nor did I need them.&#xA;&#xA;Fast forward a couple dozen years though, and I find myself running significantly older OS/2 versions - like 2.11 - on machines much older than the PC I had at the time. But since there&#39;s TCP/IP even for this old OS/2 version, I do need to make it &#34;useful&#34;, so off I go looking for IRC clients that work - because my old friend OpenChat/2 requires newer stuff to run.&#xA;&#xA;Enter ezIRC!&#xA;At some point in the mid-90s, Roger Bess released ezIRC. I don&#39;t have a good history for this program, as the only archived website seems to have been published around the time of the last known release. But of all the PM-type (GUI) IRC applications I&#39;ve managed to find, it&#39;s about the only one that works on OS/2 2.x and doesn&#39;t look absolutely bonkers.&#xA;&#xA;Problem is .. it&#39;s old. And, when exposed to &#34;modern&#34; IRC servers, UTF-8, protocol extensions and weird server CAPabilities, it gets real confused, real fast. It also does not support server passwordsa id=&#34;n3&#34;[3/a], so connecting to ZNC is not actully possible.&#xA;&#xA;After a lot of trial and error (and tcpdumping and whatnot), plus some help (and only limited amounts of mockery) from the fine people on \#znc (Libera), I arrived at a solution\^Wworkaround.&#xA;&#xA;As a bonus, this involves the use of another, only slightly newer, piece of hardware.&#xA;&#xA;The solution: Irssi&#xA;One of the possible solutions I was looking for was some kind of IRC proxy. This however does not really seem to be a thing: I found some variants over the theme, but they were either actual bouncers (already got one, thanks!) or they were solving different sets of problems - and in most cases they were horribly obsolete.&#xA;&#xA;As a side note, my requirement for a solution was that it could only involve software that could run on one of my other retro-rigs; the prime candidate being my trusty dual Pentium Pro/P2OD machine running #NetBSD. Turns out, I already had the necessary software installed!&#xA;&#xA;Irssi has a surprisingly useful feature they call Irssiproxy. Basically this sets up a separate listening port in Irssi itself, which allows connecting clients to share the associated IRC network connection with Irssi.&#xA;&#xA;How perfect is this?!? I can configure the networks i need and associated ZNC connections in Irssi, and have it listen on a set of port numbers on my LAN - and ezIRC can connect with no additional hassle.&#xA;&#xA;Except....&#xA;The only real issue I encountered (other than PEBKAC situaitons) was the aforementioned server capability messages which caused ezIRC some confusion. Basically it could join channels and send messages to them, but inbound messages were just dropped on the floor (or, in some cases, shown in the &#34;server messages&#34; window).&#xA;&#xA;I tried connection Irssi directly to the target IRC network (bypassing ZNC), which seemed to solve the problem - but also got in the way of what I tried to achieve.&#xA;&#xA;With some tcpdump and Wireshark work, I found that ZNC adds @time= to the messages. As confirmed by one of the helpful souls on IRC, this is only added if the connecting client asks for it. Luckily the solution is easy: Tell Irssi to skip sending the CAPREQ to ZNC.&#xA;&#xA;The configuration&#xA;First, the server config:&#xA;&#xA;servers = (&#xA;  {&#xA;    address = &#34;znc.example.com&#34;;&#xA;    chatnet = &#34;os2Floppy&#34;;&#xA;    port = &#34;195&#34;;&#xA;    password = &#34;user@client-id/network:password&#34;;&#xA;    usetls = &#34;yes&#34;;&#xA;    tlsverify = &#34;yes&#34;;&#xA;    nocap = &#34;yes&#34;;&#xA;  },&#xA;  {&#xA;    address = &#34;znc.example.com&#34;;&#xA;    chatnet = &#34;os2EFNet&#34;;&#xA;    port = &#34;195&#34;;&#xA;    password = &#34;user@client-id/network:password&#34;;&#xA;    usetls = &#34;yes&#34;;&#xA;    tlsverify = &#34;yes&#34;;&#xA;    nocap = &#34;yes&#34;;&#xA;  },&#xA;  {&#xA;    address = &#34;znc.example.com&#34;;&#xA;    chatnet = &#34;os2Libera&#34;;&#xA;    port = &#34;195&#34;;&#xA;    password = &#34;user@client-id/network:password&#34;;&#xA;    usetls = &#34;yes&#34;;&#xA;    tlsverify = &#34;yes&#34;;&#xA;    nocap = &#34;yes&#34;;&#xA;  }&#xA;);&#xA;&#xA;No channels configured here, as ZNC takes care of that&#xA;channels = ( );&#xA;&#xA;And then the core config; note the proxy section:&#xA;&#xA;settings = {&#xA;  core = {&#xA;    realname = &#34;Eirik Overby&#34;;&#xA;    username = &#34;ltning&#34;;&#xA;    nick = &#34;o-Ltning&#34;;&#xA;    recodetransliterate = &#34;no&#34;;&#xA;  };&#xA;  proxy = {&#xA;    irssiproxybind = &#34;192.88.99.80&#34;; # The local IP&#xA;    irssiproxyports = &#34;os2Floppy=2777 os2EFNet=2778 os2Libera=2779&#34;; # Per-chatnet listening ports&#xA;  };&#xA;};&#xA;&#xA;The chat networks, as used by the server config:&#xA;chatnets = {&#xA;  os2Floppy = { type = &#34;IRC&#34;; };&#xA;  os2EFNet = { type = &#34;IRC&#34;; };&#xA;  os2Libera = { type = &#34;IRC&#34;; };&#xA;};&#xA;&#xA;The result&#xA;And with that, I have a working config. The obligatory screenshot below.&#xA;&#xA;Good luck (if you should end up trying this yourself), and thanks for reading!&#xA;&#xA;ezIRC running on OS/2 2.11&#xA;&#xA;Footnotes&#xA;a id=&#34;fn1&#34;1/a: I spent many hours pestering IBM Norway into sending me a copy of the beta; I&#39;m sure I ended up on some list somewhere.&#xA;a id=&#34;fn2&#34;2/a: The current builds include&#xA;    A 486DX-50 running Warp 3 and Novell Netware 4.12 paravirtualized&#xA;    A 486DRx2-50 (really a 386 with a bit of cache) running Warp 3 and my BBS&#xA;    The subject of this article, a 486DLC-40 running 2.11&#xA;    A dual Pentium 233 MMX running Warp 4.52&#xA;    A 286 running OS/2 1.3 EE(!)&#xA;    A dual Pentium Pro which, when not running NetBSD or DOS runs Warp Server for e-Business&#xA;    An IBM 486slc2 - Leopard or Leopard LX-based, depending on my mood - running Warp 3&#xA;    An IBM PS/2 model P70 luggable running 2.11&#xA;a id=&#34;fn3&#34;3/a: Turns out, it does as of version 1.2o. But given all the other problems, this is mostly of academic interest at this point..&#xA;&#xA;div style=&#34;display: flex; place-content: center;&#34;table style=&#34;width: 80%; border: none; font-size: 0.8em;&#34;&#xD;&#xA;tr style=&#34;border: none;&#34;&#xD;&#xA;td align=&#34;center&#34; style=&#34;border: none;&#34; colspan=&#34;2&#34;hr&#xD;&#xA;~ love over gold ~/td&#xD;&#xA;/tr&#xD;&#xA;tr style=&#34;border: none; background-color: inherit;&#34;&#xD;&#xA;td align=&#34;left&#34; style=&#34;border: none; padding-left: 2em;&#34;a href=&#34;http://floppy.museum&#34;The Floppy Museum/a/td&#xD;&#xA;td align=&#34;right&#34; style=&#34;border: none; padding-right: 2em;&#34;a href=&#34;http://floppy.museum/thispc.htm&#34;About the Museum/a/td&#xD;&#xA;/tr/table/div]]&gt;</description>
      <content:encoded><![CDATA[<p>For anyone who has followed my escapades in retrocomputing at all, me “doing silly stuff with old stuff” should not surprise anyone. But on the list of weird problems I&#39;ve decided to tackle, this one might be among the silliest.</p>

<h2 id="wait-what">Wait, what?</h2>

<p>So I&#39;m a huge <a href="https://en.wikipedia.org/wiki/OS/2" rel="nofollow">OS/2</a> fan, as I have been since the early 90&#39;s when the first beta versions became available[<a id="n1"><a href="#fn1" rel="nofollow">1</a></a>]. The first versions were incredibly painful to run on the hardware I had: My dad&#39;s 25MHz 486 – can&#39;t remember if it was an SX or a DX – had 4MB of RAM. While this did meet the minimum requirements, it was really, really painful. My own 16MHz 386SX at home had all of <em>5</em>MB, and despite the dog slow CPU (compared to the 486), it practically <em>flew</em>.</p>

<p>Once I got past that initial pain, the wonders of this operating system did not cease to impress me. I remained a fan, and avid user, until around 2003, when I finally switched to <a href="https://freebsd.org" rel="nofollow">FreeBSD</a> even on my desktop and laptop machines.</p>

<p>It should therefore surprise exactly nobody that I&#39;m now running OS/2, 24/7 or at least part-time, on *checks notes* ...8? machines[<a id="n2"><a href="#fn2" rel="nofollow">2</a></a>].</p>

<h2 id="and-irc">And .. IRC?</h2>

<p>Back in 1996 when I started studying was when I met my first UNIX machines. The computer room at uni was full of Ultrix, HP-UX, Solaris and other things I had heard of but never actually used. I was still using OS/2 privately, but I could not help being fascinated by this new world that opened up to me.</p>

<p>Among the first applications I truly fell in love with was IRC. I don&#39;t even know what the first IRC client I used was, probably ircII or some variant thereof. I soon learned of scripts and such to improve the “user experience”, and before long I was running OpenChat/2 on my OS/2 machines. But since 1996 was already the year of OS/2 Warp 4, I <em>was</em> late to the game: Others had been on IRC for years already, and there were several older IRC applications out there that – at the time – I didn&#39;t know of, nor did I need them.</p>

<p>Fast forward a couple dozen years though, and I find myself running significantly older OS/2 versions – like 2.11 – on machines much older than the PC I had at the time. But since there&#39;s TCP/IP even for this old OS/2 version, I do need to make it “useful”, so off I go looking for IRC clients that work – because my old friend OpenChat/2 requires newer stuff to run.</p>

<h2 id="enter-ezirc">Enter ezIRC!</h2>

<p>At some point in the mid-90s, Roger Bess released ezIRC. I don&#39;t have a good history for this program, as the only <a href="https://web.archive.org/web/19981201213126/http://www.gryn.org/ezirc/" rel="nofollow">archived website</a> seems to have been published around the time of the last known release. But of all the PM-type (GUI) IRC applications I&#39;ve managed to find, it&#39;s about the only one that works on OS/2 2.x and doesn&#39;t look absolutely bonkers.</p>

<p>Problem is .. it&#39;s old. And, when exposed to “modern” IRC servers, UTF-8, protocol extensions and weird server CAPabilities, it gets real confused, real fast. It also does not support server passwords[<a id="n3"><a href="#fn3" rel="nofollow">3</a></a>], so connecting to <a href="https://wiki.znc.in/ZNC" rel="nofollow">ZNC</a> is not actully possible.</p>

<p>After a lot of trial and error (and <code>tcpdump</code>ing and whatnot), plus some help (and only limited amounts of mockery) from the fine people on #znc (Libera), I arrived at a solution\^Wworkaround.</p>

<p>As a bonus, this involves the use of another, only slightly newer, piece of hardware.</p>

<h2 id="the-solution-irssi">The solution: Irssi</h2>

<p>One of the possible solutions I was looking for was some kind of IRC proxy. This however does not really seem to be a thing: I found some variants over the theme, but they were either actual bouncers (already got one, thanks!) or they were solving different sets of problems – and in most cases they were horribly obsolete.</p>

<p>As a side note, my requirement for a solution was that it could only involve software that could run on one of my other retro-rigs; the prime candidate being my trusty dual Pentium Pro/P2OD machine running <a href="/ltning/tag:NetBSD" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">NetBSD</span></a>. Turns out, I already had the necessary software installed!</p>

<p><a href="https://irssi.org/" rel="nofollow">Irssi</a> has a surprisingly useful feature they call <a href="https://raw.githubusercontent.com/irssi/irssi/refs/heads/master/docs/proxy.txt" rel="nofollow">Irssiproxy</a>. Basically this sets up a separate listening port in Irssi itself, which allows connecting clients to share the associated IRC network connection with Irssi.</p>

<p>How perfect is this?!? I can configure the networks i need and associated ZNC connections in Irssi, and have it listen on a set of port numbers on my LAN – and ezIRC can connect with no additional hassle.</p>

<h3 id="except">Except....</h3>

<p>The only real issue I encountered (other than PEBKAC situaitons) was the aforementioned server capability messages which caused ezIRC some confusion. Basically it could join channels and send messages to them, but inbound messages were just dropped on the floor (or, in some cases, shown in the “server messages” window).</p>

<p>I tried connection Irssi directly to the target IRC network (bypassing ZNC), which seemed to solve the problem – but also got in the way of what I tried to achieve.</p>

<p>With some <code>tcpdump</code> and Wireshark work, I found that ZNC adds <code>@time=</code> to the messages. As confirmed by one of the helpful souls on IRC, this is only added if the connecting client asks for it. Luckily the solution is easy: Tell Irssi to skip sending the <code>CAPREQ</code> to ZNC.</p>

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

<p>First, the server config:</p>

<pre><code class="language-c">servers = (
  {
    address = &#34;znc.example.com&#34;;
    chatnet = &#34;os2Floppy&#34;;
    port = &#34;195&#34;;
    password = &#34;&lt;user&gt;@&lt;client-id&gt;/&lt;network&gt;:&lt;password&gt;&#34;;
    use_tls = &#34;yes&#34;;
    tls_verify = &#34;yes&#34;;
    no_cap = &#34;yes&#34;;
  },
  {
    address = &#34;znc.example.com&#34;;
    chatnet = &#34;os2EFNet&#34;;
    port = &#34;195&#34;;
    password = &#34;&lt;user&gt;@&lt;client-id&gt;/&lt;network&gt;:&lt;password&gt;&#34;;
    use_tls = &#34;yes&#34;;
    tls_verify = &#34;yes&#34;;
    no_cap = &#34;yes&#34;;
  },
  {
    address = &#34;znc.example.com&#34;;
    chatnet = &#34;os2Libera&#34;;
    port = &#34;195&#34;;
    password = &#34;&lt;user&gt;@&lt;client-id&gt;/&lt;network&gt;:&lt;password&gt;&#34;;
    use_tls = &#34;yes&#34;;
    tls_verify = &#34;yes&#34;;
    no_cap = &#34;yes&#34;;
  }
);

# No channels configured here, as ZNC takes care of that
channels = ( );
</code></pre>

<p>And then the core config; note the <code>proxy</code> section:</p>

<pre><code class="language-c">settings = {
  core = {
    real_name = &#34;Eirik Overby&#34;;
    user_name = &#34;ltning&#34;;
    nick = &#34;o-Ltning&#34;;
    recode_transliterate = &#34;no&#34;;
  };
  proxy = {
    irssiproxy_bind = &#34;192.88.99.80&#34;; # The local IP
    irssiproxy_ports = &#34;os2Floppy=2777 os2EFNet=2778 os2Libera=2779&#34;; # Per-chatnet listening ports
  };
};

# The chat networks, as used by the server config:
chatnets = {
  os2Floppy = { type = &#34;IRC&#34;; };
  os2EFNet = { type = &#34;IRC&#34;; };
  os2Libera = { type = &#34;IRC&#34;; };
};
</code></pre>

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

<p>And with that, I have a working config. The obligatory screenshot below.</p>

<p>Good luck (if you should end up trying this yourself), and thanks for reading!</p>

<p><img src="https://anduin.net/~ltning/ezirc.jpg" alt="ezIRC running on OS/2 2.11"></p>

<h2 id="footnotes">Footnotes</h2>
<ul><li><a id="fn1"><a href="#n1" rel="nofollow">1</a></a>: I spent many hours pestering IBM Norway into sending me a copy of the beta; I&#39;m sure I ended up on some list somewhere.</li>
<li><a id="fn2"><a href="#n2" rel="nofollow">2</a></a>: The current builds include
<ul><li>A 486DX-50 running Warp 3 and Novell Netware 4.12 paravirtualized</li>
<li>A 486DRx2-50 (really a 386 with a bit of cache) running Warp 3 and my BBS</li>
<li>The subject of this article, a 486DLC-40 running 2.11</li>
<li>A dual Pentium 233 MMX running Warp 4.52</li>
<li>A 286 running OS/2 1.3 EE(!)</li>
<li>A dual Pentium Pro which, when not running NetBSD or DOS runs Warp Server for e-Business</li>
<li>An IBM 486slc2 – Leopard or Leopard LX-based, depending on my mood – running Warp 3</li>
<li>An IBM PS/2 model P70 luggable running 2.11</li></ul></li>
<li><a id="fn3"><a href="#n3" rel="nofollow">3</a></a>: Turns out, it does as of version 1.2o. But given all the other problems, this is mostly of academic interest at this point..</li></ul>

<div style="display: flex; place-content: center;"><table style="width: 80%; border: none; font-size: 0.8em;">
<tr style="border: none;">
<td align="center" style="border: none;" colspan="2"><hr>
~ love over gold ~</td>
</tr>
<tr style="border: none; background-color: inherit;">
<td align="left" style="border: none; padding-left: 2em;"><a href="http://floppy.museum" rel="nofollow">The Floppy Museum</a></td>
<td align="right" style="border: none; padding-right: 2em;"><a href="http://floppy.museum/thispc.htm" rel="nofollow">About the Museum</a></td>
</tr></table></div>
]]></content:encoded>
      <guid>https://scrolls.anduin.net/ltning/os-2-connecting-ezirc-2-to-znc</guid>
      <pubDate>Sat, 17 Oct 2026 16:08:02 +0000</pubDate>
    </item>
    <item>
      <title>A Tale of failure</title>
      <link>https://scrolls.anduin.net/ltning/a-tale-of-failure</link>
      <description>&lt;![CDATA[A while back I was able to free up some space over at the Floppy Museum. Almost 200KB! In #DOS, #286 and #floppy disk terms, that&#39;s quite a bit, and I&#39;ve been scratching my head and soliciting advice for what to with all that space.&#xA;&#xA;A recurring suggestion - and one that resonates well with the retro nature of the museum website - has been to add a guestbook. And as is typical for me, my initial reaction was something along the lines of &#34;how hard can it be?&#34;.&#xA;&#xA;Famous last words.&#xA;&#xA;The problem(s)&#xA;Looking at the server specs, hardware and software come together in perfect harmony to thwart any reasonable approach to adding a web-site guestbook. There&#39;s no way to execute server-side code through the web server, so it was obvious from the start it could not be done in the traditional way.&#xA;&#xA;If you&#39;re truly curious about how the server works, do check out the above links before reading on.&#xA;&#xA;Going old(er) school&#xA;As those who lived and played before the World Wide Web took over everything will know, the Internet is for more than flash(y) web sites and point-and-click (or tap) user interfaces. Not only are those user interfaces very often abominations in their own right, web browsers have turned into more complex creatures than the operating systems they run under...but I digress.&#xA;&#xA;Having used a lot of BBSes in my youth (have you seen this awesmoe documentary? It&#39;s free!), and more recently spun up a Telnet BBS of my own, a semi-obvious idea came to mind: A Telnet-like guestbook service!&#xA;&#xA;Take 1 - rmenu&#xA;I&#39;m not even sure how I came across this tool, but I found it over at VCFed (user wiwa64). This clever little thing listens for telnet-type connections and can be used to view files and run programs on a DOS machine. It includes some not-so-rudimentary attempts at forwarding screen and keyboard activity between client and server, so interactivity is actually possible.&#xA;&#xA;This all worked surprisingly well in DOSBox. I put together a simple menu configuration for rmenu and spent too much time creating a .BAT file that would welcome and interrogate the user, then invoke sed (yes, for DOS) to parse the input and make the resulting HTML somewhat safe and readable:&#xA;&#xA;:HTML&#xA;rem This block is used if I want separate files per post&#xA;rem Uses NDOS/4DOS function to generate a unique file name&#xA;::SET OUTFILE=%@UNIQUE[.]&#xA;::SET TEE=| tee /A&#xA;::SET R=  &#xA;::SET RR=    &#xA;&#xA;rem This appends to an existing guestbook (TEE is just an append-redirect)&#xA;SET OUTFILE=&#xA;SET TEE=    &#xA;SET R=&#xA;SET RR=&#xA;&#xA;rem Output file/table header&#xA;IF NOT X%OUTFILE%==X TYPE guestsh.htm %R %OUTFILE% &#xA;ECHO TR ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34;TD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34; %TEE% guestsb.htm %RR% %OUTFILE%&#xA;&#xA;rem Output previously-created timestamp, then prepare the next table cell&#xA;ECHO %DATE %TEE% guestsb.htm %RR% %OUTFILE%&#xA;ECHO /TDTD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34; %TEE% guestsb.htm %RR% %OUTFILE%&#xA;&#xA;rem Output the name given during the interrogation, then begin the next cell&#xA;ECHO %name% %TEE% guestsb.htm %RR% %OUTFILE%&#xA;ECHO /TDTD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34; %TEE% guestsb.htm %RR% %OUTFILE%&#xA;&#xA;rem Feed the input text through sed(1) to remove HTML tags and convert newlines to br&#xA;sed -e s/[^]  //g -e s/$/br/g TMPFILE.TXT %TEE% guestsb.htm %RR% %OUTFILE%&#xA;&#xA;rem Close out the table row and optionally the whole page&#xA;ECHO /TD/TR %TEE% guestsb.htm %RR% %OUTFILE%&#xA;IF NOT X%OUTFILE%==X ECHO /TABLE/BODY/HTML %RR% %OUTFILE%&#xA;&#xA;rem Clean up, then assemble the final, update guests.htm in the HTML directory&#xA;DEL /Q TMPFILE.TXT&#xA;IF NOT X%OUTFILE%==X REN /Q %OUTFILE% %@NAME[%OUTFILE%].htm&#xA;IF EXIST \FLOPMUSE\NUL TYPE guestsh.htm guestsb.htm guestsf.htm   \FLOPMUSE\guests.htm&#xA;&#xA;GOTO END&#xA;Yes, I know, parsing HTML with regular expressions is considered harmful. And here I was thinking that would be the biggest problem .. but alas.&#xA;&#xA;So what went wrong?&#xA;The real problem with this approach became clear as soon as I deployed it on the actual museum hardware. Whereas DOSBox usually emulates a 32-bit CPU - and of course because I&#39;ve only tested in plain DOS, without DesqVIEW - the intended habitat of this contraption is quite different.&#xA;&#xA;A 286 can multitask. Not easily, not in any kind of pretty fashion - and certainly not real-mode 16-bit DOS programs. There are operating systems that run on a 286 and can multitask reasonably well. Some examples are Microsoft/SCO Xenix, Novell Netware 286 and my personal favourite, OS/2 1.x. Common to all of them is that they run entirely in protected mode, and while they all happily multitask native software, neither of them can do that with DOS programs. As far as I know, OS/2 is the only one of them that can even run DOS programs, and it does so one at a time and switches to real mode to do so.&#xA;&#xA;DesqVIEW can&#39;t multitask for sht on a 286. That is to say .. not without help. In the case of the Floppy Museum, this help comes in the form of the C&amp;% SCAT chipset. It provides hardware-assisted LIM EMS, which lets DesqVIEW map alternative memory regions into the conventional memory area (in the case of this chipset/driver, from 256 to 640KB). Along with clever use of timer interrupts and whatever else it&#39;s doing, it can actually multitask .. somewhat.&#xA;&#xA;Enough, indeed, to run an FTP server, an IRC server, whatever this project will lead to, and of course the web server - and looking pretty darned good while doing so.&#xA;&#xA;What it can&#39;t do, however, is any meaningful kind of virtualization. Specifically, it&#39;s missing what is arguably the biggest and most ground breaking feature of the 386: The virtual 8086 mode. This is a kind of virtual machine mode where DOS programs can run in a virtualized address space, and with all hardware access (I/O ports, etc.) managed by the underlying operating system (which itself is running in protected mode).&#xA;&#xA;How is this relevant for rmenu though? It turns out that the way it manages to pass screen output to the client and keyboard presses to the &#34;server&#34; expects to be all alone on the machine. It will simply return whatever the BIOS has written to the screen, and will only pass input as if it is in sole control of the keyboard buffer.&#xA;&#xA;This doesn&#39;t only affect rmenu, but also DesqVIEW itself any any other application it runs. While programs that only use BIOS screen writes can be controlled reasonably well by DesqVIEW in this situation, any program doing direct screen writes will bypass that entirely. With v86 mode, each application would have had its own virtualized screen, so rmenu could have lived in blissful ignorance thinking it was the only program running on the CPU. As it is, however, anything but the memory regions swapped in and out of EMS is effectively the Wild West: Anything written to the screen will trample all over anything that is already there, no matter which program wrote it.&#xA;&#xA;Similarly, anything trying to figure out what is on the screen will get whatever is there, whenever it shows up. And this is where rmenu gets confused: Timing is messed up by the time slicing DesqVIEW is doing, and since it&#39;s not the only thing writing to the screen, capturing output from the program it&#39;s running is disturbed.&#xA;&#xA;In the end, the experience for the user was absolutely horrendeous: Interactivity was absolutely bonkers, with input and output being horribly delayed and even botched. It worked perfectly in DOSBox, but in meeting the real world^Whardware, it came up short.&#xA;&#xA;Take 2 - an attempt at C, C++, or whatever this is&#xA;&#xA;Realising that attempting to hand screen and keyboard control over to a .BAT file and proxying input and output wasn&#39;t going to work, the obvious alternative seemed to write a standalone TCP/IP-enabled program to receive the input for the guest book.&#xA;&#xA;The mTCP package comes with source code and some sample code to help clueless people like myself get started. With some helpful hints from @mbbrutman@mastodon.sdf.org (the author), I soon had a working OpenWatcom 1.9 development environment. Being the loon I am, I wanted to do my work in DOS - I just had to work around some limitations of the DOS environment (like making compiler invocation command lines shorter by moving include- and header file paths closer) first.&#xA;&#xA;Long and unfinished story a lot shorter: I am absolutely useless at programming on this level. I haven&#39;t touched C or C++ in many decades, so everything I&#39;ve done so far is largely copying, pasting, endless amounts of trial and error, and basically behaving like a monkey who doesn&#39;t even know it&#39;s writing a novel.&#xA;&#xA;I mean .. look at this. I don&#39;t even know what I&#39;m looking at myself:&#xA;while ( !done ) {&#xA;    // Service the connection&#xA;    PACKETPROCESSSINGLE;&#xA;    Arp::driveArp( );&#xA;    Tcp::drivePackets( );&#xA;&#xA;    if ( mySocket-  isRemoteClosed( ) ) {&#xA;        fprintf( stderr, &#34;\nRemote end closed connection\n&#34; );&#xA;        done = 1;&#xA;    } else if ( recvRc   0 ) {&#xA;        msglen = strlen((char ) recvBuf);&#xA;&#xA;        strncpy(msg, (char ) recvBuf, 1);&#xA;&#xA;        switch ( msg[0] ) {&#xA;        case &#39;\n&#39;:&#xA;            if ( addedcret == 0 ) {&#xA;                nlcount++;&#xA;                addedcret = 0;&#xA;                strncpy(entry+strlen(entry), msg, 1);&#xA;                sockprintf(&#34;\r\n&#34;);&#xA;            }&#xA;            break;&#xA;        case &#39;\r&#39;:&#xA;            if ( entry[strlen(entry)-1] != &#39;\n&#39; ) {&#xA;                strncpy(entry+strlen(entry), &#34;\r\n&#34;, 2);&#xA;                nlcount++;&#xA;                nlcount++;&#xA;                addedcret = 1;&#xA;            }&#xA;            sockprintf(&#34;\r\n&#34;);&#xA;            break;&#xA;        case &#39;\b&#39;:&#xA;            if ( entry[strlen(entry)-1] != &#39;\n&#39; &amp;&amp;&#xA;              entry[strlen(entry)-1] != &#39;\n&#39;) {&#xA;                fprintf (stderr, &#34;Deleted last char\n&#34; );&#xA;                entry[strlen(entry)-1] = &#39;\0&#39;;&#xA;                sockprintf(&#34;\b&#34;);&#xA;            }&#xA;            break;&#xA;        case &#39;\127&#39;:&#xA;            if ( entry[strlen(entry)-1] != &#39;\n&#39; &amp;&amp;&#xA;              entry[strlen(entry)-1] != &#39;\n&#39;) {&#xA;                fprintf (stderr, &#34;Deleted last char\n&#34; );&#xA;                entry[strlen(entry)-1] = &#39;\0&#39;;&#xA;                sockprintf(&#34;\b&#34;);&#xA;            }&#xA;            break;&#xA;        case &#39;#&#39;:&#xA;            entry[strlen(entry)-nlcount] = &#39;\0&#39;;&#xA;            recvRc = mySocket-  recv( recvBuf, sizeof(recvBuf) );&#xA;            done = 1;&#xA;            break;&#xA;        default:&#xA;            nlcount = 0;&#xA;            strncpy(entry+strlen(entry), msg, 1);&#xA;            sockprintf(msg);&#xA;        }&#xA;    }&#xA;}&#xA;&#xA;At the moment, the basic functions of the guestbook sort of works. But then all the fun begins: speaking the Telnet protocol, handling special keys and characters, backspaces and the various end-of-line character combinations, and working with all the different telnet and netcat-like clients out there. For some more horror, consider this attempt at instructing the Telnet client about the characteristics of my &#34;server&#34;:&#xA;&#xA;  uint8t wontecho[3] = {255, 252, 1};&#xA;  uint8t willecho[3] = {255, 251, 1};&#xA;  uint8t dontecho[3] = {255, 254, 1};&#xA;  uint8t doecho[3] = {255, 253, 1};&#xA;  uint8t wontline[3] = {255, 252, 34};&#xA;  uint8t willline[3] = {255, 251, 34};&#xA;  uint8t dontline[3] = {255, 254, 34};&#xA;  uint8t doline[3] = {255, 253, 34};&#xA;  uint8t dosuppressgoahead[3] = {255, 253, 3};&#xA;  uint8t willsuppressgoahead[3] = {255, 251, 3};&#xA;  uint8t escE[2] = {27, 69};&#xA;  uint8t resp[64] = &#34;\0&#34;;&#xA;  //sockprintf((char *) dontecho);&#xA;  //mySocket-  send( wontline, sizeof(wontline) );&#xA;&#xA;  mySocket-  send( dontline, sizeof(willsuppressgoahead) );&#xA;  PACKETPROCESSSINGLE;&#xA;  Arp::driveArp( );&#xA;  Tcp::drivePackets( );&#xA;  mySocket-  recv( recvBuf, sizeof(recvBuf) );&#xA;&#xA;  mySocket-  send( dontline, sizeof(dosuppressgoahead) );&#xA;  PACKETPROCESSSINGLE;&#xA;  Arp::driveArp( );&#xA;  Tcp::drivePackets( );&#xA;  mySocket-  recv( recvBuf, sizeof(recvBuf) );&#xA;&#xA;  mySocket-  send( dontline, sizeof(dontline) );&#xA;  PACKETPROCESSSINGLE;&#xA;  Arp::driveArp( );&#xA;  Tcp::drivePackets( );&#xA;  mySocket-  recv( recvBuf, sizeof(recvBuf) );&#xA;Needless to say, it doesn&#39;t (yet) work properly. I guess I&#39;ll keep working on this..&#xA;&#xA;Desperate measures&#xA;&#xA;A final possibility would be to run a second webserver on the machine. However, all the DOS webservers I&#39;ve looked at are either too large or don&#39;t work with packet drivers. They&#39;d need a whole different set of NIC drivers and maybe even a resident TCP/IP stack, all of which is likely to go far beyond the meager memory and storage resources at my disposal.&#xA;&#xA;This option is still on the table, for the next time I have too much free time on my hands.&#xA;&#xA;div style=&#34;display: flex; place-content: center;&#34;table style=&#34;width: 80%; border: none; font-size: 0.8em;&#34;&#xD;&#xA;tr style=&#34;border: none;&#34;&#xD;&#xA;td align=&#34;center&#34; style=&#34;border: none;&#34; colspan=&#34;2&#34;hr&#xD;&#xA;~ love over gold ~/td&#xD;&#xA;/tr&#xD;&#xA;tr style=&#34;border: none; background-color: inherit;&#34;&#xD;&#xA;td align=&#34;left&#34; style=&#34;border: none; padding-left: 2em;&#34;a href=&#34;http://floppy.museum&#34;The Floppy Museum/a/td&#xD;&#xA;td align=&#34;right&#34; style=&#34;border: none; padding-right: 2em;&#34;a href=&#34;http://floppy.museum/thispc.htm&#34;About the Museum/a/td&#xD;&#xA;/tr/table/div]]&gt;</description>
      <content:encoded><![CDATA[<p>A while back I was able to free up some space over at the <a href="http://floppy.museum" rel="nofollow">Floppy Museum</a>. Almost 200KB! In <a href="/ltning/tag:DOS" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">DOS</span></a>, #286 and <a href="/ltning/tag:floppy" class="hashtag" rel="nofollow"><span>#</span><span class="p-category">floppy</span></a> disk terms, that&#39;s quite a bit, and I&#39;ve been scratching my head and <a href="https://pleroma.anduin.net/notice/AsxgzsVFmzDmCpk4kC" rel="nofollow">soliciting advice</a> for what to with all that space.</p>

<p>A recurring suggestion – and one that resonates well with the retro nature of the museum website – has been to add a guestbook. And as is typical for me, my initial reaction was something along the lines of “how hard can it be?”.</p>

<p>Famous last words.</p>

<h2 id="the-problem-s">The problem(s)</h2>

<p>Looking at <a href="http://floppy.museum/thispc.htm" rel="nofollow">the server specs</a>, <a href="http://floppy.museum/thispc.htm#hardware" rel="nofollow">hardware</a> and <a href="http://floppy.museum/thispc.htm#software" rel="nofollow">software</a> come together in perfect harmony to thwart any reasonable approach to adding a web-site guestbook. There&#39;s no way to execute server-side code through the web server, so it was obvious from the start it could not be done in the traditional way.</p>

<p>If you&#39;re truly curious about how the server works, do check out the above links before reading on.</p>

<h2 id="going-old-er-school">Going old(er) school</h2>

<p>As those who lived and played before the World Wide Web took over everything will know, the Internet is for more than flash(y) web sites and point-and-click (or tap) user interfaces. Not only are those user interfaces very often abominations in their own right, web browsers have turned into more complex creatures than the operating systems they run under...but I digress.</p>

<p>Having used a lot of <a href="https://en.wikipedia.org/wiki/Bulletin_board_system" rel="nofollow">BBSes</a> in my youth (have you seen <a href="http://www.bbsdocumentary.com/" rel="nofollow">this awesmoe documentary</a>? It&#39;s free!), and more recently spun up a <a href="http://bbs.anduin.net/bbs.htm" rel="nofollow">Telnet BBS</a> of my own, a semi-obvious idea came to mind: A Telnet-like guestbook service!</p>

<h2 id="take-1-rmenu">Take 1 – <code>rmenu</code></h2>

<p>I&#39;m not even sure how I came across this tool, but I found it over at <a href="https://forum.vcfed.org/index.php?threads/remote-control-of-dos-box.14324/" rel="nofollow">VCFed (user wiwa64)</a>. This clever little thing listens for telnet-type connections and can be used to view files and run programs on a DOS machine. It includes some not-so-rudimentary attempts at forwarding screen and keyboard activity between client and server, so interactivity is actually possible.</p>

<p>This all worked surprisingly well in DOSBox. I put together a simple menu configuration for <code>rmenu</code> and spent too much time creating a <code>.BAT</code> file that would welcome and interrogate the user, then invoke <a href="https://en.wikipedia.org/wiki/Sed" rel="nofollow"><code>sed</code></a> (yes, for DOS) to parse the input and make the resulting HTML somewhat safe and readable:</p>

<pre><code class="language-dos">:HTML
rem This block is used if I want separate files per post
rem Uses NDOS/4DOS function to generate a unique file name
::SET OUTFILE=%@UNIQUE[.]
::SET _TEE=`| tee /A`
::SET _R=`&gt;`
::SET _RR=`&gt;&gt;`

rem This appends to an existing guestbook (_TEE is just an append-redirect)
SET OUTFILE=
SET _TEE=`&gt;&gt;`
SET _R=
SET _RR=

rem Output file/table header
IF NOT X%OUTFILE%==X TYPE guests_h.htm %_R %OUTFILE% 
ECHO `&lt;TR ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34;&gt;&lt;TD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34;&gt;` %_TEE% guests_b.htm %_RR% %OUTFILE%

rem Output previously-created timestamp, then prepare the next table cell
ECHO %_DATE %_TEE% guests_b.htm %_RR% %OUTFILE%
ECHO `&lt;/TD&gt;&lt;TD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34;&gt;` %_TEE% guests_b.htm %_RR% %OUTFILE%

rem Output the name given during the interrogation, then begin the next cell
ECHO %name% %_TEE% guests_b.htm %_RR% %OUTFILE%
ECHO `&lt;/TD&gt;&lt;TD ALIGN=&#34;LEFT&#34; VALIGN=&#34;TOP&#34;&gt;` %_TEE% guests_b.htm %_RR% %OUTFILE%

rem Feed the input text through sed(1) to remove HTML tags and convert newlines to &lt;br&gt;
sed -e `s/&lt;[^&gt;]*&gt;//g` -e `s/$/&lt;br&gt;/g` TMPFILE.TXT %_TEE% guests_b.htm %_RR% %OUTFILE%

rem Close out the table row and optionally the whole page
ECHO `&lt;/TD&gt;&lt;/TR&gt;` %_TEE% guests_b.htm %_RR% %OUTFILE%
IF NOT X%OUTFILE%==X ECHO `&lt;/TABLE&gt;&lt;/BODY&gt;&lt;/HTML&gt;` %_RR% %OUTFILE%

rem Clean up, then assemble the final, update guests.htm in the HTML directory
DEL /Q TMPFILE.TXT
IF NOT X%OUTFILE%==X REN /Q %OUTFILE% %@NAME[%OUTFILE%].htm
IF EXIST \FLOPMUSE\NUL TYPE guests_h.htm guests_b.htm guests_f.htm &gt; \FLOPMUSE\guests.htm

GOTO END
</code></pre>

<p>Yes, I know, <a href="https://stackoverflow.com/a/1732454" rel="nofollow">parsing HTML with regular expressions is considered harmful</a>. And here I was thinking that would be the biggest problem .. but alas.</p>

<h3 id="so-what-went-wrong">So what went wrong?</h3>

<p>The real problem with this approach became clear as soon as I deployed it on the actual museum hardware. Whereas DOSBox usually emulates a 32-bit CPU – and of course because I&#39;ve only tested in plain DOS, without DesqVIEW – the intended habitat of this contraption is quite different.</p>

<p>A 286 can multitask. Not easily, not in any kind of pretty fashion – and certainly not real-mode 16-bit DOS programs. There are operating systems that run on a 286 and can multitask reasonably well. Some examples are Microsoft/SCO Xenix, Novell Netware 286 and my personal favourite, OS/2 1.x. Common to all of them is that they run entirely in <a href="https://en.wikipedia.org/wiki/Protected_mode" rel="nofollow">protected mode</a>, and while they all happily multitask native software, neither of them can do that with DOS programs. As far as I know, OS/2 is the only one of them that can even <em>run</em> DOS programs, and it does so one at a time and switches to real mode to do so.</p>

<p>DesqVIEW can&#39;t multitask for sh*t on a 286. That is to say .. not without help. In the case of the Floppy Museum, this help comes in the form of the C&amp;% SCAT chipset. It provides hardware-assisted <a href="https://en.wikipedia.org/wiki/Expanded_memory" rel="nofollow">LIM EMS</a>, which lets DesqVIEW map alternative memory regions into the conventional memory area (in the case of this chipset/driver, from 256 to 640KB). Along with clever use of timer interrupts and whatever else it&#39;s doing, it <em>can</em> actually multitask .. somewhat.</p>

<p>Enough, indeed, to run an FTP server, an IRC server, whatever this project will lead to, and of course the web server – and <a href="https://bbs.anduin.net/img/floppymuseum_med.jpg" rel="nofollow">looking pretty darned good</a> while doing so.</p>

<p>What it can&#39;t do, however, is any meaningful kind of virtualization. Specifically, it&#39;s missing what is arguably the biggest and most ground breaking feature of the 386: The <a href="https://en.wikipedia.org/wiki/Virtual_8086_mode" rel="nofollow">virtual 8086 mode</a>. This is a kind of virtual machine mode where DOS programs can run in a virtualized address space, and with all hardware access (I/O ports, etc.) managed by the underlying operating system (which itself is running in protected mode).</p>

<p>How is this relevant for <code>rmenu</code> though? It turns out that the way it manages to pass screen output to the client and keyboard presses to the “server” expects to be all alone on the machine. It will simply return whatever the BIOS has written to the screen, and will only pass input as if it is in sole control of the keyboard buffer.</p>

<p>This doesn&#39;t only affect <code>rmenu</code>, but also DesqVIEW itself any any other application it runs. While programs that only use BIOS screen writes can be controlled reasonably well by DesqVIEW in this situation, any program doing direct screen writes will bypass that entirely. With v86 mode, each application would have had its own virtualized screen, so <code>rmenu</code> could have lived in blissful ignorance thinking it was the only program running on the CPU. As it is, however, anything but the memory regions swapped in and out of EMS is effectively the Wild West: Anything written to the screen will trample all over anything that is already there, no matter which program wrote it.</p>

<p>Similarly, anything trying to figure out what <em>is</em> on the screen will get whatever is there, whenever it shows up. And this is where <code>rmenu</code> gets confused: Timing is messed up by the time slicing DesqVIEW is doing, and since it&#39;s not the only thing writing to the screen, capturing output from the program it&#39;s running is disturbed.</p>

<p>In the end, the experience for the user was absolutely horrendeous: Interactivity was absolutely bonkers, with input and output being horribly delayed and even botched. It worked perfectly in DOSBox, but in meeting the real world^Whardware, it came up short.</p>

<h2 id="take-2-an-attempt-at-c-c-or-whatever-this-is">Take 2 – an attempt at C, C++, or whatever this is</h2>

<p>Realising that attempting to hand screen and keyboard control over to a .BAT file and proxying input and output wasn&#39;t going to work, the obvious alternative seemed to write a standalone TCP/IP-enabled program to receive the input for the guest book.</p>

<p>The <a href="https://www.brutman.com/mTCP/mTCP.html" rel="nofollow">mTCP</a> package comes with source code and some sample code to help clueless people like myself get started. With some helpful hints from <a href="https://scrolls.anduin.net/@/mbbrutman@mastodon.sdf.org" class="u-url mention" rel="nofollow">@<span>mbbrutman@mastodon.sdf.org</span></a> (the author), I soon had a working <a href="https://www.openwatcom.org/" rel="nofollow">OpenWatcom 1.9</a> development environment. Being the loon I am, I wanted to do my work in DOS – I just had to work around some limitations of the DOS environment (like making compiler invocation command lines shorter by moving include- and header file paths closer) first.</p>

<p>Long and unfinished story a lot shorter: I am absolutely useless at programming on this level. I haven&#39;t touched C <em>or</em> C++ in many decades, so everything I&#39;ve done so far is largely copying, pasting, endless amounts of trial and error, and basically behaving like a monkey who doesn&#39;t even know it&#39;s writing a novel.</p>

<p>I mean .. look at this. I don&#39;t even know what I&#39;m looking at myself:</p>

<pre><code class="language-c">while ( !done ) {
    // Service the connection
    PACKET_PROCESS_SINGLE;
    Arp::driveArp( );
    Tcp::drivePackets( );

    if ( mySocket-&gt;isRemoteClosed( ) ) {
        fprintf( stderr, &#34;\nRemote end closed connection\n&#34; );
        done = 1;
    } else if ( recvRc &gt; 0 ) {
        msglen = strlen((char *) recvBuf);

        strncpy(msg, (char *) recvBuf, 1);

        switch ( msg[0] ) {
        case &#39;\n&#39;:
            if ( added_cret == 0 ) {
                nlcount++;
                added_cret = 0;
                strncpy(entry+strlen(entry), msg, 1);
                sock_printf(&#34;\r\n&#34;);
            }
            break;
        case &#39;\r&#39;:
            if ( entry[strlen(entry)-1] != &#39;\n&#39; ) {
                strncpy(entry+strlen(entry), &#34;\r\n&#34;, 2);
                nlcount++;
                nlcount++;
                added_cret = 1;
            }
            sock_printf(&#34;\r\n&#34;);
            break;
        case &#39;\b&#39;:
            if ( entry[strlen(entry)-1] != &#39;\n&#39; &amp;&amp;
              entry[strlen(entry)-1] != &#39;\n&#39;) {
                fprintf (stderr, &#34;Deleted last char\n&#34; );
                entry[strlen(entry)-1] = &#39;\0&#39;;
                sock_printf(&#34;\b&#34;);
            }
            break;
        case &#39;\127&#39;:
            if ( entry[strlen(entry)-1] != &#39;\n&#39; &amp;&amp;
              entry[strlen(entry)-1] != &#39;\n&#39;) {
                fprintf (stderr, &#34;Deleted last char\n&#34; );
                entry[strlen(entry)-1] = &#39;\0&#39;;
                sock_printf(&#34;\b&#34;);
            }
            break;
        case &#39;#&#39;:
            entry[strlen(entry)-nlcount] = &#39;\0&#39;;
            recvRc = mySocket-&gt;recv( recvBuf, sizeof(recvBuf) );
            done = 1;
            break;
        default:
            nlcount = 0;
            strncpy(entry+strlen(entry), msg, 1);
            sock_printf(msg);
        }
    }
}
</code></pre>

<p>At the moment, the basic functions of the guestbook sort of works. But then all the fun begins: speaking the Telnet protocol, handling special keys and characters, backspaces and the various end-of-line character combinations, and working with all the different telnet and netcat-like clients out there. For some more horror, consider this attempt at instructing the Telnet client about the characteristics of my “server”:</p>

<pre><code class="language-c">  uint8_t wont_echo[3] = {255, 252, 1};
  uint8_t will_echo[3] = {255, 251, 1};
  uint8_t dont_echo[3] = {255, 254, 1};
  uint8_t do_echo[3] = {255, 253, 1};
  uint8_t wont_line[3] = {255, 252, 34};
  uint8_t will_line[3] = {255, 251, 34};
  uint8_t dont_line[3] = {255, 254, 34};
  uint8_t do_line[3] = {255, 253, 34};
  uint8_t do_suppress_goahead[3] = {255, 253, 3};
  uint8_t will_suppress_goahead[3] = {255, 251, 3};
  uint8_t escE[2] = {27, 69};
  uint8_t resp[64] = &#34;\0&#34;;
  //sock_printf((char *) dont_echo);
  //mySocket-&gt;send( wont_line, sizeof(wont_line) );

  mySocket-&gt;send( dont_line, sizeof(will_suppress_goahead) );
  PACKET_PROCESS_SINGLE;
  Arp::driveArp( );
  Tcp::drivePackets( );
  mySocket-&gt;recv( recvBuf, sizeof(recvBuf) );

  mySocket-&gt;send( dont_line, sizeof(do_suppress_goahead) );
  PACKET_PROCESS_SINGLE;
  Arp::driveArp( );
  Tcp::drivePackets( );
  mySocket-&gt;recv( recvBuf, sizeof(recvBuf) );

  mySocket-&gt;send( dont_line, sizeof(dont_line) );
  PACKET_PROCESS_SINGLE;
  Arp::driveArp( );
  Tcp::drivePackets( );
  mySocket-&gt;recv( recvBuf, sizeof(recvBuf) );
</code></pre>

<p>Needless to say, it doesn&#39;t (yet) work properly. I guess I&#39;ll keep working on this..</p>

<h2 id="desperate-measures">Desperate measures</h2>

<p>A final possibility would be to run a second webserver on the machine. However, all the DOS webservers I&#39;ve looked at are either too large or don&#39;t work with packet drivers. They&#39;d need a whole different set of NIC drivers and maybe even a resident TCP/IP stack, all of which is likely to go far beyond the meager memory and storage resources at my disposal.</p>

<p>This option is still on the table, for the next time I have too much free time on my hands.</p>

<div style="display: flex; place-content: center;"><table style="width: 80%; border: none; font-size: 0.8em;">
<tr style="border: none;">
<td align="center" style="border: none;" colspan="2"><hr>
~ love over gold ~</td>
</tr>
<tr style="border: none; background-color: inherit;">
<td align="left" style="border: none; padding-left: 2em;"><a href="http://floppy.museum" rel="nofollow">The Floppy Museum</a></td>
<td align="right" style="border: none; padding-right: 2em;"><a href="http://floppy.museum/thispc.htm" rel="nofollow">About the Museum</a></td>
</tr></table></div>
]]></content:encoded>
      <guid>https://scrolls.anduin.net/ltning/a-tale-of-failure</guid>
      <pubDate>Sat, 03 May 2025 18:58:34 +0000</pubDate>
    </item>
    <item>
      <title>What mattered to me</title>
      <link>https://scrolls.anduin.net/ltning/what-mattered-to-me</link>
      <description>&lt;![CDATA[It used to matter to me that the big fish did not eat all the small fish.&#xA;This is still true.&#xA;&#xA;It used to matter to me that technology won out because it was better, more fit for purpose, and/or enabled, invited and inspired new ways to think.&#xA;This is still true.&#xA;&#xA;It used to matter to me that people made informed choices for themselves, and were allowed to do so by their peers.&#xA;This is still true.&#xA;&#xA;None of this came to pass.&#xA;&#xA;div style=&#34;display: flex; place-content: center;&#34;table style=&#34;width: 80%; border: none; font-size: 0.8em;&#34;&#xD;&#xA;tr style=&#34;border: none;&#34;&#xD;&#xA;td align=&#34;center&#34; style=&#34;border: none;&#34; colspan=&#34;2&#34;hr&#xD;&#xA;~ love over gold ~/td&#xD;&#xA;/tr&#xD;&#xA;tr style=&#34;border: none; background-color: inherit;&#34;&#xD;&#xA;td align=&#34;left&#34; style=&#34;border: none; padding-left: 2em;&#34;a href=&#34;http://floppy.museum&#34;The Floppy Museum/a/td&#xD;&#xA;td align=&#34;right&#34; style=&#34;border: none; padding-right: 2em;&#34;a href=&#34;http://floppy.museum/thispc.htm&#34;About the Museum/a/td&#xD;&#xA;/tr/table/div]]&gt;</description>
      <content:encoded><![CDATA[<p>It used to matter to me that the big fish did not eat all the small fish.
This is still true.</p>

<p>It used to matter to me that technology won out because it was better, more fit for purpose, and/or enabled, invited and inspired new ways to think.
This is still true.</p>

<p>It used to matter to me that people made informed choices for themselves, and were allowed to do so by their peers.
This is still true.</p>

<p>None of this came to pass.</p>

<div style="display: flex; place-content: center;"><table style="width: 80%; border: none; font-size: 0.8em;">
<tr style="border: none;">
<td align="center" style="border: none;" colspan="2"><hr>
~ love over gold ~</td>
</tr>
<tr style="border: none; background-color: inherit;">
<td align="left" style="border: none; padding-left: 2em;"><a href="http://floppy.museum" rel="nofollow">The Floppy Museum</a></td>
<td align="right" style="border: none; padding-right: 2em;"><a href="http://floppy.museum/thispc.htm" rel="nofollow">About the Museum</a></td>
</tr></table></div>
]]></content:encoded>
      <guid>https://scrolls.anduin.net/ltning/what-mattered-to-me</guid>
      <pubDate>Tue, 22 Apr 2025 21:16:22 +0000</pubDate>
    </item>
  </channel>
</rss>