fix many internal links

Fixed a number of internal links that need to go to the GitHub site. Ran `make www` to perform the above.
2024-04-24 02:37:42 -07:00
1 parent b84f8cbe28
commit e7d69bc232
@@ -71,8 +71,8 @@ to:
    for(signal=0;*k *x * __FILE__ *i;) do {
 ```
-To see what we mean look at the [original source file](decot.orig.c). The
+To see what we mean look at the [original source file](%%REPO_URL%%/1984/decot/decot.orig.c). The
-[alternate code](decot.alt.c) is the version that has this modification. The
+[alternate code](%%REPO_URL%%/1984/decot/decot.alt.c) is the version that has this modification. The
 fixed version has instead:
 ``` <!---c-->
@@ -406,8 +406,8 @@ changed it from:</p>
 <pre><code>    for(signal=0;*k * x * __FILE__ *i;) do {</code></pre>
 <p>to:</p>
 <pre><code>    for(signal=0;*k *x * __FILE__ *i;) do {</code></pre>
-<p>To see what we mean look at the <a href="decot.orig.c">original source file</a>. The
+<p>To see what we mean look at the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1984/decot/decot.orig.c">original source file</a>. The
-<a href="decot.alt.c">alternate code</a> is the version that has this modification. The
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1984/decot/decot.alt.c">alternate code</a> is the version that has this modification. The
 fixed version has instead:</p>
 <pre><code>    for(signal=0;*k *= * __FILE__ *i;) do {</code></pre>
 <h2 id="authors-remarks">Author’s remarks:</h2>
@@ -125,8 +125,8 @@ favorite entries!
 In 2023 remarks were discovered from [Sjoerd
 Mullender](../../authors.html#Sjoerd_Mullender), one of the authors, and so was the
 program that was used by the authors to generate the array that he referred to.
-Because [a.out.h](a.out.h), which [gentab.c](gentab.c) uses, is not available in
+Because [a.out.h](%%REPO_URL%%/1984/mullender/a.out.h), which [gentab.c](%%REPO_URL%%/1984/mullender/gentab.c) uses,
-all systems (like macOS) and more importantly because we wanted it to be as
+is not available in all systems (like macOS) and more importantly because we wanted it to be as
 close to as the original as possible we used a copy of
 <https://raw.githubusercontent.com/dspinellis/unix-history-repo/Research-Release/usr/include/a.out.h>
 in the *fabulous* [Unix History
@@ -386,7 +386,7 @@ format randomly. So we wrote a program to choose an appropriate format at
 random. The program we wrote for that follows.  This program actually also
 extracted the machine code from the object file.
-NOTE from judges: see [gentab.c](gentab.c) for a copy of this file that can be
+NOTE from judges: see [gentab.c](%%REPO_URL%%/1984/mullender/gentab.c) for a copy of this file that can be
 compiled in modern systems.
 ``` <!---c-->
@@ -456,8 +456,8 @@ favorite entries!</p>
 <p>In 2023 remarks were discovered from <a href="../../authors.html#Sjoerd_Mullender">Sjoerd
 Mullender</a>, one of the authors, and so was the
 program that was used by the authors to generate the array that he referred to.
-Because <a href="a.out.h">a.out.h</a>, which <a href="gentab.c">gentab.c</a> uses, is not available in
+Because <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1984/mullender/a.out.h">a.out.h</a>, which <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1984/mullender/gentab.c">gentab.c</a> uses,
-all systems (like macOS) and more importantly because we wanted it to be as
+is not available in all systems (like macOS) and more importantly because we wanted it to be as
 close to as the original as possible we used a copy of
 <a href="https://raw.githubusercontent.com/dspinellis/unix-history-repo/Research-Release/usr/include/a.out.h" class="uri">https://raw.githubusercontent.com/dspinellis/unix-history-repo/Research-Release/usr/include/a.out.h</a>
 in the <em>fabulous</em> <a href="https://github.com/dspinellis/unix-history-repo/tree/Research-Release">Unix History
@@ -661,7 +661,7 @@ formats for each constant in the resulting array, and we wanted to choose the
 format randomly. So we wrote a program to choose an appropriate format at
 random. The program we wrote for that follows. This program actually also
 extracted the machine code from the object file.</p>
-<p>NOTE from judges: see <a href="gentab.c">gentab.c</a> for a copy of this file that can be
+<p>NOTE from judges: see <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1984/mullender/gentab.c">gentab.c</a> for a copy of this file that can be
 compiled in modern systems.</p>
 <pre><code>    #include &lt;stdio.h&gt;
    #include &lt;a.out.h&gt;
@@ -27,7 +27,7 @@ if nothing else.
 Due to the different [conflicting problems](compilers.html) with gcc and clang, we
 instead offer the problematic code as an alternate version whereas
-[marshall.c](marshall.c) has both the infinite loop and the complicated arg to
+[marshall.c](%%REPO_URL%%/1986/marshall/marshall.c) has both the infinite loop and the complicated arg to
 `_exit()` commented out, changing the value passed into `_exit()` to `1`.
@@ -393,7 +393,7 @@ if nothing else.</p>
 <h2 id="alternate-code">Alternate code:</h2>
 <p>Due to the different <a href="compilers.html">conflicting problems</a> with gcc and clang, we
 instead offer the problematic code as an alternate version whereas
-<a href="marshall.c">marshall.c</a> has both the infinite loop and the complicated arg to
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1986/marshall/marshall.c">marshall.c</a> has both the infinite loop and the complicated arg to
 <code>_exit()</code> commented out, changing the value passed into <code>_exit()</code> to <code>1</code>.</p>
 <h3 id="alternate-build">Alternate build:</h3>
 <p>To see if your compiler has the problems noted:</p>
@@ -8,15 +8,15 @@ You may then wish to look at the Author's remarks for even more details.
 Instructions for use: Run `make` to compile entries (it is possible that on
 System V or non-Unix systems the Makefile needs to be changed).
-The Makefile for ["Best self modifying program"](fubar/indx.html) always uses
+The Makefile for ["Best self modifying program"](fubar/index.html) always uses
 the most portable version because there is no loss of functionality in
-using it.  In the case of the ["Best game"](tromp/README) entry, however, some
+using it.  In the case of the ["Best game"](tromp/index.html) entry, however, some
 functionality is lost in the portable version and so its Makefile uses
 the original program.
 This year, the [Best of Show](jar.2/index.html) was given to the most useful program.
-The ["Strangest abuse of the rules"](jar.1/indx.html) award was given this year
+The ["Strangest abuse of the rules"](jar.1/index.html) award was given this year
 to stress the fact that starting in 1990, compiling entries must result a
 regular executable file.
@@ -369,13 +369,13 @@ Look at the winning source and try to figure how it does what it does!
 You may then wish to look at the Author’s remarks for even more details.</p>
 <p>Instructions for use: Run <code>make</code> to compile entries (it is possible that on
 System V or non-Unix systems the Makefile needs to be changed).</p>
-<p>The Makefile for <a href="fubar/indx.html">“Best self modifying program”</a> always uses
+<p>The Makefile for <a href="fubar/index.html">“Best self modifying program”</a> always uses
 the most portable version because there is no loss of functionality in
-using it. In the case of the <a href="tromp/README">“Best game”</a> entry, however, some
+using it. In the case of the <a href="tromp/index.html">“Best game”</a> entry, however, some
 functionality is lost in the portable version and so its Makefile uses
 the original program.</p>
 <p>This year, the <a href="jar.2/index.html">Best of Show</a> was given to the most useful program.</p>
-<p>The <a href="jar.1/indx.html">“Strangest abuse of the rules”</a> award was given this year
+<p>The <a href="jar.1/index.html">“Strangest abuse of the rules”</a> award was given this year
 to stress the fact that starting in 1990, compiling entries must result a
 regular executable file.</p>
 <p><a href="rules.txt">Rules</a> and <a href="../years.html#1989">results</a> were posted to
@@ -92,7 +92,7 @@ of the program which we include below:
    e(x){x<0?write(2,"?\n$ "-x/4,2),x+1||(exit(1),0):5;}
 ```
-and in the [alternate source](tbr.alt.c). It has the `exit()` returns `void` fix
+and in the [alternate source](%%REPO_URL%%/1990/tbr/tbr.alt.c). It has the `exit()` returns `void` fix
 as well as the `gets()` to `fgets()` fix applied to it to make it functionally
 equivalent like the authors intended.
@@ -436,7 +436,7 @@ of the program which we include below:</p>
    0),e(open(*i,0)):5,t?dup2(t,1)D t):i[2]?9 D
    1),e(creat(i[2],438)):2,e(execvp(*u,u))):e(chdir(u[1])*2):6;}
    e(x){x&lt;0?write(2,&quot;?\n$ &quot;-x/4,2),x+1||(exit(1),0):5;}</code></pre>
-<p>and in the <a href="tbr.alt.c">alternate source</a>. It has the <code>exit()</code> returns <code>void</code> fix
+<p>and in the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1990/tbr/tbr.alt.c">alternate source</a>. It has the <code>exit()</code> returns <code>void</code> fix
 as well as the <code>gets()</code> to <code>fgets()</code> fix applied to it to make it functionally
 equivalent like the authors intended.</p>
 <h3 id="alternate-build">Alternate build:</h3>
@@ -61,7 +61,7 @@ game:
 ```
 The shell script determines when the game is over, and automatically replaces
-`merlyn.c` (copied from [westley.c](westley.c) with an improved version after a
+`merlyn.c` (copied from [westley.c](%%REPO_URL%%/1991/westley/westley.c) with an improved version after a
 loss).  The [ttt](%%REPO_URL%%/1991/westley/ttt.sh) script has two other modes:
 ``` <!---sh-->
@@ -406,7 +406,7 @@ game:</p>
    ./ttt [first_move]  # &lt;- omit first arg to have it move first
    ./ttt next_move</code></pre>
 <p>The shell script determines when the game is over, and automatically replaces
-<code>merlyn.c</code> (copied from <a href="westley.c">westley.c</a> with an improved version after a
+<code>merlyn.c</code> (copied from <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1991/westley/westley.c">westley.c</a> with an improved version after a
 loss). The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1991/westley/ttt.sh">ttt</a> script has two other modes:</p>
 <pre><code>    ./ttt quitgame      # cancel any game in progress, revert to merlyn.c
    ./ttt clobber       # remove all game files, revert to original source</code></pre>
@@ -57,7 +57,7 @@ judges this Email message:
 ```
 The [albert.alt.c](%%REPO_URL%%/1992/albert/albert.c) file is the fixed file, whereas the
-[albert.c](1992/albert/1992/albert/lbert.c) is the file before applying Leo Broukhis' fix.
+[albert.c](%%REPO_URL%%/1992/albert/1992/albert/lbert.c) is the file before applying Leo Broukhis' fix.
 ### Alternate build:
@@ -407,7 +407,7 @@ judges this Email message:</p>
    Both albert and albert.orig loop without printing anything, although
    the first factor is 101 and is usually found in an instant.</code></pre>
 <p>The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/albert/albert.c">albert.alt.c</a> file is the fixed file, whereas the
-<a href="1992/albert/1992/albert/lbert.c">albert.c</a> is the file before applying Leo Broukhis’ fix.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/albert/1992/albert/lbert.c">albert.c</a> is the file before applying Leo Broukhis’ fix.</p>
 <h3 id="alternate-build">Alternate build:</h3>
 <p>To compile this alternate version:</p>
 <pre><code>    make alt</code></pre>
@@ -58,7 +58,7 @@ Then using the proper dictionary file it does:
 where `/usr/share/dict/words` is the dictionary file.
 Then it uses the [mkdict.sh](%%REPO_URL%%/1992/gson/mkdict.sh) script to create a dictionary file out
-of the files [README.md](%%REPO_URL%%/1992/gson/README.md), [try.sh](try.sh) (itself) and
+of the files [index.html](%%REPO_URL%%/1992/gson/index.html), [try.sh](%%REPO_URL%%/1992/gson/try.sh) (itself) and
 [Makefile](%%REPO_URL%%/1992/gson/Makefile) and it repeats the same commands as above. In the case no
 dictionary file can be found in the first step it only runs the commands once
 with the created dictionary file.
@@ -406,7 +406,7 @@ following locations though the last one is more ironic:</p>
    ./ag paul e tsongas         &lt; /usr/share/dict/words</code></pre>
 <p>where <code>/usr/share/dict/words</code> is the dictionary file.</p>
 <p>Then it uses the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/gson/mkdict.sh">mkdict.sh</a> script to create a dictionary file out
-of the files <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/gson/README.md">README.md</a>, <a href="try.sh">try.sh</a> (itself) and
+of the files <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/gson/index.html">index.html</a>, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/gson/try.sh">try.sh</a> (itself) and
 <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/gson/Makefile">Makefile</a> and it repeats the same commands as above. In the case no
 dictionary file can be found in the first step it only runs the commands once
 with the created dictionary file.</p>
@@ -77,7 +77,7 @@ you have with the program.  :-)
 NOTE: Some systems do not need (or have) `-lsocket -lresolv -lnsl`.
 Such systems should compile using `-lX11`.
-NOTE: The original winning source [kivinen.orig.c](kivinen.orig.c) assumed that
+NOTE: The original winning source [kivinen.orig.c](%%REPO_URL%%/1992/kivinen/kivinen.orig.c) assumed that
 `exit(3)` returned a value which cause problems for some systems where `exit(3)` returns
 `void`.  The file [kivinen.c](%%REPO_URL%%/1992/kivinen/kivinen.c) avoids this problem.
@@ -414,7 +414,7 @@ code.</p>
 you have with the program. :-)</p>
 <p>NOTE: Some systems do not need (or have) <code>-lsocket -lresolv -lnsl</code>.
 Such systems should compile using <code>-lX11</code>.</p>
-<p>NOTE: The original winning source <a href="kivinen.orig.c">kivinen.orig.c</a> assumed that
+<p>NOTE: The original winning source <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/kivinen/kivinen.orig.c">kivinen.orig.c</a> assumed that
 <code>exit(3)</code> returned a value which cause problems for some systems where <code>exit(3)</code> returns
 <code>void</code>. The file <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/kivinen/kivinen.c">kivinen.c</a> avoids this problem.</p>
 <h2 id="authors-remarks">Author’s remarks:</h2>
@@ -22,7 +22,7 @@ NOTE: Some compilers have had trouble optimizing this entry.
 NOTE: Some systems need to compile with `-ltermcap` as well as `-lcurses`.
-NOTE: The original winning source [marangon.orig.c](marangon.orig.c), found in
+NOTE: The original winning source [marangon.orig.c](%%REPO_URL%%/1992/marangon/marangon.orig.c), found in
 the archive directory, assumed that `exit(3)` returns a value which cause
 problems for some systems where `exit(3)` returns a `void`.  The file
 [marangon.c](%%REPO_URL%%/1992/marangon/marangon.c) avoids this problem.
@@ -387,7 +387,7 @@ numbers as needed. It is the side effects that get you into trouble. If you
 are not careful, you may find things “at 6’s and 7’s”. :-)</p>
 <p>NOTE: Some compilers have had trouble optimizing this entry.</p>
 <p>NOTE: Some systems need to compile with <code>-ltermcap</code> as well as <code>-lcurses</code>.</p>
-<p>NOTE: The original winning source <a href="marangon.orig.c">marangon.orig.c</a>, found in
+<p>NOTE: The original winning source <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/marangon/marangon.orig.c">marangon.orig.c</a>, found in
 the archive directory, assumed that <code>exit(3)</code> returns a value which cause
 problems for some systems where <code>exit(3)</code> returns a <code>void</code>. The file
 <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1992/marangon/marangon.c">marangon.c</a> avoids this problem.</p>
@@ -49,7 +49,7 @@ and
 to another value like 18 or 19 which you can do with the alternate code.
-In the case of [rince.alt2.c](rince.alt2.c) the movement keys for the default
+In the case of [rince.alt2.c](%%REPO_URL%%/1993/rince/rince.alt2.c) the movement keys for the default
 game are slightly different and will feel more familiar to vi users: `h` for
 left, `l` for right (that was not changed) and `k` for fire (that was also not
 changed).
@@ -397,7 +397,7 @@ value 17 in the lines:</p>
 <p>and</p>
 <pre><code>    refresh(),c=select(k,&amp;y,0,0,(v.tv_usec=1&lt;&lt;17,&amp;v))?getch():0;</code></pre>
 <p>to another value like 18 or 19 which you can do with the alternate code.</p>
-<p>In the case of <a href="rince.alt2.c">rince.alt2.c</a> the movement keys for the default
+<p>In the case of <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1993/rince/rince.alt2.c">rince.alt2.c</a> the movement keys for the default
 game are slightly different and will feel more familiar to vi users: <code>h</code> for
 left, <code>l</code> for right (that was not changed) and <code>k</code> for fire (that was also not
 changed).</p>
@@ -84,8 +84,8 @@ in an effort to get around the 1994 version of Rule 2:
       followed by either whitespace or end of file, must be <= 1536.
 ```
-Indeed, when [dodsond2.gz](dodsond2.gz) is uncompressed, it is identical
+Indeed the [dodsond2.c](%%REPO_URL%%/1994/dodsond2/dodsond2.c) file is
-to the [dodsond2.c](%%REPO_URL%%/1994/dodsond2/dodsond2.c) file you find today.
+what you find today.
 The IOCCC rules and guidelines were modified in an effort to address
 this fun "_abuse of the rules_" style hack.
@@ -414,8 +414,8 @@ in an effort to get around the 1994 version of Rule 2:</p>
 <pre><code>    2) Your entry must be &lt;= 3217 bytes in length.  The number of characters
       excluding whitespace (tab, space, newline), and excluding any ; { or }
       followed by either whitespace or end of file, must be &lt;= 1536.</code></pre>
-<p>Indeed, when <a href="dodsond2.gz">dodsond2.gz</a> is uncompressed, it is identical
+<p>Indeed the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/dodsond2/dodsond2.c">dodsond2.c</a> file is
-to the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/dodsond2/dodsond2.c">dodsond2.c</a> file you find today.</p>
+what you find today.</p>
 <p>The IOCCC rules and guidelines were modified in an effort to address
 this fun “<em>abuse of the rules</em>” style hack.</p>
 <p>It is worth noting that without this compression “<em>abuse of the rules</em>”
@@ -51,7 +51,7 @@ source file is self documenting.  :-)
 From time to time, run `ps(1)` and look at the new processes.
 See [shapiro.html](shapiro.html) for more information in the internals of this program.
-The file [shapiro.alt.c](shapiro.alt.c) contains a non-obfuscated version of
+The file [shapiro.alt.c](%%REPO_URL%%/1994/shapiro/shapiro.alt.c) contains a non-obfuscated version of
 this program. Note that this alt file is currently missing. See
 [1994 shapiro bugs](../../bugs.html#1994_shapiro) for details.
@@ -67,18 +67,18 @@ The basic theme (pun) of this program is:
 ```
-My entry ([shapiro.c](shapiro.c)) is mostly comments, formatted in the shape of a
+My entry ([shapiro.c](%%REPO_URL%%/1994/shapiro/shapiro.c)) is mostly comments, formatted in the shape of a
 clock. If you strip out the comments and look at the code you will
 quickly realize that the comments were the important part and that
 the code does very little (see pun above). It writes (to `stdout`)
-another C program ([shapiro_t2.c](shapiro_t2.c)). This is the first level of
+another C program ([shapiro_t2.c](%%REPO_URL%%/1994/shapiro/shapiro_t2.c)). This is the first level of
 obfuscation.
-The second program ([shapiro_t2.c](shapiro_t2.c) (use `make everything` first)
+The second program ([shapiro_t2.c](%%REPO_URL%%/1994/shapiro/shapiro_t2.c) (use `make everything` first)
 prints a clock in the upper right hand corner of your VTxxx/ANSI display.
 Most of the surface obfuscation in the second program
-([shapiro_t2.c](shapiro_t2.c)) was an attempt to make it as small as possible.
+([shapiro_t2.c](%%REPO_URL%%/1994/shapiro/shapiro_t2.c)) was an attempt to make it as small as possible.
 You should be able to see around this with `cb(1)` and some more intelligent
 variable names.  Once you get past this you will realize that the third level of
 obfuscation is a six member client/server hierarchy.  (See the
@@ -403,23 +403,23 @@ bring it back to the foreground.</p>
 source file is self documenting. :-)</p>
 <p>From time to time, run <code>ps(1)</code> and look at the new processes.</p>
 <p>See <a href="shapiro.html">shapiro.html</a> for more information in the internals of this program.
-The file <a href="shapiro.alt.c">shapiro.alt.c</a> contains a non-obfuscated version of
+The file <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/shapiro/shapiro.alt.c">shapiro.alt.c</a> contains a non-obfuscated version of
 this program. Note that this alt file is currently missing. See
 <a href="../../bugs.html#1994_shapiro">1994 shapiro bugs</a> for details.</p>
 <h2 id="authors-remarks">Author’s remarks:</h2>
 <p>The basic theme (pun) of this program is:</p>
 <pre><code>    &quot;This time (everything) is not where it should be.&quot;
          ~~~~</code></pre>
-<p>My entry (<a href="shapiro.c">shapiro.c</a>) is mostly comments, formatted in the shape of a
+<p>My entry (<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/shapiro/shapiro.c">shapiro.c</a>) is mostly comments, formatted in the shape of a
 clock. If you strip out the comments and look at the code you will
 quickly realize that the comments were the important part and that
 the code does very little (see pun above). It writes (to <code>stdout</code>)
-another C program (<a href="shapiro_t2.c">shapiro_t2.c</a>). This is the first level of
+another C program (<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/shapiro/shapiro_t2.c">shapiro_t2.c</a>). This is the first level of
 obfuscation.</p>
-<p>The second program (<a href="shapiro_t2.c">shapiro_t2.c</a> (use <code>make everything</code> first)
+<p>The second program (<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/shapiro/shapiro_t2.c">shapiro_t2.c</a> (use <code>make everything</code> first)
 prints a clock in the upper right hand corner of your VTxxx/ANSI display.</p>
 <p>Most of the surface obfuscation in the second program
-(<a href="shapiro_t2.c">shapiro_t2.c</a>) was an attempt to make it as small as possible.
+(<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/shapiro/shapiro_t2.c">shapiro_t2.c</a>) was an attempt to make it as small as possible.
 You should be able to see around this with <code>cb(1)</code> and some more intelligent
 variable names. Once you get past this you will realize that the third level of
 obfuscation is a six member client/server hierarchy. (See the
@@ -77,7 +77,7 @@ event of a system shutdown or reboot.
 ### Bugs
-The alternative version ([weisberg.alt.c](weisberg.alt.c)) has never actually run
+The alternative version ([weisberg.alt.c](%%REPO_URL%%/1994/weisberg/weisberg.alt.c)) has never actually run
 to completion. After running for close to a week it had reached somewhere around
 250e6, when it became necessary to test the reboot feature (which worked
 remarkably well). As there are quite a lot of primes to be found before reaching
@@ -413,7 +413,7 @@ run to completion, so as an added feature, it will cease execution upon
 reception of a signal (any will do equally well), as well as in the
 event of a system shutdown or reboot.</p>
 <h3 id="bugs">Bugs</h3>
-<p>The alternative version (<a href="weisberg.alt.c">weisberg.alt.c</a>) has never actually run
+<p>The alternative version (<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/weisberg/weisberg.alt.c">weisberg.alt.c</a>) has never actually run
 to completion. After running for close to a week it had reached somewhere around
 250e6, when it became necessary to test the reboot feature (which worked
 remarkably well). As there are quite a lot of primes to be found before reaching
@@ -85,7 +85,7 @@ compiler line you will get a different message (**as compiler errors** due to
 _This is independent of the previous (if there are any previous) compiler
 invocations_ specified. That means with the right compiler invocation you can
-'win' the game on the first go. But unlike the [alternate code](westley.alt.c),
+'win' the game on the first go. But unlike the [alternate code](%%REPO_URL%%/1994/westley/westley.alt.c),
 you do not run this program as it will not compile: instead it will be compiler
 error messages; there is no running the program.
@@ -427,7 +427,7 @@ compiler line you will get a different message (<strong>as compiler errors</stro
 (escape the dungeon) and others will show something else.</p>
 <p><em>This is independent of the previous (if there are any previous) compiler
 invocations</em> specified. That means with the right compiler invocation you can
-‘win’ the game on the first go. But unlike the <a href="westley.alt.c">alternate code</a>,
+‘win’ the game on the first go. But unlike the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1994/westley/westley.alt.c">alternate code</a>,
 you do not run this program as it will not compile: instead it will be compiler
 error messages; there is no running the program.</p>
 <h3 id="original-try">Original try:</h3>
@@ -104,7 +104,7 @@ of operators, and control constructs.
 The compiler has absolutely no error checking, so things
 can go wrong.  But this is what seasoned C programmers are used to.
-Assume that the interpreter has been compiled (source in [august.c](august.c)
+Assume that the interpreter has been compiled (source in [august.c](%%REPO_URL%%/1996/august/august.c)
 and binary in august) and that we have this in a file named `test.oc`:
 ``` <!---c-->
@@ -429,7 +429,7 @@ syntax). Arrays can only be global. There is a rather limited set
 of operators, and control constructs.</p>
 <p>The compiler has absolutely no error checking, so things
 can go wrong. But this is what seasoned C programmers are used to.</p>
-<p>Assume that the interpreter has been compiled (source in <a href="august.c">august.c</a>
+<p>Assume that the interpreter has been compiled (source in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/august/august.c">august.c</a>
 and binary in august) and that we have this in a file named <code>test.oc</code>:</p>
 <pre><code>    main() { putchar(&#39;!&#39;); putchar(&#39;\n&#39;); exit(0); }</code></pre>
 <p>We can then compile by</p>
@@ -48,7 +48,7 @@ make it easier to see what is happening.
 ## Judges' remarks:
 The original program, with only slight obfuscations, may be found
-in [hunni.c](%%REPO_URL%%/1996/schweikh2/hunni.c). The perl script, [hunni.pl](hunni.pl), obfuscates it
+in [hunni.c](%%REPO_URL%%/1996/schweikh2/hunni.c). The perl script, [hunni.pl](%%REPO_URL%%/1996/schweikh2/hunni.pl), obfuscates it
 appropriately. Try:
 ``` <!---sh-->
@@ -399,7 +399,7 @@ ctrl-c to stop it. Observe how the <a href="https://github.com/ioccc-src/temp-te
 make it easier to see what is happening.</p>
 <h2 id="judges-remarks">Judges’ remarks:</h2>
 <p>The original program, with only slight obfuscations, may be found
-in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/schweikh2/hunni.c">hunni.c</a>. The perl script, <a href="hunni.pl">hunni.pl</a>, obfuscates it
+in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/schweikh2/hunni.c">hunni.c</a>. The perl script, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/schweikh2/hunni.pl">hunni.pl</a>, obfuscates it
 appropriately. Try:</p>
 <pre><code>    less hunni.c
@@ -131,7 +131,7 @@ situation and effectively halts.
 I had the pleasure and pain to find all three behaviours on the
 platforms I use regularly.
-The file [schweikh3.info](schweikh3.info) has output for
+The file [schweikh3.info](%%REPO_URL%%/1996/schweikh3/schweikh3.info) has output for
 Solaris, FreeBSD and Linux. The Linux case is interesting in that
 the behaviour changes when `calloc(3)` is used instead of `malloc(3)`.
 For FreeBSD, the last two lines are not messages stemming from
@@ -463,7 +463,7 @@ situation and effectively halts.</p></li>
 </ol>
 <p>I had the pleasure and pain to find all three behaviours on the
 platforms I use regularly.</p>
-<p>The file <a href="schweikh3.info">schweikh3.info</a> has output for
+<p>The file <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/schweikh3/schweikh3.info">schweikh3.info</a> has output for
 Solaris, FreeBSD and Linux. The Linux case is interesting in that
 the behaviour changes when <code>calloc(3)</code> is used instead of <code>malloc(3)</code>.
 For FreeBSD, the last two lines are not messages stemming from
@@ -60,8 +60,8 @@ This 1-line program allows various analog ASCII clocks to print out the time;
 the ASCII clock is built from the command line arguments, so many different
 types of clocks can be generated.  Three different scripts are supplied to print
 out a grandfather clock, a mantle clock that runs backwards, and a linear-style
-clock.  (they are [clock1.sh](clock1.sh), [clock2.sh](clock2.sh), and
+clock.  (they are [clock1.sh](%%REPO_URL%%/1996/westley/clock1.sh), [clock2.sh](%%REPO_URL%%/1996/westley/clock2.sh), and
-[clock3.sh](clock3.sh))
+[clock3.sh](%%REPO_URL%%/1996/westley/clock3.sh))
 The program makes the following (possibly non-portable) assumptions:
@@ -406,8 +406,8 @@ understand it.</p>
 the ASCII clock is built from the command line arguments, so many different
 types of clocks can be generated. Three different scripts are supplied to print
 out a grandfather clock, a mantle clock that runs backwards, and a linear-style
-clock. (they are <a href="clock1.sh">clock1.sh</a>, <a href="clock2.sh">clock2.sh</a>, and
+clock. (they are <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/westley/clock1.sh">clock1.sh</a>, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/westley/clock2.sh">clock2.sh</a>, and
-<a href="clock3.sh">clock3.sh</a>)</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/westley/clock3.sh">clock3.sh</a>)</p>
 <p>The program makes the following (possibly non-portable) assumptions:</p>
 <ol type="1">
 <li><p>Command line strings can be written to.</p></li>
@@ -219,7 +219,7 @@ Here is a brief description of the scenery files:
    Simple obstacle course.  Try to fly over the buildings and under the
    bridges.
- [pyramids.sc](1998/banks/pyramids.sc)
+- [pyramids.sc](%%REPO_URL%%/1998/banks/pyramids.sc)
    Fly over the tombs of the ancient Pharaohs, in this (fictitious) Egyptian
    landscape.
@@ -519,7 +519,7 @@ to your right.</p></li>
 <li><p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/banks/bb.sc">bb.sc</a></p>
 <p>Simple obstacle course. Try to fly over the buildings and under the
 bridges.</p></li>
-<li><p><a href="1998/banks/pyramids.sc">pyramids.sc</a></p>
+<li><p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/banks/pyramids.sc">pyramids.sc</a></p>
 <p>Fly over the tombs of the ancient Pharaohs, in this (fictitious) Egyptian
 landscape.</p></li>
 <li><p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/banks/river.sc">river.sc</a></p>
@@ -28,7 +28,7 @@ Enter an expression on standard input.  To try some we have selected:
 ### Historical aside:
-At a time the code in [fanf.c](%%REPO_URL%%/1998/fanf/fanf.c) was that of [fanf.orig.c](fanf.orig.c)
+At a time the code in [fanf.c](%%REPO_URL%%/1998/fanf/fanf.c) was that of [fanf.orig.c](%%REPO_URL%%/1998/fanf/fanf.orig.c)
 but to get this to compile in modern systems it had to be translated to what you
 now see. The intermediate steps can still be performed but they might be
 different from the past. This should be kept in mind as you read the below
@@ -388,7 +388,7 @@ this entry because of the system resources it requires.</p>
 <pre><code>    ./try.sh</code></pre>
 <h2 id="judges-remarks">Judges’ remarks:</h2>
 <h3 id="historical-aside">Historical aside:</h3>
-<p>At a time the code in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/fanf/fanf.c">fanf.c</a> was that of <a href="fanf.orig.c">fanf.orig.c</a>
+<p>At a time the code in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/fanf/fanf.c">fanf.c</a> was that of <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/fanf/fanf.orig.c">fanf.orig.c</a>
 but to get this to compile in modern systems it had to be translated to what you
 now see. The intermediate steps can still be performed but they might be
 different from the past. This should be kept in mind as you read the below
@@ -78,7 +78,7 @@ the letter `e`.
 The IOCCC has been no stranger to this concept.  Several winning entries from
 years past have accomplished interesting feats while completely avoiding the use
 of certain C constructs thought to be essential (e.g.
-[1988/robison.c](../../1988/robison/robison.c)).  If you have to ask why they would
+[1988/robison.c](%%REPO_URL%%/1988/robison/robison.c)).  If you have to ask why they would
 do it this way, then this contest just isn't for you.
 In continuing with this fine tradition, this program suggests to the ANSI
@@ -422,7 +422,7 @@ the letter <code>e</code>.</p>
 <p>The IOCCC has been no stranger to this concept. Several winning entries from
 years past have accomplished interesting feats while completely avoiding the use
 of certain C constructs thought to be essential (e.g.
-<a href="../../1988/robison/robison.c">1988/robison.c</a>). If you have to ask why they would
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1988/robison/robison.c">1988/robison.c</a>). If you have to ask why they would
 do it this way, then this contest just isn’t for you.</p>
 <p>In continuing with this fine tradition, this program suggests to the ANSI
 committee some new practical simplifications for the C language. For instance,
@@ -247,7 +247,7 @@ printer when you make a hard copy of the source...
 While I'm at it, the rules state that only space, tab and newline
 are ignored for the count (plus '`{`', '`}`', '`;`' followed by
-whitespace). The [mkentry.c](../mkentry.c) program, however, uses `isspace(3)` which
+whitespace). The [mkentry.c](%%REPO_URL%%/1998/mkentry.c) program, however, uses `isspace(3)` which
 returns nonzero for `\v` and `\f` and other characters as well.
 I could have used a lot more `^K` and `^L` probably undetected by your
 counter but decided to err on the side of safety. I use a perl
@@ -546,7 +546,7 @@ something new – after 14 years of IOCCC. Be sure to have enough paper in your
 printer when you make a hard copy of the source…</p>
 <p>While I’m at it, the rules state that only space, tab and newline
 are ignored for the count (plus ‘<code>{</code>’, ‘<code>}</code>’, ‘<code>;</code>’ followed by
-whitespace). The <a href="../mkentry.c">mkentry.c</a> program, however, uses <code>isspace(3)</code> which
+whitespace). The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1998/mkentry.c">mkentry.c</a> program, however, uses <code>isspace(3)</code> which
 returns nonzero for <code>\v</code> and <code>\f</code> and other characters as well.
 I could have used a lot more <code>^K</code> and <code>^L</code> probably undetected by your
 counter but decided to err on the side of safety. I use a perl
@@ -45,7 +45,7 @@ them for the next IOCCC.
 We believe you will be impressed with this year's winning entries.  The [Best of
 Show](jarijyrki/index.html) is a fine example of compact obfuscation. But don't
 ignore the shorter entries.  The [Best Small Program](natori/index.html), [Best
-Abuse of User](briddlebane/README.m) and [Most Complete Program](tomx/index.html)
+Abuse of User](briddlebane/index.html) and [Most Complete Program](tomx/index.html)
 are well worth studying in detail.
 The [Worst Abuse of the Rules](dlowe/index.html) is technically allowed by the
@@ -324,7 +324,7 @@ en/decoded[^4].
 ### Photo Credit:
 [Lavabus](%%REPO_URL%%/2000/bmeyer/lavabus.pgm) was taken by the judge, Landon Curt Noll.
-[Michael](%%REPO_URL%%/2000/bmeyer/michael.pgm) was provided by the author.  [Lenna](lenna.glic) is a
+[Michael](%%REPO_URL%%/2000/bmeyer/michael.pgm) was provided by the author.  [Lenna](%%REPO_URL%%/2000/bmeyer/lenna.glic) is a
 1972 Playboy centerfold (doesn't show very much).
 - [michael.pgm](%%REPO_URL%%/2000/bmeyer/michael.pgm)
@@ -584,7 +584,7 @@ interval halving repeated until the precise value has been
 en/decoded<a href="#fn5" class="footnote-ref" id="fnref5" role="doc-noteref"><sup>5</sup></a>.</p>
 <h3 id="photo-credit">Photo Credit:</h3>
 <p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/bmeyer/lavabus.pgm">Lavabus</a> was taken by the judge, Landon Curt Noll.
-<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/bmeyer/michael.pgm">Michael</a> was provided by the author. <a href="lenna.glic">Lenna</a> is a
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/bmeyer/michael.pgm">Michael</a> was provided by the author. <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/bmeyer/lenna.glic">Lenna</a> is a
 1972 Playboy centerfold (doesn’t show very much).</p>
 <ul>
 <li><p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/bmeyer/michael.pgm">michael.pgm</a></p>
@@ -394,7 +394,7 @@ that was not supplied with the entry.</li>
 them for the next IOCCC.</p>
 <p>We believe you will be impressed with this year’s winning entries. The <a href="jarijyrki/index.html">Best of
 Show</a> is a fine example of compact obfuscation. But don’t
-ignore the shorter entries. The <a href="natori/index.html">Best Small Program</a>, <a href="briddlebane/README.m">Best
+ignore the shorter entries. The <a href="natori/index.html">Best Small Program</a>, <a href="briddlebane/index.html">Best
 Abuse of User</a> and <a href="tomx/index.html">Most Complete Program</a>
 are well worth studying in detail.</p>
 <p>The <a href="dlowe/index.html">Worst Abuse of the Rules</a> is technically allowed by the
@@ -81,7 +81,7 @@ Julian day number (given as a fractional component). E.g.:
 A Julian day is defined to be the number of days since January 1st in the year
 `-4712 (4713BC)`, starting at 12 noon GMT.[^1][^2] To aid conversion from our time
 (GMT) to Julian day there is a supplementary program named:
-[dmy2jd.c](dmy2jd.c). Run it with `day_float month year` as arguments. See below
+[dmy2jd.c](%%REPO_URL%%/2000/rince/dmy2jd.c). Run it with `day_float month year` as arguments. See below
 for an example.
@@ -417,7 +417,7 @@ Julian day number (given as a fractional component). E.g.:</p>
 <p>A Julian day is defined to be the number of days since January 1st in the year
 <code>-4712 (4713BC)</code>, starting at 12 noon GMT.<a href="#fn1" class="footnote-ref" id="fnref1" role="doc-noteref"><sup>1</sup></a><a href="#fn2" class="footnote-ref" id="fnref2" role="doc-noteref"><sup>2</sup></a> To aid conversion from our time
 (GMT) to Julian day there is a supplementary program named:
-<a href="dmy2jd.c">dmy2jd.c</a>. Run it with <code>day_float month year</code> as arguments. See below
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/rince/dmy2jd.c">dmy2jd.c</a>. Run it with <code>day_float month year</code> as arguments. See below
 for an example.</p>
 <h3 id="what-it-does">What it does</h3>
 <p>This program plots the positions of the four <a href="https://en.wikipedia.org/wiki/Galilean_moons">Galilean
@@ -61,7 +61,7 @@ FWIW, none of the compilers with which this entry was tested issued
 any warnings or errors in compiling the source for this entry.  No
 doubt lint will complain about the goto statements.
-The [schneiderwent.data](schneiderwent.data) file consists of the values used by
+The [schneiderwent.data](%%REPO_URL%%/2000/schneiderwent/schneiderwent.data) file consists of the values used by
 multiple table lookups.  This is simply good program design, separating these
 values from the source code allows changing the behavior of the application
 without recompiling.
@@ -406,7 +406,7 @@ keeping the program portable between different encodings
 <p>FWIW, none of the compilers with which this entry was tested issued
 any warnings or errors in compiling the source for this entry. No
 doubt lint will complain about the goto statements.</p>
-<p>The <a href="schneiderwent.data">schneiderwent.data</a> file consists of the values used by
+<p>The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/schneiderwent/schneiderwent.data">schneiderwent.data</a> file consists of the values used by
 multiple table lookups. This is simply good program design, separating these
 values from the source code allows changing the behavior of the application
 without recompiling.</p>
@@ -4,7 +4,7 @@
    make
 ```
-There is an alternate version that compiles [mkentry.c](mkentry.c). See
+There is an alternate version that compiles [mkentry.c](%%REPO_URL%%/2000/mkentry.c). See
 [alternate code](#alternate-code).
@@ -379,7 +379,7 @@ Location: <a href="../../location.html#NL">NL</a> - <em>Kingdom of the Netherlan
 <!-- BEFORE: 1st line of markdown file: 2000/tomx/README.md -->
 <h2 id="to-build">To build:</h2>
 <pre><code>    make</code></pre>
-<p>There is an alternate version that compiles <a href="mkentry.c">mkentry.c</a>. See
+<p>There is an alternate version that compiles <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/mkentry.c">mkentry.c</a>. See
 <a href="#alternate-code">alternate code</a>.</p>
 <h2 id="to-use">To use:</h2>
 <pre><code>    ./tomx</code></pre>
@@ -41,7 +41,7 @@ For more detailed information see [2001 anonymous bugs](../../bugs.html#2001_ano
 ```
 NOTE: if the 32-bit version cannot be compiled the script will at least compile
-[anonymous.ten](anonymous.ten.c) as a 64-bit program (or whatever your system is
+[anonymous.ten](%%REPO_URL%%/2001/anonymous/anonymous.ten.c) as a 64-bit program (or whatever your system is
 set to) and run it directly.
 What happens if the x86 program has already been modified by this program? The
@@ -105,7 +105,7 @@ run x86 programs on any machine (x86 or otherwise).
 I have included a simple [10 Green
 Bottles](https://www.bbc.co.uk/teach/school-radio/nursery-rhymes-ten-green-bottles/zncyt39)
-[program](anonymous.ten.c) (try `make anonymous.ten`), meant to be compiled as a
+[program](%%REPO_URL%%/2001/anonymous/anonymous.ten.c) (try `make anonymous.ten`), meant to be compiled as a
 32-bit ELF binary.
 The program `anonymous.ten`, and its source (dull) are included as data
@@ -129,7 +129,7 @@ generates a translation for an entire basic block of subject code.  Having
 translated a block once, the target code generated will be stored for the
 duration that the program is running, so as a program runs its performance will
 improve, as it hits cached blocks.  This can be clearly observed in the example
-program, [anonymous.ten](anonymous.ten.c) which must be compiled as a 32-bit ELF
+program, [anonymous.ten](%%REPO_URL%%/2001/anonymous/anonymous.ten.c) which must be compiled as a 32-bit ELF
 binary.
 ### Target Library Linking:
@@ -221,7 +221,7 @@ Despite all this, the translator is not exclusively limited to running the
 [anonymous.ten](%%REPO_URL%%/2001/anonymous/anonymous.ten.c) program.  Other trivial **_x86_** programs may
 run on the translator \- and I have successfully run a wide range of "Hello
 World" programs, including one of last year's IOCCC entries,
-[tomx](2001/anonymous/2000/tomx/tomx.c).
+[tomx](%%REPO_URL%%/2000/tomx/index.html).
 ### Complete Program
@@ -398,7 +398,7 @@ whatever your system will compile them as).</p>
    ./anonymous anonymous.bed # if able to compile as a 32-bit (-m32) ELF binary</code></pre>
 <p>NOTE: if the 32-bit version cannot be compiled the script will at least compile
-<a href="anonymous.ten.c">anonymous.ten</a> as a 64-bit program (or whatever your system is
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2001/anonymous/anonymous.ten.c">anonymous.ten</a> as a 64-bit program (or whatever your system is
 set to) and run it directly.</p>
 <p>What happens if the x86 program has already been modified by this program? The
 judges’ remarks below might give you a hint!</p>
@@ -436,7 +436,7 @@ quite a lot of bit twiddling.</p>
 run x86 programs on any machine (x86 or otherwise).</p>
 <p>I have included a simple <a href="https://www.bbc.co.uk/teach/school-radio/nursery-rhymes-ten-green-bottles/zncyt39">10 Green
 Bottles</a>
-<a href="anonymous.ten.c">program</a> (try <code>make anonymous.ten</code>), meant to be compiled as a
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2001/anonymous/anonymous.ten.c">program</a> (try <code>make anonymous.ten</code>), meant to be compiled as a
 32-bit ELF binary.</p>
 <p>The program <code>anonymous.ten</code>, and its source (dull) are included as data
 files.</p>
@@ -453,7 +453,7 @@ generates a translation for an entire basic block of subject code. Having
 translated a block once, the target code generated will be stored for the
 duration that the program is running, so as a program runs its performance will
 improve, as it hits cached blocks. This can be clearly observed in the example
-program, <a href="anonymous.ten.c">anonymous.ten</a> which must be compiled as a 32-bit ELF
+program, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2001/anonymous/anonymous.ten.c">anonymous.ten</a> which must be compiled as a 32-bit ELF
 binary.</p>
 <h3 id="target-library-linking">Target Library Linking:</h3>
 <p>The program runs <a href="https://en.wikipedia.org/wiki/Dynamic_linker">dynamically
@@ -523,7 +523,7 @@ turn up - yes the program is unsafe, but it’s pretty cool anyway.</p>
 <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2001/anonymous/anonymous.ten.c">anonymous.ten</a> program. Other trivial <strong><em>x86</em></strong> programs may
 run on the translator - and I have successfully run a wide range of “Hello
 World” programs, including one of last year’s IOCCC entries,
-<a href="2001/anonymous/2000/tomx/tomx.c">tomx</a>.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2000/tomx/index.html">tomx</a>.</p>
 <h3 id="complete-program">Complete Program</h3>
 <p>The fact that the program makes calls to <code>system(3)</code> and <code>execv(3)</code> may imply that
 it is not a complete program in its own right, as it relies on other programs.
@@ -82,7 +82,10 @@ put `#!/usr/local/bin/otcc` at the beginning of your C source if
 you installed `otcc` at this place.
-### Supported C language subset (read joint example [bellard.otccex.c](bellard.otccex.c) to have an introduction to `OTCC` dialect)
+### Supported C language subset
 Read joint example [bellard.otccex.c](%%REPO_URL%%/2001/bellard/bellard.otccex.c) to have an introduction to `OTCC` dialect.
 #### Expressions:
@@ -418,7 +418,8 @@ program.</p>
 <p>An alternate syntax is to use it as a script interpreter: you can
 put <code>#!/usr/local/bin/otcc</code> at the beginning of your C source if
 you installed <code>otcc</code> at this place.</p>
-<h3 id="supported-c-language-subset-read-joint-example-bellard.otccex.c-to-have-an-introduction-to-otcc-dialect">Supported C language subset (read joint example <a href="bellard.otccex.c">bellard.otccex.c</a> to have an introduction to <code>OTCC</code> dialect)</h3>
+<h3 id="supported-c-language-subset">Supported C language subset</h3>
 <p>Read joint example <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2001/bellard/bellard.otccex.c">bellard.otccex.c</a> to have an introduction to <code>OTCC</code> dialect.</p>
 <h4 id="expressions">Expressions:</h4>
 <ul>
 <li><p>binary operators, in decreasing precedence order: <code>*</code> <code>/</code> <code>%</code>,
@@ -26,11 +26,11 @@ local compiler.
 ## Remarks on some of the entries
 We believe you will be impressed with this year's winning entries.  The [Best of
-Show](gavin/gavin.c) is a fine example of obfuscation.  But don't ignore the
+Show](%%REPO_URL%%/2004/gavin/gavin.c) is a fine example of obfuscation.  But don't ignore the
 other entries!  There are games, utilities, programs that output images and a
 one-liner that is incredibly compact for what it can do.
-The [Best Abuse of the Guidelines](hibachi/hibachi.c) is technically allowed by
+The [Best Abuse of the Guidelines](%%REPO_URL%%/2004/hibachi/hibachi.c) is technically allowed by
 the rules.  This year we allowed it due to the functionality of the program.
 This entry is one of the easiest to read!  Don't assume you can win using
 similar techniques next time ... :-)
@@ -379,11 +379,11 @@ require certain changes (add or remove a library, add or remove a
 yours is lacking, you may need to compile using <code>gcc</code> instead of your
 local compiler.</p>
 <h2 id="remarks-on-some-of-the-entries">Remarks on some of the entries</h2>
-<p>We believe you will be impressed with this year’s winning entries. The <a href="gavin/gavin.c">Best of
+<p>We believe you will be impressed with this year’s winning entries. The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/gavin/gavin.c">Best of
 Show</a> is a fine example of obfuscation. But don’t ignore the
 other entries! There are games, utilities, programs that output images and a
 one-liner that is incredibly compact for what it can do.</p>
-<p>The <a href="hibachi/hibachi.c">Best Abuse of the Guidelines</a> is technically allowed by
+<p>The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/hibachi/hibachi.c">Best Abuse of the Guidelines</a> is technically allowed by
 the rules. This year we allowed it due to the functionality of the program.
 This entry is one of the easiest to read! Don’t assume you can win using
 similar techniques next time … :-)</p>
@@ -114,7 +114,7 @@ last digit.  Four common elements could also be written as isotopes of these two
 elements: `He=Np(2)`, `Li=Pu(2)`, `Ta=Np(3)`, `W=Pu(3)`.  To get other isotopes,
 just change the last digit to a number greater than 3.
-The build file runs [jdalbec.c](jdalbec.c) through the C preprocessor to
+The build file runs [jdalbec.c](%%REPO_URL%%/2004/jdalbec/jdalbec.c) through the C preprocessor to
 generate `jdalbec2.c` which is edited with `ed(1)` (NOTE: this was
 changed to `sed(1)` by the judges) to correct spacing
 problems caused by some preprocessors.  (I know you dislike
@@ -122,10 +122,10 @@ Perl, but hopefully `ed(1)` is more common on POSIX systems than
 Perl.)  The build file then compiles `jdalbec2.c]` and removes the
 `jdalbec2.c` file.
-I don't think I'm abusing the [Makefile](Makefile) since I'm only using
+I don't think I'm abusing the [Makefile](%%REPO_URL%%/2004/jdalbec/Makefile) since I'm only using
 147 characters including white space and the `#!` line and the
 commented-out `-Wall -pedantic` options.  The [rules for the
-16th IOCCC](../../2004/rules.txt) specifically allowed build files up to 521
+16th IOCCC](../rules.txt) specifically allowed build files up to 521
 characters long so my build file would have been well within
 those limits.  (What typo?)
@@ -168,7 +168,7 @@ line 3 because the compiler doesn't recognize `??=` as a trigraph
 step on Mac OS X 10.1.5 generates lots of errors in "smart
 mode" because the "smart" preprocessor expects a valid C
 program as input.  The preprocessor then falls back to "basic
-mode" and preprocesses [jdalbec.c](jdalbec.c) successfully without warnings,
+mode" and preprocesses [jdalbec.c](%%REPO_URL%%/2004/jdalbec/jdalbec.c) successfully without warnings,
 even when I give it the `-Wall -pedantic` options.  It inserts
 a lot of extra spaces into the preprocessed output which is
 why I had to add the `ed(1)` step to the build file so that my
@@ -191,7 +191,7 @@ it if you run into this feature (e.g. `1 22` instead of `122`).
 Note: you may want to be sure you have a limit on stack size before trying this.
 I ran it on Debian GNU/Linux and discovered that I did not have a limit on stack
 size.  Fortunately the OS killed the process after a while.  This is consistent
-with the output of [1996/schweikh3](../../1996/schweikh3/schweikh3.c) which reports
+with the output of [1996/schweikh3](%%REPO_URL%%/1996/schweikh3/schweikh3.c) which reports
 `malloc lied!  Killed by SIGKILL`.
 Arguments matching `/(?:.){10,}/` will produce visually incorrect
@@ -434,16 +434,16 @@ the <a href="https://en.wikipedia.org/wiki/Isotope">isotopes</a> of
 last digit. Four common elements could also be written as isotopes of these two
 elements: <code>He=Np(2)</code>, <code>Li=Pu(2)</code>, <code>Ta=Np(3)</code>, <code>W=Pu(3)</code>. To get other isotopes,
 just change the last digit to a number greater than 3.</p>
-<p>The build file runs <a href="jdalbec.c">jdalbec.c</a> through the C preprocessor to
+<p>The build file runs <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/jdalbec/jdalbec.c">jdalbec.c</a> through the C preprocessor to
 generate <code>jdalbec2.c</code> which is edited with <code>ed(1)</code> (NOTE: this was
 changed to <code>sed(1)</code> by the judges) to correct spacing
 problems caused by some preprocessors. (I know you dislike
 Perl, but hopefully <code>ed(1)</code> is more common on POSIX systems than
 Perl.) The build file then compiles <code>jdalbec2.c]</code> and removes the
 <code>jdalbec2.c</code> file.</p>
-<p>I don’t think I’m abusing the <a href="Makefile">Makefile</a> since I’m only using
+<p>I don’t think I’m abusing the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/jdalbec/Makefile">Makefile</a> since I’m only using
 147 characters including white space and the <code>#!</code> line and the
-commented-out <code>-Wall -pedantic</code> options. The <a href="../../2004/rules.txt">rules for the
+commented-out <code>-Wall -pedantic</code> options. The <a href="../rules.txt">rules for the
 16th IOCCC</a> specifically allowed build files up to 521
 characters long so my build file would have been well within
 those limits. (What typo?)</p>
@@ -481,7 +481,7 @@ line 3 because the compiler doesn’t recognize <code>??=</code> as a trigraph
 step on Mac OS X 10.1.5 generates lots of errors in “smart
 mode” because the “smart” preprocessor expects a valid C
 program as input. The preprocessor then falls back to “basic
-mode” and preprocesses <a href="jdalbec.c">jdalbec.c</a> successfully without warnings,
+mode” and preprocesses <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/jdalbec/jdalbec.c">jdalbec.c</a> successfully without warnings,
 even when I give it the <code>-Wall -pedantic</code> options. It inserts
 a lot of extra spaces into the preprocessed output which is
 why I had to add the <code>ed(1)</code> step to the build file so that my
@@ -500,7 +500,7 @@ it if you run into this feature (e.g. <code>1 22</code> instead of <code>122</c
 <p>Note: you may want to be sure you have a limit on stack size before trying this.
 I ran it on Debian GNU/Linux and discovered that I did not have a limit on stack
 size. Fortunately the OS killed the process after a while. This is consistent
-with the output of <a href="../../1996/schweikh3/schweikh3.c">1996/schweikh3</a> which reports
+with the output of <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/1996/schweikh3/schweikh3.c">1996/schweikh3</a> which reports
 <code>malloc lied!  Killed by SIGKILL</code>.</p>
 <p>Arguments matching <code>/(?:.){10,}/</code> will produce visually incorrect
 results (e.g., generation 1 starting from 1111111111 will
@@ -101,7 +101,7 @@ the number from 8 to 11 that it thinks it is.
  (`$?` is shorter)
 * I achieved some additional tightness by `-Dif=while` which saved 9 chars of
-source for the cost of 11 chars of the [Makefile](Makefile). If I used
+source for the cost of 11 chars of the [Makefile](%%REPO_URL%%/2004/kopczynski/Makefile). If I used
 `-DB=while`, I would save 12 for 10, but I think the current version is slightly
 more obfuscated.
@@ -436,7 +436,7 @@ number, not to print it, but if you think otherwise, the change would
 cost 6 chars (plus a warning). Change <code>return *</code> to <code>printf("%d",*)</code>.
 (<code>$?</code> is shorter)</p></li>
 <li><p>I achieved some additional tightness by <code>-Dif=while</code> which saved 9 chars of
-source for the cost of 11 chars of the <a href="Makefile">Makefile</a>. If I used
+source for the cost of 11 chars of the <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2004/kopczynski/Makefile">Makefile</a>. If I used
 <code>-DB=while</code>, I would save 12 for 10, but I think the current version is slightly
 more obfuscated.</p></li>
 </ul>
@@ -58,7 +58,7 @@ more while it puzzles out some puzzles: all in a puzzling way!
 `goto` the source and notice the lack of functions.  Jump
 to the `switch` statement.  And if you think you can puzzle
-it out better, try solving [insane1.sudoku](insane1.sudoku) all by yourself!
+it out better, try solving [insane1.sudoku](%%REPO_URL%%/2005/aidan/insane1.sudoku) all by yourself!
 NOTE: the tarball named by the author was removed after the files were extracted
 into the directory instead. These files were added to the list.
@@ -91,7 +91,7 @@ Yes, the test suite really is 5 times larger than the actual program. You
 ### IMPORTANT
-[test.perl](test-su.perl) expects the `sudoku-sf` executable to be
+[test.perl](%%REPO_URL%%/2005/aidan/test-su.perl) expects the `sudoku-sf` executable to be
 `./sudoku-sf`. If you rename it, please change the definition of `$sf_exe` to the
 new name.
@@ -219,7 +219,7 @@ Input should be the numbers of each row in turn. Empty spaces can be
 represented as a full stop (`.`) or zero (`0`). Other characters are ignored, so you
 can cut-and-paste either example sudoku puzzle from above into a file and feed that
 in. The program has coped with every input I've thrown at it so far, including
-an [empty grid](blank.sudoku).
+an [empty grid](%%REPO_URL%%/2005/aidan/blank.sudoku).
 Output consists of the problem, then the solution, then a message of
 success/failure.
@@ -402,7 +402,7 @@ which is slower, inelegant and not as obscure, as the author put it.</p>
 more while it puzzles out some puzzles: all in a puzzling way!</p>
 <p><code>goto</code> the source and notice the lack of functions. Jump
 to the <code>switch</code> statement. And if you think you can puzzle
-it out better, try solving <a href="insane1.sudoku">insane1.sudoku</a> all by yourself!</p>
+it out better, try solving <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/aidan/insane1.sudoku">insane1.sudoku</a> all by yourself!</p>
 <p>NOTE: the tarball named by the author was removed after the files were extracted
 into the directory instead. These files were added to the list.</p>
 <h2 id="authors-remarks">Author’s remarks:</h2>
@@ -425,7 +425,7 @@ generate new ones.</p>
 <p>Yes, the test suite really is 5 times larger than the actual program. You
 <strong>did</strong> say this was a parody of the software development process…</p>
 <h3 id="important">IMPORTANT</h3>
-<p><a href="test-su.perl">test.perl</a> expects the <code>sudoku-sf</code> executable to be
+<p><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/aidan/test-su.perl">test.perl</a> expects the <code>sudoku-sf</code> executable to be
 <code>./sudoku-sf</code>. If you rename it, please change the definition of <code>$sf_exe</code> to the
 new name.</p>
 <p><strong>Judges’ note</strong>: this was done.</p>
@@ -507,7 +507,7 @@ really.</p>
 represented as a full stop (<code>.</code>) or zero (<code>0</code>). Other characters are ignored, so you
 can cut-and-paste either example sudoku puzzle from above into a file and feed that
 in. The program has coped with every input I’ve thrown at it so far, including
-an <a href="blank.sudoku">empty grid</a>.</p>
+an <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2005/aidan/blank.sudoku">empty grid</a>.</p>
 <p>Output consists of the problem, then the solution, then a message of
 success/failure.</p>
 <p><code>./aidan U</code> may also be useful, or perhaps not. It only took
@@ -37,8 +37,8 @@ NOTE: if your move is invalid it will prompt you again but not show the board.
 This entry turns a newspaper solitaire puzzle into a competition
 with the computer to be the last one moving.  The object of this
-[source code](meyer.c) is to determine the object of this [source
+[source code](%%REPO_URL%%/2006/meyer/meyer.c) is to determine the object of this [source
-code](meyer.c)!  :-)
+code](%%REPO_URL%%/2006/meyer/meyer.c)!  :-)
 ## Author's remarks:
@@ -393,7 +393,7 @@ more text to fulfil the program’s requirements.</p>
 <h2 id="judges-remarks">Judges’ remarks:</h2>
 <p>This entry turns a newspaper solitaire puzzle into a competition
 with the computer to be the last one moving. The object of this
-<a href="meyer.c">source code</a> is to determine the object of this <a href="meyer.c">source
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/meyer/meyer.c">source code</a> is to determine the object of this <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/meyer/meyer.c">source
 code</a>! :-)</p>
 <h2 id="authors-remarks">Author’s remarks:</h2>
 <p>I wanted submit a fully object-oriented entry to show that
@@ -4,7 +4,7 @@
    make
 ```
-There is an alt version based on the author's remarks. The [try.sh](try.sh)
+There is an alt version based on the author's remarks. The [try.sh](%%REPO_URL%%/2006/sykes1/try.sh)
 script will compile and use both but if you wish to see the alternate version
 specifically see the [Alternate code](#alternate-code) section below.
@@ -127,7 +127,7 @@ Some instructions on how to make your own cube are in the included file
 [bedlam-cubes.pdf](%%REPO_URL%%/2006/sykes1/bedlam-cubes.pdf).
 Finally, you can watch it being solved with the video file here,
-[solving-bedlam-cube.mp4](solving-bedlam-cube.mp4).
+[solving-bedlam-cube.mp4](%%REPO_URL%%/2006/sykes1/solving-bedlam-cube.mp4).
 Actually the cube is very hard to do by hand by just trying to fit it
 together.  I have not known anyone to solve it in this way - you would
@@ -379,7 +379,7 @@ Location: <a href="../../location.html#FI">FI</a> - <em>Republic of Finland </em
 <!-- BEFORE: 1st line of markdown file: 2006/sykes1/README.md -->
 <h2 id="to-build">To build:</h2>
 <pre><code>    make</code></pre>
-<p>There is an alt version based on the author’s remarks. The <a href="try.sh">try.sh</a>
+<p>There is an alt version based on the author’s remarks. The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/sykes1/try.sh">try.sh</a>
 script will compile and use both but if you wish to see the alternate version
 specifically see the <a href="#alternate-code">Alternate code</a> section below.</p>
 <h3 id="bugs-and-misfeatures">Bugs and (Mis)features:</h3>
@@ -431,7 +431,7 @@ puzzles in wikipedia</a>.</p>
 <p>Some instructions on how to make your own cube are in the included file
 <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/sykes1/bedlam-cubes.pdf">bedlam-cubes.pdf</a>.</p>
 <p>Finally, you can watch it being solved with the video file here,
-<a href="solving-bedlam-cube.mp4">solving-bedlam-cube.mp4</a>.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/sykes1/solving-bedlam-cube.mp4">solving-bedlam-cube.mp4</a>.</p>
 <p>Actually the cube is very hard to do by hand by just trying to fit it
 together. I have not known anyone to solve it in this way - you would
 need to be very lucky indeed to stumble across a solution.</p>
@@ -102,8 +102,8 @@ and a disk controller, just like at the start of the personal computers
 revolution (circa 1975).
 It needs an initial memory image to do something usable, so along with it you
-will find two files: ([C.BASIC](C.BASIC) and [C.BIOS](C.BIOS)). Rename
+will find two files: ([C.BASIC](%%REPO_URL%%/2006/toledo2/C.BASIC) and [C.BIOS](%%REPO_URL%%/2006/toledo2/C.BIOS)). Rename
-[C.BASIC](C.BASIC) to `C` (NOTE from the judges: running `make` will do this for
+[C.BASIC](%%REPO_URL%%/2006/toledo2/C.BASIC) to `C` (NOTE from the judges: running `make` will do this for
 you), run the emulator, and et voila! you have the public domain Palo Alto [Tiny
 BASIC](https://en.wikipedia.org/wiki/Tiny_BASIC) (by Li-Chen Wang), published in
 the very first volume of [Dr. Dobb's
@@ -432,8 +432,8 @@ removes those files extracted from the <code>KAYPROII.ZIP</code>.</p>
 and a disk controller, just like at the start of the personal computers
 revolution (circa 1975).</p>
 <p>It needs an initial memory image to do something usable, so along with it you
-will find two files: (<a href="C.BASIC">C.BASIC</a> and <a href="C.BIOS">C.BIOS</a>). Rename
+will find two files: (<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/toledo2/C.BASIC">C.BASIC</a> and <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/toledo2/C.BIOS">C.BIOS</a>). Rename
-<a href="C.BASIC">C.BASIC</a> to <code>C</code> (NOTE from the judges: running <code>make</code> will do this for
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2006/toledo2/C.BASIC">C.BASIC</a> to <code>C</code> (NOTE from the judges: running <code>make</code> will do this for
 you), run the emulator, and et voila! you have the public domain Palo Alto <a href="https://en.wikipedia.org/wiki/Tiny_BASIC">Tiny
 BASIC</a> (by Li-Chen Wang), published in
 the very first volume of <a href="https://en.wikipedia.org/wiki/Dr._Dobb%27s_Journal">Dr. Dobb’s
@@ -96,7 +96,7 @@ When evolving forwards, the code can only be stopped by killing the
 program. In the reverse direction, the code can either be killed, or
 will stop if it is unable to find a grid that will evolve to the
 current layout correctly. This can be seen with the sample grid
-[t0_3.dat](t0_3.dat), which evolves backwards by only two time-steps.
+[t0_3.dat](%%REPO_URL%%/2011/blakely/t0_3.dat), which evolves backwards by only two time-steps.
 ### Input file format
@@ -428,7 +428,7 @@ line option.</p>
 program. In the reverse direction, the code can either be killed, or
 will stop if it is unable to find a grid that will evolve to the
 current layout correctly. This can be seen with the sample grid
-<a href="t0_3.dat">t0_3.dat</a>, which evolves backwards by only two time-steps.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/blakely/t0_3.dat">t0_3.dat</a>, which evolves backwards by only two time-steps.</p>
 <h3 id="input-file-format">Input file format</h3>
 <p>The input files must encode a square grid, where a space is a dead cell, and any
 other character (except a newline) is a living cell. Each line must be the same
@@ -228,11 +228,11 @@ The learning rate is hard-coded as `0.3`. No momentum factor is used.
 #### png corpora
-The [png-1](dlowe-aux-data/png-1) corpus was obtained by manually scraping the first results from an
+The `dlowe-aux-data/png-1` corpus was obtained by manually scraping the first results from an
 <https://images.google.com> search for "[obfuscate
 filetype:png](https://www.google.com/search?tbm=isch&sxsrf=APwXEddqIU34dxhpFdjEa8emNJKHOkYXIw:1683049158657&source=hp&biw=1727&bih=952&ei=xkpRZLapJtLKkPIP8bGKkAk&iflsig=AOEireoAAAAAZFFY1nbmEUHC74Cmg3Y5rEY1fV7ocASZ&ved=0ahUKEwi2pPXZltf-AhVSJUQIHfGYApIQ4dUDCAY&uact=5&oq=obfuscate+filetype:png&gs_lcp=CgNpbWcQA1AAWABg9QNoAHAAeACAATuIATuSAQExmAEAoAECoAEBqgELZ3dzLXdpei1pbWc&sclient=img&q=obfuscate&tbs=ift:png)".
-The [png-0](dlowe-aux-data/png-0) corpus was obtained by manually scraping the first results from an
+The `dlowe-aux-data/png-0` corpus was obtained by manually scraping the first results from an
 <https://images.google.com> search for "[obfuscate
 filetype:gif](https://www.google.com/search?tbm=isch&ved=2ahUKEwiowo7fltf-AhXSOUQIHbJBDp8Q2-cCegQIABAA&oq=obfuscate+filetype:gif&gs_lcp=CgNpbWcQAzoECCMQJzoHCAAQigUQQzoFCAAQgARQ6AxY6AxgpBJoAHAAeACAAUeIAcsBkgEBM5gBAKABAaoBC2d3cy13aXotaW1nwAEB&sclient=img&ei=0UpRZOiVINLzkPIPsoO5-Ak&bih=952&biw=1727&q=obfuscate&tbs=ift:gif)".
@@ -242,7 +242,7 @@ filetype:gif](https://www.google.com/search?tbm=isch&ved=2ahUKEwiowo7fltf-AhXSOU
 * Using these corpora as a training set;
 * Using several hundred random `*.gif` and `*.png` files from my home computer as
  a test set;
-* Using the included [earlystop.pl](earlystop.pl) script, which stopped after 67000 iterations.
+* Using the included [earlystop.pl](%%REPO_URL%%/2011/dlowe/earlystop.pl) script, which stopped after 67000 iterations.
 * Interpreting output of `> 0.5` as "probably a `.png`" and `< 0.5` as "probably
  a `.gif`"
@@ -303,7 +303,7 @@ representation and [endianness](https://en.wikipedia.org/wiki/Endianness).
 Making sure not to
 [overfit](https://en.wikipedia.org/wiki/Overfitting#Machine_learning) the network to the training data is a bit of a
-black art. I have enclosed [earlystop.pl](earlystop.pl), a wrapper script that implements
+black art. I have enclosed [earlystop.pl](%%REPO_URL%%/2011/dlowe/earlystop.pl), a wrapper script that implements
 a simple 'early stopping' algorithm; other techniques are possible.
 Bad input (e.g. nonexistent files, non-numeric number of iterations, etc.)
@@ -501,10 +501,10 @@ assigned a target value of <code>0</code>. Intervening directories (if any) will
 intermediate target values.</p>
 <p>The learning rate is hard-coded as <code>0.3</code>. No momentum factor is used.</p>
 <h4 id="png-corpora">png corpora</h4>
-<p>The <a href="dlowe-aux-data/png-1">png-1</a> corpus was obtained by manually scraping the first results from an
+<p>The <code>dlowe-aux-data/png-1</code> corpus was obtained by manually scraping the first results from an
 <a href="https://images.google.com" class="uri">https://images.google.com</a> search for “<a href="https://www.google.com/search?tbm=isch&amp;sxsrf=APwXEddqIU34dxhpFdjEa8emNJKHOkYXIw:1683049158657&amp;source=hp&amp;biw=1727&amp;bih=952&amp;ei=xkpRZLapJtLKkPIP8bGKkAk&amp;iflsig=AOEireoAAAAAZFFY1nbmEUHC74Cmg3Y5rEY1fV7ocASZ&amp;ved=0ahUKEwi2pPXZltf-AhVSJUQIHfGYApIQ4dUDCAY&amp;uact=5&amp;oq=obfuscate+filetype:png&amp;gs_lcp=CgNpbWcQA1AAWABg9QNoAHAAeACAATuIATuSAQExmAEAoAECoAEBqgELZ3dzLXdpei1pbWc&amp;sclient=img&amp;q=obfuscate&amp;tbs=ift:png">obfuscate
 filetype:png</a>”.</p>
-<p>The <a href="dlowe-aux-data/png-0">png-0</a> corpus was obtained by manually scraping the first results from an
+<p>The <code>dlowe-aux-data/png-0</code> corpus was obtained by manually scraping the first results from an
 <a href="https://images.google.com" class="uri">https://images.google.com</a> search for “<a href="https://www.google.com/search?tbm=isch&amp;ved=2ahUKEwiowo7fltf-AhXSOUQIHbJBDp8Q2-cCegQIABAA&amp;oq=obfuscate+filetype:gif&amp;gs_lcp=CgNpbWcQAzoECCMQJzoHCAAQigUQQzoFCAAQgARQ6AxY6AxgpBJoAHAAeACAAUeIAcsBkgEBM5gBAKABAaoBC2d3cy13aXotaW1nwAEB&amp;sclient=img&amp;ei=0UpRZOiVINLzkPIPsoO5-Ak&amp;bih=952&amp;biw=1727&amp;q=obfuscate&amp;tbs=ift:gif">obfuscate
 filetype:gif</a>”.</p>
 <h5 id="results">Results:</h5>
@@ -512,7 +512,7 @@ filetype:gif</a>”.</p>
 <li>Using these corpora as a training set;</li>
 <li>Using several hundred random <code>*.gif</code> and <code>*.png</code> files from my home computer as
 a test set;</li>
-<li>Using the included <a href="earlystop.pl">earlystop.pl</a> script, which stopped after 67000 iterations.</li>
+<li>Using the included <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/dlowe/earlystop.pl">earlystop.pl</a> script, which stopped after 67000 iterations.</li>
 <li>Interpreting output of <code>&gt; 0.5</code> as “probably a <code>.png</code>” and <code>&lt; 0.5</code> as “probably
 a <code>.gif</code>”</li>
 </ul>
@@ -554,7 +554,7 @@ figuring out how to concatenate directory and file names…).</p>
 representation and <a href="https://en.wikipedia.org/wiki/Endianness">endianness</a>.</p>
 <p>Making sure not to
 <a href="https://en.wikipedia.org/wiki/Overfitting#Machine_learning">overfit</a> the network to the training data is a bit of a
-black art. I have enclosed <a href="earlystop.pl">earlystop.pl</a>, a wrapper script that implements
+black art. I have enclosed <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/dlowe/earlystop.pl">earlystop.pl</a>, a wrapper script that implements
 a simple ‘early stopping’ algorithm; other techniques are possible.</p>
 <p>Bad input (e.g. nonexistent files, non-numeric number of iterations, etc.)
 tends to result in empty output.</p>
@@ -159,7 +159,7 @@ cannot execute deeper recursions).
 First of all, as you see, my program looks like a nonogram puzzle which results
 in a [dragon](https://www.draconika.com/what). This puzzle is
-[dragon.nono](dragon.nono) in this directory. Note that whitespaces in character
+[dragon.nono](%%REPO_URL%%/2011/hamaji/dragon.nono) in this directory. Note that whitespaces in character
 literals, a string literal, and around the last comment are considered as a
 boxed cell.
@@ -214,7 +214,7 @@ The second author wrote the picture, and the first author did
 everything else.
 The [dragon.nono](%%REPO_URL%%/2011/hamaji/dragon.nono), [samurai.nono](%%REPO_URL%%/2011/hamaji/samurai.nono), and
-[penguin.nono](penguin.nono) were written by the
+[penguin.nono](%%REPO_URL%%/2011/hamaji/penguin.nono) were written by the
 second author.  The [soccer.nono](%%REPO_URL%%/2011/hamaji/soccer.nono) was from a wikipedia entry.  This file
 is important because it requires some guesses.  The [random.nono](%%REPO_URL%%/2011/hamaji/random.nono) file
 was randomly generated. The [codegolf.nono](%%REPO_URL%%/2011/hamaji/codegolf.nono) and
@@ -484,7 +484,7 @@ cannot execute deeper recursions).</p>
 <h3 id="obfuscation">Obfuscation</h3>
 <p>First of all, as you see, my program looks like a nonogram puzzle which results
 in a <a href="https://www.draconika.com/what">dragon</a>. This puzzle is
-<a href="dragon.nono">dragon.nono</a> in this directory. Note that whitespaces in character
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/dragon.nono">dragon.nono</a> in this directory. Note that whitespaces in character
 literals, a string literal, and around the last comment are considered as a
 boxed cell.</p>
 <p>My program is based on bit operations for speed. Many values are
@@ -518,7 +518,7 @@ dragon</a>.</p>
 <p>The second author wrote the picture, and the first author did
 everything else.</p>
 <p>The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/dragon.nono">dragon.nono</a>, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/samurai.nono">samurai.nono</a>, and
-<a href="penguin.nono">penguin.nono</a> were written by the
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/penguin.nono">penguin.nono</a> were written by the
 second author. The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/soccer.nono">soccer.nono</a> was from a wikipedia entry. This file
 is important because it requires some guesses. The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/random.nono">random.nono</a> file
 was randomly generated. The <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/hamaji/codegolf.nono">codegolf.nono</a> and
@@ -8,7 +8,7 @@ If you have X11 installed:
 There are two alternate versions, [one that lets you reconfigure the
 controls and size](%%REPO_URL%%/2011/toledo/toledo.alt.c) and [another that should work for
-Windows](toledo.alt2.c). See [Alternate code](#alternate-code) below.
+Windows](%%REPO_URL%%/2011/toledo/toledo.alt2.c). See [Alternate code](#alternate-code) below.
 ## To use:
@@ -381,7 +381,7 @@ Location: <a href="../../location.html#MX">MX</a> - <em>United Mexican States</e
 <p>If you have X11 installed:</p>
 <pre><code>    make</code></pre>
 <p>There are two alternate versions, <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/toledo/toledo.alt.c">one that lets you reconfigure the
-controls and size</a> and <a href="toledo.alt2.c">another that should work for
+controls and size</a> and <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2011/toledo/toledo.alt2.c">another that should work for
 Windows</a>. See <a href="#alternate-code">Alternate code</a> below.</p>
 <h2 id="to-use">To use:</h2>
 <pre><code>    ./toledo</code></pre>
@@ -102,13 +102,13 @@ etc.
 The program expects *two command line arguments*. The first is a (very long)
 string that contains the complete music font. Note that it should *not* be given
 as a file name! The default music font is encoded in the (whitespace of the)
-source code [deckmyn.c](deckmyn.c) itself. Therefore, the first argument should
+source code [deckmyn.c](%%REPO_URL%%/2012/deckmyn/deckmyn.c) itself. Therefore, the first argument should
 be `"`` ` ``cat deckmyn.c`` ` ``"`.
 The second command line argument is the music itself. This is, again, a string,
 not a file name. All music signs are entered as tokens of exactly 3 characters.
 The code is very sensitive to bad spacing! So for instance, if you use a file to
-write the music (e.g. [example\_greensleeves](example_greensleeves)), you should
+write the music (e.g. [example\_greensleeves](%%REPO_URL%%/2012/deckmyn/example_greensleeves)), you should
 keep in mind that newline is also a character!
@@ -427,12 +427,12 @@ etc.</p>
 <p>The program expects <em>two command line arguments</em>. The first is a (very long)
 string that contains the complete music font. Note that it should <em>not</em> be given
 as a file name! The default music font is encoded in the (whitespace of the)
-source code <a href="deckmyn.c">deckmyn.c</a> itself. Therefore, the first argument should
+source code <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2012/deckmyn/deckmyn.c">deckmyn.c</a> itself. Therefore, the first argument should
 be <code>"``</code> <code>cat deckmyn.c</code> <code>``"</code>.</p>
 <p>The second command line argument is the music itself. This is, again, a string,
 not a file name. All music signs are entered as tokens of exactly 3 characters.
 The code is very sensitive to bad spacing! So for instance, if you use a file to
-write the music (e.g. <a href="example_greensleeves">example_greensleeves</a>), you should
+write the music (e.g. <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2012/deckmyn/example_greensleeves">example_greensleeves</a>), you should
 keep in mind that newline is also a character!</p>
 <h3 id="internals">Internals</h3>
 <p>To minimise the memory footprint, this program has <em>no variable declarations at
@@ -512,7 +512,7 @@ and detract from its single minded focus on conciseness.
 It implicitly declares functions `read(2)`, `write(2)`, `exit(3)` and
 `calloc(3)`, the latter two incompatibly.  32 bit and 64 bit executables are
-separate [Makefile](Makefile) targets, involving a change from `int` to `long`
+separate [Makefile](%%REPO_URL%%/2012/tromp/Makefile) targets, involving a change from `int` to `long`
 and from a hardcoded sizeof of 4 to 8.
 The program has been tested to work correctly on Linux/Solaris/MacOSX both in 32
@@ -715,7 +715,7 @@ many warnings;</p>
 and detract from its single minded focus on conciseness.</p>
 <p>It implicitly declares functions <code>read(2)</code>, <code>write(2)</code>, <code>exit(3)</code> and
 <code>calloc(3)</code>, the latter two incompatibly. 32 bit and 64 bit executables are
-separate <a href="Makefile">Makefile</a> targets, involving a change from <code>int</code> to <code>long</code>
+separate <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2012/tromp/Makefile">Makefile</a> targets, involving a change from <code>int</code> to <code>long</code>
 and from a hardcoded sizeof of 4 to 8.</p>
 <p>The program has been tested to work correctly on Linux/Solaris/MacOSX both in 32
 and 64 bits.</p>
@@ -333,7 +333,7 @@ rate in which the terminal application receives and displays the [ANSI
 escape](https://en.wikipedia.org/wiki/ANSI_escape_code) sequences.
 To slow down the output programmatically, introduce a delay immediately after the
-call to `fflush(3)` (N.B: see [birken.alt.c](birken.alt.c)).
+call to `fflush(3)` (N.B: see [birken.alt.c](%%REPO_URL%%/2013/birken/birken.alt.c)).
 ### IOCCC Size Tool Bug
@@ -600,7 +600,7 @@ represented using ANSI escape sequences.</p>
 rate in which the terminal application receives and displays the <a href="https://en.wikipedia.org/wiki/ANSI_escape_code">ANSI
 escape</a> sequences.</p>
 <p>To slow down the output programmatically, introduce a delay immediately after the
-call to <code>fflush(3)</code> (N.B: see <a href="birken.alt.c">birken.alt.c</a>).</p>
+call to <code>fflush(3)</code> (N.B: see <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/birken/birken.alt.c">birken.alt.c</a>).</p>
 <h3 id="ioccc-size-tool-bug">IOCCC Size Tool Bug</h3>
 <p>This program demonstrates how to exploit a bug in IOCCC size tool version
 <code>2013-07-30-v17</code>. The first line of the program is reproduced below.</p>
@@ -112,7 +112,7 @@ little endian machine.
 ### RULE 2 ABUSE DISCLAIMER
- [cable3.c](cable3.c) is 4043 bytes in length (half an 8086).
+- [cable3.c](%%REPO_URL%%/2013/cable3/cable3.c) is 4043 bytes in length (half an 8086).
 - `iocccsize -i < cable3.c` returns 1977 (the year the 4.77 MHz 8086 CPU was announced)
 - Therefore, any suspicions the judges may have regarding rule 2 non-compliance
 might be well-intentioned but are groundless.
@@ -445,7 +445,7 @@ and somewhat useless beast. For everyone else, please run the emulator on a
 little endian machine.</p>
 <h3 id="rule-2-abuse-disclaimer">RULE 2 ABUSE DISCLAIMER</h3>
 <ul>
-<li><a href="cable3.c">cable3.c</a> is 4043 bytes in length (half an 8086).</li>
+<li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/cable3/cable3.c">cable3.c</a> is 4043 bytes in length (half an 8086).</li>
 <li><code>iocccsize -i &lt; cable3.c</code> returns 1977 (the year the 4.77 MHz 8086 CPU was announced)</li>
 <li>Therefore, any suspicions the judges may have regarding rule 2 non-compliance
 might be well-intentioned but are groundless.</li>
@@ -88,7 +88,7 @@ and with Apple clang version 15.0.0 (clang-1500.0.40.1) in 2023, we get:
 Which one is correct? :)
-NOTE: `make all` will compile [fun.c](fun.c) but to provide a different compiler
+NOTE: `make all` will compile [fun.c](%%REPO_URL%%/2013/dlowe/fun.c) but to provide a different compiler
 you can do something like:
 ``` <!---sh-->
@@ -113,13 +113,13 @@ A tiny implementation of command-line
    sparkline of file sizes: ▁▁▁▃▃▂▁▂▁▁▉
 ```
-NOTE: this has been provided in [slflen.sh](slflen.sh) so you can try:
+NOTE: this has been provided in [slflen.sh](%%REPO_URL%%/2013/dlowe/slflen.sh) so you can try:
 ``` <!---sh-->
    ./slflen.sh
 ```
-instead (which [try.sh](try.sh) also uses).
+instead (which [try.sh](%%REPO_URL%%/2013/dlowe/try.sh) also uses).
 ### Description
@@ -413,7 +413,7 @@ with one argument or when all arguments are equal?</p>
 <p>and with Apple clang version 15.0.0 (clang-1500.0.40.1) in 2023, we get:</p>
 <pre><code>    1840985120 -2033041452 35979112</code></pre>
 <p>Which one is correct? :)</p>
-<p>NOTE: <code>make all</code> will compile <a href="fun.c">fun.c</a> but to provide a different compiler
+<p>NOTE: <code>make all</code> will compile <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/dlowe/fun.c">fun.c</a> but to provide a different compiler
 you can do something like:</p>
 <pre><code>    make CC=clang fun</code></pre>
 <h2 id="authors-remarks">Author’s remarks:</h2>
@@ -426,9 +426,9 @@ you can do something like:</p>
    $ echo &quot;sparkline of file lengths: $(wc -c * | awk &#39;{print $1}&#39; | xargs ./sparkl)&quot;
    sparkline of file sizes: ▁▁▁▃▃▂▁▂▁▁▉</code></pre>
-<p>NOTE: this has been provided in <a href="slflen.sh">slflen.sh</a> so you can try:</p>
+<p>NOTE: this has been provided in <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/dlowe/slflen.sh">slflen.sh</a> so you can try:</p>
 <pre><code>    ./slflen.sh</code></pre>
-<p>instead (which <a href="try.sh">try.sh</a> also uses).</p>
+<p>instead (which <a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/dlowe/try.sh">try.sh</a> also uses).</p>
 <h3 id="description">Description</h3>
 <p>This is a handy little tool for visualizing numeric series from the
 command line, using sparklines. Pass it a numeric series as arguments, and
@@ -142,7 +142,7 @@ So, try:
    ./endoh4 < endoh4.c # or ./endoh4.sh
 ```
-The [solids/](solids/) directory includes various solid data:
+The `solids` directory includes various solid data:
 - [Platonic solids](http://en.wikipedia.org/wiki/Platonic_solid)
 - [Archimedean solids](http://en.wikipedia.org/wiki/Archimedean_solid)
@@ -451,7 +451,7 @@ clang, and tcc. It should not trigger warnings with <code>-pedantic</code> and <
 <p>The shape of this code is the geometric net of a regular tetrahedron.
 So, try:</p>
 <pre><code>    ./endoh4 &lt; endoh4.c # or ./endoh4.sh</code></pre>
-<p>The <a href="solids/">solids/</a> directory includes various solid data:</p>
+<p>The <code>solids</code> directory includes various solid data:</p>
 <ul>
 <li><a href="http://en.wikipedia.org/wiki/Platonic_solid">Platonic solids</a></li>
 <li><a href="http://en.wikipedia.org/wiki/Archimedean_solid">Archimedean solids</a></li>
@@ -1,4 +1,4 @@
-## To build:
+%%REPO_URL%%/## To build:
 ``` <!---sh-->
    make all
@@ -137,7 +137,7 @@ don't have to worry about conflicts.
 * [hou.c](%%REPO_URL%%/2013/hou/hou.c) uses compression to get around the size limit. The compression
 fully complies with the rules and the guidelines (at least the portion that
 shows up in `grep` "size limit"). Please see the [Self-imposed
-restrictions](self-imposed-restrictions) section below for more details.
+restrictions](#self-imposed-restrictions) section below for more details.
 * `hou` does not terminate (as suggested by the second line of rule 6).
@@ -587,7 +587,7 @@ editing program, you need to convert the RGB values from 0..255 to 0..1, and
 then put here its <em>square</em>. Only after squaring you get your real, lively color
 in the output image, as opposed to something dull and gray. You need to do the
 same if you use a RGB texture. And yes, that is the what the second <code>sqrt(3)</code> in
-<a href="../a.c">a.c</a> is for.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/hou/a.c">a.c</a> is for.</p>
 <p>Going beyond colors, to define a shiny or dull surface one needs more maths, and
 optics. The jargon for this is <code>BRDF</code> (Bidirectional Reflectance Distribution
 Function) importance sampling. Let’s continue to function <code>o</code>.</p>
@@ -312,7 +312,7 @@ editing program, you need to convert the RGB values from 0..255 to 0..1, and
 then put here its *square*. Only after squaring you get your real, lively color
 in the output image, as opposed to something dull and gray. You need to do the
 same if you use a RGB texture. And yes, that is the what the second `sqrt(3)` in
-[a.c](../a.c) is for.
+[a.c](%%REPO_URL%%/2013/hou/a.c) is for.
 Going beyond colors, to define a shiny or dull surface one needs more maths, and
 optics. The jargon for this is `BRDF` (Bidirectional Reflectance Distribution
@@ -377,7 +377,7 @@ Location: <a href="../../location.html#CN">CN</a> - <em>People’s Republic of C
 <!-- START: this line starts content for HTML phase 21 by: bin/pandoc-wrapper.sh via bin/md2html.sh -->
 <!-- BEFORE: 1st line of markdown file: 2013/hou/README.md -->
-<h2 id="to-build">To build:</h2>
+<p>https://github.com/ioccc-src/temp-test-ioccc/blob/master/## To build:</p>
 <pre><code>    make all</code></pre>
 <h3 id="bugs-and-misfeatures">Bugs and (Mis)features:</h3>
 <p>The current status of this entry is:</p>
@@ -468,7 +468,7 @@ don’t have to worry about conflicts.</li>
 <ul>
 <li><a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2013/hou/hou.c">hou.c</a> uses compression to get around the size limit. The compression
 fully complies with the rules and the guidelines (at least the portion that
-shows up in <code>grep</code> “size limit”). Please see the <a href="self-imposed-restrictions">Self-imposed
+shows up in <code>grep</code> “size limit”). Please see the <a href="#self-imposed-restrictions">Self-imposed
 restrictions</a> section below for more details.</li>
 <li><code>hou</code> does not terminate (as suggested by the second line of rule 6).</li>
 </ul>
@@ -60,7 +60,7 @@ Escape to quit.)
 Unfortunately, it became very difficult to get this entry to fit within the
 size requirements. It's currently at 2052 according to
-[iocccsize](../iocccsize.c), and 3994 according to `wc -c`.
+[iocccsize](%%REPO_URL%%/2014/iocccsize.c), and 3994 according to `wc -c`.
 The current size is a third to a half of the original size before compression.
@@ -414,7 +414,7 @@ Escape to quit.)</p>
 <h3 id="known-compiler-warnings">KNOWN COMPILER WARNINGS</h3>
 <p>Unfortunately, it became very difficult to get this entry to fit within the
 size requirements. It’s currently at 2052 according to
-<a href="../iocccsize.c">iocccsize</a>, and 3994 according to <code>wc -c</code>.</p>
+<a href="https://github.com/ioccc-src/temp-test-ioccc/blob/master/2014/iocccsize.c">iocccsize</a>, and 3994 according to <code>wc -c</code>.</p>
 <p>The current size is a third to a half of the original size before compression.</p>
 <p>Due to this, it does LOTS of things that subtly annoy compilers!</p>
 <ul>