<div dir="ltr"><div dir="ltr">





<p class="gmail-p1" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">Derek</span></p>
<p class="gmail-p2" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica;min-height:17px"><span class="gmail-s1" style="font-kerning:none"></span><br></p>
<p class="gmail-p1" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">Thanks for your thoughts on specification-based programming. I had similar thoughts in a recent exercise, performing data acquisition from cryptocurrency exchanges.<span class="gmail-Apple-converted-space"> </span></span></p>
<p class="gmail-p1" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">From this exercise, I have concluded:</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s2" style="font-kerning:none"><b>1. The essential skill of the systems engineer is now to accurately specify intent. </b>E</span><span class="gmail-s1" style="font-kerning:none">ngineering is become an act of applied philosophy. A poorly defined spec is not just a technical failure; it is a failure of self-knowledge. The ancient task of "knowing thyself" is no longer just a spiritual pursuit—it is the functional prerequisite for developing the massive systems of systems that are now possible.</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none"><b>2. We need to let go of the detail. </b>We are transitioning to the point where we don't read the code generated by the AI. I didn’t.</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">Looking to the future, I can see where, given the scale of systems we are now capable of building, human involvement in specification writing may end with the concept of operations, all subordinate specs being reviewed by a gaggle of independent AIs -<span class="gmail-Apple-converted-space">  </span>without human input.</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">3. <b>If your application is not novel, it is possible to get the spec right the first (or second) time.<span class="gmail-Apple-converted-space"> </span></b><b></b></span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">I think we are approaching this reality. There is not much new under the sun. Coworkers such as Claude have seen it all before. They have the practical wisdom. Not only required features in a given app domain, but also options for implementation. Claude made me aware of essential fintech stuff I've never heard of - hash sequence logging - say what ???<span class="gmail-Apple-converted-space"> </span></span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">4. <b>We are touching the sublime.</b> Developing this app, I discovered the vast knowledge base available through Claude with the same sense of awe I once felt during a 3000 ft night descent over the glowing lights of Manhattan into New Jersey's Trenton-Mercer Airport.</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">In the presence of the sublime, I felt small but was uplifted by the infinity of knowledge now available to me.</span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">The experience had a profoundly redemptive effect. A restored sense of frictionless moving forward, where essential fintech features can be discovered and architecture options chosen in minutes, without hours of research and bugs are fixed in seconds.<span class="gmail-Apple-converted-space"> In essence, the AI pushes the masses of computer implementation detail out of the way, allowing you to focus on functionality.</span></span></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none">An enjoyable, low-anxiety development environment, partnered with a mentor pleased to answer my stupid questions with courtesy and encouragement. The absence of human failings that destroy creativity is refreshing. It has led me to a compelling thought experiment.<span class="gmail-Apple-converted-space"> </span></span></p>
<p class="gmail-p4" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica;min-height:17px"><span class="gmail-s1" style="font-kerning:none"></span><br></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none"><i>What if we stripped human failings out of systems projects, staffing them with rational, infinitely knowledgeable, courteous Claude agents? Surely productivity would go through the roof!</i></span></p>
<p class="gmail-p4" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica;min-height:17px"><span class="gmail-s1" style="font-kerning:none"><i></i></span><br></p>
<p class="gmail-p3" style="margin:0px;font-style:normal;font-variant:normal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stretch:normal;font-size:14px;line-height:normal;font-family:Helvetica"><span class="gmail-s1" style="font-kerning:none"><i>Les</i></span></p></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Fri, 19 Jun 2026 at 22:15, Derek M Jones <<a href="mailto:derek@knosof.co.uk">derek@knosof.co.uk</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Martyn,<br>
<br>
> The paper emphasises that it's not N-version programming as originally conceived, and the author makes no claims about <br>
> achieving high dependability or fault-tolerance.<br>
<br>
This paper uses the Knight-Leveson Launch Interceptor Program Specification<br>
for N-Version Programming with Coding Agents<br>
<a href="https://arxiv.org/abs/2606.20158" rel="noreferrer" target="_blank">https://arxiv.org/abs/2606.20158</a><br>
<br>
and some thoughts on specification based programming<br>
<a href="https://shape-of-code.com/2026/06/14/specification-based-programming/" rel="noreferrer" target="_blank">https://shape-of-code.com/2026/06/14/specification-based-programming/</a><br>
<br>
and a paper from earlier this week asks:<br>
"Do programming languages still matter to your AI coding agent teammate?<br>
Evidence at scale from chess engines"<br>
<a href="https://arxiv.org/abs/2606.13763" rel="noreferrer" target="_blank">https://arxiv.org/abs/2606.13763</a><br>
<br>
-- <br>
Derek M. Jones           Evidence-based software engineering<br>
blog:<a href="https://shape-of-code.com" rel="noreferrer" target="_blank">https://shape-of-code.com</a><br>
<br>
_______________________________________________<br>
The System Safety Mailing List<br>
<a href="mailto:systemsafety@TechFak.Uni-Bielefeld.DE" target="_blank">systemsafety@TechFak.Uni-Bielefeld.DE</a><br>
Manage your subscription: <a href="https://lists.techfak.uni-bielefeld.de/mailman/listinfo/systemsafety" rel="noreferrer" target="_blank">https://lists.techfak.uni-bielefeld.de/mailman/listinfo/systemsafety</a><br>
</blockquote></div></div>