I stalled out with the writing. I stalled out right on the precipice of driving home the most important piece of coding… of expressing myself in a way that has an automating and thus self-accelerating component… that I have ever done.
Briefly, in my time at a company from my past I was firing on all cylinders like this and made a system that lifted the company up by the scruff of its neck and took it into the future. Or at least ensured it’s survival long enough to be purchased by another company for more than just picking it apart for pieces.
This go-around, it’s riding the wave of the anti-ReactJS movement, a move towards greater simplifcation in web development systems, but for the purpose of making powerful localhost apps, LLMs and all, which exceed the capabilities of things you can do when hobbled by enterprise architecture and scaling concerns. No app is an island. But if you design it so that localhost is an island, a whole host of tricks open up to you that just aren’t available otherwise. That stuff is covered in depth other places, so this is just a reminder to myself about what I’m doing here.
My last post, which was supposed to be about the driving the Pipeline Process to Completion got derailed by a combination of my own exhaustion catching up with me (I’ve been driving myself very hard), and the client work having still to take priority. And it’s not just easily planned around client work. It’s additional stuff being tossed my way because… well, because life. So again, it’s my failure to factor in Murphy’s Law and default FUBAR states. The world ain’t ideal. That’s why I start out with:
- What’s most broken
- Where do you get the biggest bang for the buck
- What plates need to be spun
One of the nuances here that I learned is that the nature of the way you spin those plates is ever-so important. You either spin them in such a way that frees up cycles between re-spins or in a way that tightens the gap and you are left doing nothing but spinning plates at the expense of ever fixing what’s most broken or doing any of those things where you get the biggest bang for the buck. That’s the busywork trap, and is clearly what I’m trying to avoid getting trapped in.
If I do, it’s game over.
And one can almost view themselves in a cat and mouse game of those forces collaborating unintentionally, merely fate being what it is, against you versus the you that was put into this world to achieve your Ikigai. Something is preventing you from producing what the world needs as a result of doing what you are good at, what you love doing and what you can get paid for.
This I think it feels in this world, and I could be wrong and would like to be proven so, that so few people achieve anything like their Ikigai. Most folks die on some alter of compromise because all pursuit of Ikigai has these anvils falling on your head as if dropped by Wile E. Coyote on the Roadrunner. Of course, none of them ever hit the Roadrunner, and there’s a lot strategic and tactical inspiration in that. Be fast. Know what you’re about. And have a great sense of humor regarding those anvils dropping all around you. Know them for what they are and don’t let them change that MEEP MEEP spirit.
Okay, so it’s really about client work today.
But it’s also really about this tiny last step that makes the client work the first pipeline. It’s a RED LIGHT / GREEN LIGHT warning system for what’s out of whack in your website.
I mean how perfect is that?
I also have to produce some of those Link-Graph diagrams that provide insight about a website, but that’s going to be a big pipeline porting process, which I have the next 4 days as beautifully focused, concentrated time for.
These 4 days of concentrated work time is something I truly have to be thankful for. The timing could not be more perfect.
Where I lost steam and left off was setting the stage for my most difficult next step with a clarity, and therefore confidence, I need to keep the code from getting out of my control.
I’m using those effin code assistant “genies” to help, and I am therefore overreaching my innate technical abilities. Again, there’s that concept of hitting above my weight class. I don’t have the kind of pulsing alien brain that people like Paul Graham or David Heinemeier Hansson seem to have. I use those two examples because they write influentially about this stuff, Paul about the meta programming aspects of LISP, and DHH in bucking the ReactJS trend. Of late, I’d probably throw Jeremy Howard into this category. I’ve created incredibly junior entry-level versions of some of the admirable things they’ve created. And now that I’m up to my next project, my intention is to hit at one weight class up. I’m beefing up with the code assistant genies.
This is where I left off 2 days ago with perfect clarity of thought:
Claude: I notice several key “hot spots” where implementation details need to crystallize, while maintaining the elegantly simple design principles you’ve outlined:
- Card State Serialization Pattern
# Not yet implemented but needs consistent pattern:
card_state = {
"card1": {
"url": "https://example.com" # Single field example
},
"card2": {
"option1": "selected_value",
"option2": ["multiple", "selections"] # Multi-field example
}
}
- Pipeline State Transitions
# Current placeholder locations in WorkFlow class:
async def process_url(self, request):
# Need to standardize how we:
# 1. Validate incoming state
# 2. Update pipeline JSON blob
# 3. Return only placeholder for next card
- Card Placeholder Generation
# Need consistent pattern for all cards:
return Card(id="card2") # Currently ad-hoc
# Should be standardized helper method:
def create_placeholder(self, card_id): ...
- State Recovery Points
# Pipeline table schema already supports this:
{
"url": str, # Primary key
"data": str, # JSON blob
"created": str,
"updated": str
}
# But need consistent recovery implementation
Navigating Priorities and Presentations
But the work I do today for TWO DIFFERENT Monday presentations, I have to be able to fake having what the automation process will produce. Two things, which themselves must fit into even higher-priority projects coming down from… from… from… Perplexity?
Perplexity: The expression “coming down from on high” typically refers to orders, directives, or decisions that originate from people in positions of authority or power within an organization or hierarchy[1][3]. It is often used humorously or ironically to describe instructions or mandates that are imposed by senior management or leadership without input from lower-level employees[2][5].
For example:
- “An order came down from on high that lunch breaks were to be reduced to 30 minutes.”
- “The new policy came down from on high, leaving employees frustrated with the lack of consultation.”
The phrase combines two idioms:
- “From on high” - meaning from a position of authority or power[5]
- “Coming down” - implying the top-down nature of the communication[4]
This expression emphasizes the distance between decision-makers and those affected by the decisions, often suggesting a lack of understanding or consideration for ground-level realities[3][5].
Citations:
[1] https://preply.com/en/question/what-does-to-bring-hell-down-from-on-high-mean-65724
[2] https://www.ldoceonline.com/dictionary/from-on-high
[3] https://idioms.thefreedictionary.com/on+high
[4] https://www.reddit.com/r/EnglishLearning/comments/yellvh/what_is_the_meaning_of_come_down_in_this_context/
[5] https://idioms.thefreedictionary.com/from+on+high
[6] https://dictionary.cambridge.org/us/dictionary/english/on-high
Balancing Priorities and Performance
Yeah, so there you go. I have to be able to report on progress on my main on high directive which I so agree with as needing to be the company priority, because without being able to shove more work through a smaller pipeline we satisfy fewer customers to higher degrees. There’s a whale-hunting versus not-whale-hunting discussion here that I’m trying to nullify or make moot by having blurring the line between harpooning and trolling. Can you still use whale murdering metaphors? Well, I’ll fry that fish when I kill 2 birds with one stone.
The Art of Simultaneous Problem-Solving
Anyhoo, it’s a complex multidimensional puzzle. I’m solving multiple equations simultaneously, but I must do that without slowing myself down, haha! And that means brushing off a little PT Barnum Penn & Teller showmanship. It’s not about the miracle being accurately performed, it’s about the perception of the miracle being accurately performed, and in my case covering for just enough time to accurately perform the miracle.
The Importance of Authentic Data
But the data needs to be good. The trick needs to be truly and accurately customized to each client. When they look at the Google Slide, because unfortunately that’s where the trick needs to be performed, the client needs to see their own special case screaming back out to them saying “These people really look at and know your site better than you do.”
Career Reflections and Scientific Achievement
And that is my today. It’s not the next and wonderful and life-changing step of completing a piece of work, which at in my mid-50s has got to be my Magnum Opus and the output of my optimum stopping distance algorithm. Whereas most people would have flipped the switch by 28 going into their midlife crisis… what’s that quote about physicists all having done all their greatest work by 26 years old or not at all… Perplexity?
Perplexity: Einstein once said, “A person who has not made his great contribution to science before the age of 30 will never do so.”[1] However, a recent study has shown that this view is increasingly outdated. Researchers found that while young scientists did make groundbreaking contributions in the early 20th century, particularly during the quantum mechanics revolution, the age of peak scientific creativity has significantly shifted[1][2].
By the year 2000, great scientific work before age 30 had become extremely rare. In physics, the average age for Nobel Prize-winning work is now 48, compared to the earlier era when many physicists made critical breakthroughs in their 20s. The study suggests this change is due to the increasing complexity of scientific knowledge, with scientists now requiring more time to build upon existing foundational research[1][2].
Interestingly, the early 20th century was unique. In 1923, about 31 percent of physicists did their breakthrough work by age 30, and by 1934, 78 percent did so by age 40. However, these percentages declined throughout the rest of the century[1].
Citations:
[1] https://www.nbcnews.com/id/wbna45198217
[2] https://www.telegraphindia.com/india/sorry-albert-40-is-new-30-study-finds-einstein-s-age-rule-needn-t-hold-true-now/cid/323889
[3] https://www.asl-associates.com/einsteinquotes.htm
[4] https://www.easybib.com/guides/quotes-facts-stats/albert-einstein/
[5] https://www.physicsforums.com/threads/greatest-work-of-physicist-done-in-their-20s.666255/
[6] https://www.deseret.com/2018/3/14/20641870/22-witty-and-inspiring-albert-einstein-quotes/
Breaking the Sisyphus Cycle
…so in my mid-50s I’m basically an old coot who thinks he’s a tyke. So effin be it. Keeps me young and all that. The trick is to not remain Sisyphus, but rather to have a moment of self-awareness and a cycle-breaking agency moment. Know why it’s a not-so-virtuous cycle that keeps you rolling the boulder up the mountain, create an anchor-and-lever system that can both:
- Give you a momentary respite
- Allow you to bank your next bit of progress, once recovered
Strategic Progress and Automation
So these couple of slides I’m making today are an image of the output of this system, only enough so to give me that momentary respite… today and today only. Take consolation in the fact that you can make major boulder-rolling progress starting tomorrow in a way that can never be taken back from you again.
I am indeed analogous to Sisyphus figuring out how to automate his boulder-rolling task. It’s not even automation. It’s repeatable workflows where you can leave of and pick up in the middle of, where the steps of the workflow might be human or might be machine. Sometimes I’ll hit an API and do some Pandas data tricks, and sometimes I’ll pop open a browser window using all the user’s local Botify and SEMRush stored profile data to end-run authentication and APIs with a bit of instruction about what to click-to-save and where to save it. Semi-automation!
The Power of Metaphor
I like this Sisyphus metaphor so much it’s going to replace my Icarus jumping off the cliff and learning to fly metaphor, which is what I used previously for those hitting above your weight class moments. Too dangerous! Too reckless! Sisyphus has been at it awhile and is the definition of stupidity, right? Doing things over and over the same way, the hard and wrong way without learning from your experiences, means you lack intelligence. Sure, Sisyphus was cursed to his fate, but it seems to me he was locked into acting stupid, and with a nice dose of self-awareness, agency, and some strategic and tactile thinking, he can break free of the curse. Am I right?
Okay, today’s post officially becomes De-dumbing Sisyphus.
Seizing Opportunities
Even today, I’ve got a bit of wiggle room to work with. Thanksgiving is tomorrow, and then Friday off and a blissful weekend. Life circumstances being what they are at the moment, this means the introvert’s dream of 4 solid days of focus! Not merely that, but 4 solid days of focus with next step spelled out crystal clear to achieve greater than I ever have in my professional career, by virtue of the complete re-alignment with the evolving world of AI SEO. This is big. As big as anything in my life second only to having a child, for whom doing right by this career elevation step is all about.
Personal Growth and Motivation
It’s not easy to communicate these things, but it needs to be said. My young-person energy in my 20s which was all about seeing if I could perform at that next level up, becomes in my 50s about trying to hit it all home in a way that I can open up options for them in their life. Sure, I’ll enjoy these second-act vibes and the whole Rocky Balboa bouncing back and rising to the occasion feeling. But it’s also about having a reason to do it. We still have tons of time to work with here. What doesn’t kill you makes you stronger. I know there’s a generation who thinks what doesn’t kill you makes you damaged goods wronged by society and permanently entitled. But to that I say “I shall taunt the fragile flower generation a second time-a!”
The Concept of Anti-Fragility
Anti-fragile. Thank you Taleb-something for writing that book that encapsulates why the fragile flower generation of algorthmically riled-angry YouTubians is so wrong-headed and funny and why “what doesn’t kill you makes you stronger” is spot-on right. What was his name?
Perplexity: Nassim Nicholas Taleb wrote the book “Antifragile: Things That Gain from Disorder”[1][2]. This book is part of Taleb’s Incerto series, which explores themes of uncertainty, probability, and decision-making in a complex world[2][3].
“Antifragile” introduces and expands on the concept of antifragility, which refers to things that not only withstand chaos and volatility but actually benefit from it[2]. Taleb argues that many aspects of life, from biological systems to economic structures, can exhibit antifragile properties[1].
The book covers a wide range of topics, including:
- How certain systems gain strength from stress and disorder
- The limitations of predictive models in complex environments
- The importance of optionality and the “barbell strategy” in decision-making
- Critiques of modern approaches to risk management and policy-making
- Examples of antifragility in various fields such as medicine, economics, and urban planning[1][2]
Taleb’s writing style in “Antifragile” combines narrative elements, philosophical discussions, and scientific commentary to present his ideas[3]. The book aims to provide a framework for thriving in an uncertain world by embracing and harnessing the power of volatility and randomness[2].
Citations:
[1] https://psychiatryresource.com/bookreviews/antifragile-review
[2] https://www.barnesandnoble.com/w/antifragile-nassim-nicholas-taleb/1110867836
[3] https://en.wikipedia.org/wiki/Nassim_taleb
[4] https://www.goodreads.com/book/show/13530973-antifragile
And let’s follow up with a nice “this too shall pass” origins story…
The Origins of “This Too Shall Pass”
Perplexity: “This too shall pass” is a profound philosophical phrase that reflects the transient nature of life, emphasizing that both good and bad moments are temporary. The phrase originates from Persian literature, with the original Persian saying “این نیز بگذرد” (in nīz bogzarad)[1].
Historical Context
The expression gained widespread popularity through a legendary story involving King Solomon. According to the tale, Solomon sought a ring that would make a sad man happy and a happy man sad, ultimately inscribing the phrase “Gam ze ya’avor” (Hebrew for “This too shall pass”) as a reminder of life’s impermanence[1].
Cultural Significance
The phrase carries deep philosophical meaning:
- It reminds people that no emotional state is permanent
- Provides comfort during difficult times
- Encourages resilience and perspective
Notable Popularization
Abraham Lincoln famously referenced the phrase in an 1859 speech, describing it as a profound wisdom that captures both the humbling nature of pride and the consolation in times of affliction[3][4]. Lincoln, who struggled with depression, personally understood the phrase’s transformative power of hope.
The saying transcends cultural boundaries, appearing in Persian, Jewish, and Western traditions, ultimately serving as a universal reminder that life’s circumstances are always changing[2][6].
Citations:
[1] https://www.christianity.com/wiki/bible/this-too-shall-pass.html
[2] https://www.reddit.com/r/AskHistorians/comments/90vu05/what_is_the_origin_of_the_phrase_this_too_shall/
[3] https://yoprolibrary.substack.com/p/this-too-shall-pass-where-does-the
[4] https://humbledollar.com/2020/03/this-too-shall-pass/
[5] https://www.alefbet.com/blogs/blog/this-too-shall-pass-the-story-the-meaning-and-you
[6] https://www.ansheamunim.org/gam_zeh_yaavor
[7] https://ovlcollection.com/blogs/news/the-meaning-of-this-too-shall-pass
Preserving Deep Messages
So I am totally aware that these deeper messages are going to get lost in the noise of day-to-day work journal entires. But fear not! I’ve already done my tests to confirm that these big-token-window LLM (monster genies) like Gemini Advanced Pro can take my whole journal, every single post, and parse through it finding patterns and order, proposing the books I can “spin” out of it. Mike’s Book of Future-Proofing Tech Strategies & Tactics will certainly be one of them, but there are plenty of sub-books here that I shall ensure are extractable.
Life’s Spinning Plates
I’ll do that work for you, so you don’t have to. Just know that as one of my life’s spinning plates has reached optimal stopping distance and is being taken from the plate-spinning act and moved to the plate display mantle, other plates are being put into the rotation. The extracted stories will, I hope, have a wee bit of possible redemptive power here and there if you are willing to have an open mind about it. I’m sorry for the past, and my only power-up’s now to help me on the challenging adventure that is life is just coffee, and I’m going to go get me some now to dive into today with renewed vigor!
The Journey Continues
Fast-forward to Thanksgiving. Yesterday was another bust. While I caught up on sleep, I was totally distracted processing the whirlwind of the past few weeks (months) of bearing down on the biggest coding project of my life, re-settling myself onto FastHTML which has the same feeling of durability as HTML itself. It’s the union of timeless aspects of Python and HTML, using HTMX, a proposed new protocol and completely functioning library and technology, to extend HTML’s simple hyperlinked hypertext intent to the next logical level, by which anything can make requests and handle responses from anything in the document object model (DOM) of the browser, thereby turning the simple client/server sequential model of the original web into… into… well, I can’t even ask Perplexity anymore because their feature that lets you copy the entire markdown response citations and all (the best feature ever!) disappeared.
Oops.
Adapting to Change
So much for my favorite blogging technique of gathering Perplexity’s view on things as I go, blending in the well reasoned and well cited response. Maybe it’s time to look at ChatGPT Search as an alternative to Perplexity. Didn’t think I’d be doing that for awhile, but the removal of that key critical feature of Perplexity is a slap in the face, and hey there’s no loyalties here using any particular AI service. It’s really about bouncing around to all of them, seeing what they have to offer, using each as appropriate, and when it comes time for a real job, use the lowest-bidder approach so long as the output quality is sufficient.
Managing AI Like Subcontractors
It’s just like being a general contractor managing subcontractors. You have to have a vision of the overall job and know what sub-skills it’s going to take to get the job done. Then you look over your list of subcontractors and run the job past them to see what they could make of it. If you don’t like their response to your proposal, you move onto the next. We as general contractors of jobs to be performed by AIs are in the catbird position of squeezing these mechanical turks, these automated Fiverr’s, for the best juice. It is up to us to make sure the process of doing so doesn’t use any custom API’s, so that you can round robin between them and erode all their respective competitive moats to nothing.
The Power of Local Intelligence
In such a world, the power shifts back to those who imagine a project. Move as much of the process as you can to inside the private walls of localhost and LAN, employing your in-house local LLMs to do most of the busywork. Intelligence is like electricity. It’s not no-cost, but it is a commodity that is ubiquitous and ever present, flowing like water and infinitely multi-source-able.
Reassessing and Moving Forward
Pshwew! Okay, so my November 27th post becomes my November 28th Thanksgiving post as well, because I didn’t finish what I needed to finish yesterday, and I am altering my approach. Given this incredible opening up of a 4-day stretch, during which I most certainly have to do client work, but mock-up’s of “real things”, I step back and re-evaluate. Don’t lose any ground. Just… well, as the topic of this post now reads: de-dumb Sisyphus. That’s the criteria.
The Journey of Synthesis
And this is just the deep breath getting started. That’s what people I interact with hate to hear. Just being at the beginning and that long stretches of time are not an assurance of the completion of “Jump / How High?” style work that permeates the contracting and consulting world. Rather, when you have someone who’s realizing they’re synthesizing a lifetime of lessons into a powerful new methodology and approach, they… well, they can’t relate. Let’s keep it at that.
Fighting Feature Creep
Now for me personally, I feel I am engaged in a battle against feature creep and the very information architects who make everything too difficult with a project that is guilty of exactly what they say! Self-perpetrated hypocrisy is what I’m worried about. Then I remind myself how much of diverse different systems, from a Django to a Celery to a LangChain I’ve condensed down to under 7K lines of code, and at least half of that comments and LLM coercion to see things my way. And it’s all based on the anti-information-architecture-astronaut tech of FastHTML/HTMX, with a keen eye towards stripping out all client-side framework complications making everything observable in the server-side web server console output and logs… pshwew! Exhale… point being, I really am putting a fine-edged minimalism and transparency of process on all this which is already kick-ass in a first pass, but which over time is likely to become much smaller. The entire opposite of feature creep. Complexity excising!
The Right Track
And then I realize I am on the right track. Especially considering how I’m layering in the local LLM and the just-in-time hot prompt injection to give it Neo Kung Fu like matrix downloads of exactly what it needs to know exactly when it needs to know it, based on how the user navigates the user interface… WOW!
The Future of Web Frameworks
Nobody knows the web framework I build.
Keep much of it proprietary and behind a private GitHub repo, always offering your employer right of first refusal to planting their flag in the buzz that is likely to surround it if and when this particular Pipulate fork I’ve named Botifython is made FOSS again.
Forever-Forward Technology
I have now what I call “forever-forward” tech. From here forever after. Sure, things might change. But Linux is the pee in the pool of tech. It’s in and it’s not coming out without a total draining. And a such, there’s areas of Linux worth focusing on, because they’re the bits that really really matter. And among those is instant system creation. Instantiation of entire systems with all their file systems, databases, servers and services. The entire kitten caboodle, local area network (LAN). This is traditionally done with complex deploy scripts like Ansible, or with some variety of Docker images and complex compositing, or virtual machines (VMs), or some variety of Kubernetes, which is really all that same stuff. Sometimes with k8s as a simplified version, but all still quite complex.
The Simple Solution
The easy way to do this is with a single text file, and for that there is precisely 2 great techs: NixOS/nix which innovated the thing, and guix, the GNU Free Software Foundation’s version of it because it’s so good that the people who started the whole free and open source software revolution think it’s worth copying. And insider secret: nobody knows this!!! For lightweight everyday use with single individuals seeking a sort of digital sovereignty and perhaps a touch of locahost superpowers, nothing beats Infrastructure as Code (IaC). Honestly, nothing beats IaC for the enterprise either, but they’re not done with their Docker honeymoon yet.
Wrapping Up / Banking Wins
So, as I sip my coffee and brace myself for another day of client work and coding adventures, I find solace in the knowledge that even Sisyphus, that ultimate symbol of futility, can find a way to break free from his eternal struggle. And perhaps, with a bit of ingenuity and a healthy dose of defiance, we can all learn to “dedumb” ourselves, to escape the loops of repetitive toil, and to build a future where our efforts are not just Sisyphean struggles, but meaningful steps towards a more fulfilling and impactful existence. And hey, if I can figure out how to bank a few of those hard-won victories along the way, well, that’s just icing on the cake. (Stay tuned for the next article, where I’ll delve deeper into the art of “banking wins” and building a future-proof tech career.)