You don't actually know your heroes. What you know is an image of them sold to you through media, internet, television and from all the marketing or advertising means. It's a fake persona (totally made up) tailored for you to feel a certain way about the person. The PR machinery is actively laundering people's reputations.

I could take specific examples but the truth is all your heroes are humans. Humans do right & wrong things on different levels. From any objective level, some of them are goddamn evil. And you can't get successful without entering a grey area.

Never underestimate the capacity of a human to be evil. Especially when they have a lot of power. It's very tempting to commit debauchery when you know that there are no consequences & you can get away with it.

The worship of heroes comes with the power to control masses. To take them away from the truth and make them half-headed idiots. To reduce them down to a number which serves a so called higher purpose. And the loss of individuality.

Think on your own. Evaluate from different perspectives in order to seek the truth no matter how painful the process is. Or better yet, don't follow anyone. Don't endorse anyone that you don't know personally for a very long time.

It's very important to highlight that it's our natural tendency to seek out heroes and be hopeful of the world and when the mask falls off, we only see an imperfect human. Truth be told, it is we that are idiots to even hope for something magical or special. Sometimes it's us; the people who are ready to fall for the magic trick even if they know that magic doesn't exist. So knowing that you can be fooled or rather you are very eager to be fooled is important.

Be wary of whom you admire. Because it's a mask. And you don't know them as a person but a persona!

You have seen the Simplificators. These guys say that Excel is just a GUI wrapper with some cool formulas over SQL. Or SQL is just a CLI wrapper with some constraints over text files. Well, I know nobody says that but you get the idea. You’ll quickly identify these guys when you see them.

Their basic premise is to frame an idea sans the complexity it took to create and the complexity it possesses in an ignorant way. Or worse, to put the idea down, deliberately. Basically make the idea seem banal.

Devil lies in the detail. The nuances of an idea make it what it is. And when you bring down complexity and oversimplify, you’re losing substance.

It’s like saying humans are moving water. It’s technically true but we don’t think moving water when we think humans.

The value of an entity lies in the experience of it & not the description of it. You can’t give a person a complete accurate picture of an entity using comprehension. Descriptions are the map, experience is the terrain. When you walk the terrain, you feel the minute details of it. And you can’t capture all of it in the map itself. Your description is bound to fall short of the experience.

I don’t want to get into what their act of simplification stems from but this is an argument against the act. It’s deeply insulting to compress a thing of breadth with a few sentences. It doesn’t do justice to the entity & is ignorant to say the least.

Serious thinkers know when to stop simplifying. They know when a sentence must become a paragraph, when a paragraph must become practice, and when the only honest answer is “you have to experience it.”

In cybersecurity, you need to know what to look for and where. Otherwise you're searching for a needle in a desert; unusual enough to be seen but unlikely to be found.

Humans are like horses. Straight vision. Can see in only one direction at a time.

Machines on the other hand, only look for what you tell them to. They can't decide on their own about what to look for. If they could, they'd replace humans. They can in fact replace a lot of human straight looking cases. Repetitive bureaucratic stuff should be replaced with more efficient processes that reduce the load to do the actual activity that brings out the most from an activity in an 80/20 sense. But those repetitive case handling scripts need to be written by humans.

A lot of the times, cybersecurity is just looking and searching for common patterns using common methods, tools & processes. It's very standard to be repeatable but hard to be replicated by machines.

What machines can't handle:

The volume of data matters a lot: Like I said, the machine only looks for what it has been told to look for. The human can look for any sort of weirdness. If the human has convinced themselves that there’s something weird going on in a sample, they would work by listing all the possibilities and elimination of the improbable. While on the other hand, a machine would not look way beyond. Hell if you’re using an LLM, you’re one erronous instruction, context overflow or prompt injection away from a hallucination.

Sometimes the Horse Vision becomes an important factor itself. Machines can’t & don’t ignore the noise but due to the virtue of this Horse Vision Humans do.

The specific knowledge of the professional isn't on the internet & is only gained through experience. If specifically comparing it with a generalist LLM, most of it has been trained on generic datasets. It doesn’t have the specialized knowledge to understand things the way an expert would do. They simulate the thinking process but can’t simulate the best possible experience.

Catchy hooks, generic subjects and mediocre coverage, all of these things actually get immense views to your content.

It seems like people aren’t reading to inform themselves but to conform to the most proximate theory they can wrap their heads around. Or maybe it was always like this. Internet just amplified this effect.

This is particularly true after two sortof revolutions in the content field: 1) SEO based writing 2) AI Slop Articles.

SEO based writing pushed people to write stuff that search engines pushed. Search engines worked on what was clickable stuff. So either something extreme / polarizing or something relatable, something of that sort that would get the clicks.

AI based writing pushed people to be on content cycles. Reading an AI Slop article is like talking to a mechanical robot. You feel like you’ve read something but you never actually understood anything. It’s vague, diplomatic & weird. It’s so non-personal that you can sense that an AI must have written it. Far from the conversational English & tonality, it shifts to a weird sort of ornamental but mechanical English. Apart from that, there’s no fencing of concepts. The content could be understood as anything but not distinctively something that’s distinguishable from others. AI tries to say everything at once, which results in saying nothing at all.

People aren’t writing to express. It’s all the numbers game now. They wanna get heard and get some visibility. Some want to make money and for that they’re in a way, forced to take a mathematically calculated route of SEO Optimization & AI Slop Schedule Cycled Articles. Nothing wrong about that, except, it’s not interesting in my humble opinion.

Me? I’m more interested in the old school blogging style. I don’t want to be gramatically correct. I don’t want to be smart or clever. I want to express what I’m thinking while using clear articulation.

Has it ever happened to you that you took your phone or laptop to accomplish some task & the currently opened windows put you in an unproductive blackhole?

GUIs are a mess full of recall value sans priority. The thing is we only perform better if we focus on one task at a time & GUI suggests you (subconsciously) many tasks.

Initially, I made another workspace in my Macbook but then it got just as cluttered and eventually I figured out it was the GUIs fault.

The colours are too shiny & the notifications + tabs tell you again & again that you could be doing more or better things.

I don’t use a task manager app when I want to focus because of too many colours & it’s too suggestive.

This distraction is mentioned in the book Deep Work as “Context Switch” and the effect of this is called “Attention Residue”. Both of which are associated with Shallow work.

So, I’ve made a rule now. Whenever I want to focus, I move away from GUI to CLI.

The best thing about CLI is nothing happens until I do it. There’s no suggestion but only wait for my instructions.

When I’m learning stuff, I often find myself mugging things up to just remember things of value while not getting satisfactory understanding & depth of things.

It’s just learning things for the moment. Though this is immensely useful at the moment but doesn’t make a lot of sense in the long run. Connecting the dots gives far better learning experience.

You can learn it better if you create or recreate something.

There’s a concept in learning called Active Recall.

Active recall is a high-efficiency, evidence-based learning strategy that strengthens long-term memory by requiring learners to actively retrieve information from their brain rather than passively reviewing it.

When you look at a study material for a time long enough, you develop a familiarity. However this familiarity doesn’t guarantee the ability to recreate or apply the leanings from material when needed. So, passive looking / reviewing doesn’t help much.

Active Recall on the other hand, requires you to work your brain & fetch that information. Your brain only stores stuff that it thinks is important. Active Recall signals your brain that this particular information is important so you’re more likely to remember / store that piece of information.

The process of Active Recall is slow & painful but a very rewarding one. I would even argue that this is the most genuine form of learning.

I even found this true in programming. Say you want to learn something, try creating a project around it from scratch. Doesn’t matter how much back & forth you have to do in documentation, notes & articles. You’ll learn much more than you’ll ever learn in a theoretical skimming session.

To apply this, go through something in an easygoing way. Now, put it aside & take pen & paper and recreate what you read. That’s it!

I was writing this blog not because I had something to say but I felt like I had to say something. I wanted to have an opinion on a lot of things and especially on the talks of the town in Tech. I was mostly correct about my opinions & predictions but it felt like I was doing it for the sake of being a journalist which I’m not. I never had the deep knowledge of these tech things. (Although a lot of times my knowledge was deeper than a lot of folks that worked with that tech)

When you have such broad scope, you cease to dive deep into a subject.

I never learned and understood things with depth. I had amazing breadth, I still do; I’m a generalist. But eventually I realized to create value for myself, I have to be a T-shaped engineer. Depth in at least one thing with breadth in others.

Right now, I’m doing my masters in CyberSecurity domain so I chose this to be the depth point. But after diving deep in Cyber, I realized that it’s a broad one complementing my earlier style. An accident of convergence. But it requires some level of depth in some things. The Fundamentals.

This blog is documentation of my understanding of things sans AI SLOP.

These are my perspectives on engineering. They could be debatable and sometimes wrong too. But they’re mine. I’m not aiming at being correct, just raw & real.