There's this thing people say now: that the tools are changing us. That we're losing something. That the person who used to sit down and think through a problem is being slowly replaced by the person who asks a machine to think for them, and honestly, I get why that lands. It's not an unreasonable fear.

But I want to push back on it for a second. Not because I think it's entirely wrong, but because I think the framing is off.

When you pick up a new skill—really pick it up, not just skim a tutorial—the person you were before that skill starts to fade. That's not loss. That's refinement. The old version of you doesn't get destroyed so much as it gets polished down into something that can do more, see more, hold more weight without cracking.

I think about carpentry sometimes. Not because I'm any good at it [No, dear, I cannot make that mid-century Asian-inspired credenza you've been eyeballing and not-so-subtly hintdropping about.], but because the metaphor holds up. There's a guy who's never touched a jointer plane, never attempted a dovetail joint, never really understood why wood moves the way it does. And then one day he learns. He picks up the plane, figures out the angle, starts producing joints that actually fit. The old version of him—the one who would've just grabbed a nail gun and called it done, that guy is gone. And nobody mourns him, because the new version is better. Not better in some abstract motivational-poster way, but functionally, practically, observably more capable.

That's what tools do when they work right. They don't replace you. They refine you.

There's a line I keep coming back to—and I'm not going to dress it up with attribution because either it lands for you or it doesn't—about putting away childish things. When I was a child, I spoke as a child, I reasoned as a child. But at some point you set those things down. Not because they were bad, but because you outgrew them. You don't want milk anymore; you want a steak.

And I think that's closer to what's actually happening with AI tools than the narrative we keep hearing. The narrative says erosion—as though something is being worn away against our will, ground down by a force we didn't invite. But what if it's formation? What if the discomfort people feel isn't loss at all but the growing pains of becoming someone who can operate at a higher level—someone who's been shaped by the tools they chose to pick up?

That reframing matters to me. Because erosion is passive. It happens to you. Formation is something you participate in. You show up. You do the work. You let the process change you because you trust that the person on the other side of it will be more than the person who went in.

I'm not naïve about the counterpoint. And the counterpoint is real.

When you stop doing the thinking yourself, when you just accept the output and move on, no critical evaluation, no "wait, is this actually right," you're not being sharpened. You're going dull. The edge that made you useful, the part of you that could sit with a hard problem and reason through it, that atrophies if you never use it. And the thing about atrophy is that it's quiet. You don't notice it happening until one day you reach for a capacity that used to be there and it's just...not.

The light slowly goes out.

So the question isn't whether to use the tools. That cheese has long since been eaten, and I'm on it—I'm literally using an AI tool to help me write this, which is either ironic or perfectly appropriate depending on how generous you're feeling. The question is whether the tools are shaping you or just doing things for you. Because those are not the same thing. [And no, using AI to proofread is not the only thing I've done for this piece...it was getting a bit rambly, so I used it to trim things while retaining my voice. It was...helpful.]

A tool that shapes you leaves you sharper than it found you. A tool you lean on so hard that you forget how to stand isn't a tool anymore. That's called a crutch. And crutches are fine when you're healing, but nobody wants to use one forever.

I don't have a clean answer here. I don't think there is one. But my instinct—and it's just that, an instinct—is that the safeguard isn't in the tool itself. It's in the question you ask after the tool gives you its answer:

Is this reasonable?

Three words. That's it. If you can still ask that question and mean it, if you can look at what the machine produced and bring your own judgment to bear on whether it's actually good, then you're still in the game. You're still being formed. The tool is doing what tools are supposed to do.

But the moment you stop asking, the moment the output just flows through you, frictionless, that's when the old fear earns its keep.

I'd rather be refined than eroded. I think most people would. The difference is just whether you're paying attention.

I am, by every measurable standard, more capable than I was a year ago. That's not nothing. The AI tools I've folded into my workflow, largely Claude Code and OpenAI, have genuinely expanded what I can accomplish — whole categories of work that I could not have dreamed of asking the first questions of research about, let alone executing, now get done. I find it useful not to inflate this: it's not really me doing the work. It's me prompting something that does the work. But the output lands in my name, on my desk, and I've learned to navigate that distinction without too much philosophical hand-wringing about it. After all, I mean I am the one formulating the informed-enough questions to get the ball rolling; the project wouldn't progress without me.

What I Did Not Anticipate Was the Accounting.

The thing about becoming more capable is that people notice. And when people do notice, they load you up accordingly. I may be five times more productive than I was twelve months ago — and I am, without question, five times busier. That's the kind of math that does not sum to any total in your favor. Somewhere between the tool and the output and the organizing of the input and output files I'm generating, the day disappears. Keeping track of what the AI has helped me produce is, on some days, a full enough job by itself, leaving little time to muse or ideate. It can be an endless loop of iteration, making the next newer version of a thing that wasn't needed until someone had the capability to do it. kinda a Jurassic Park thing — so consumed with whether we could do whatever, we never stopped to consider if we should [ or however the quote is supposed to go...positive Kevin will correct it ;) ]

So Stay Put the Boulders

It's easier to move pebbles. Low-hanging fruit, quick wins, the small tasks that give you the brief satisfaction of checking something off — yes, each one inches you a step closer to a goal. But I've lost sight of the goal because I'm head-down in the pebbles. The bigger items, the ones that require sitting with a question long enough for something useful to happen, get deferred to some later time when things quiet down. Things don't quiet down.

I used to be able to sit in a restaurant or a café — the steam off a coffee cup, the smell of roasting garlic, the sound of passing cars blurring into white noise — and find my way into something like a flow state. Reflective. A little slow. The kind of thinking where the bigger picture starts to arrange itself. I wrote during that time, and the writing helped. It has always helped me find clarity on the things I actually care about moving.

I Haven't Written Anything Outside of Work in Over a Year.

There's something worth naming in that irony. The tools that were supposed to free me up to have more time to cogitate and muse have kept me busier with smaller things. Whether that's a feature of the tools or a bug of how I've let myself use them — I genuinely don't know. I haven't had time to figure it out.

I have pressing deadlines today, as it happens. This thought will have to wait.

Which is, I realize, exactly the problem.

Ready to elevate your coffee game? While the "best" cup of coffee is a matter of personal taste, there are a few brewing methods that consistently stand out for their ability to produce a delicious and satisfying cup. Let's count down to the best as we explore the top three brewing methods that every coffee lover should know.

3. The AeroPress: For a Quick and Versatile Cup

The AeroPress is the new kid on the block, but it has quickly become a favorite among coffee lovers for its speed, portability, and versatility. This unique brewer uses a combination of immersion and pressure to create a smooth, rich, and low-acidity cup of coffee in about a minute.

What You'll Need:

• AeroPress
• AeroPress filters
• Finely ground coffee
• Hot water (185°F/85°C)
• Digital scale
• Timer
• Mug

How to Brew:

1. Prepare Your AeroPress: Place a filter in the filter cap and twist it onto the AeroPress chamber. Place the chamber on top of your mug.
2. Add Coffee and Water: Add 17 grams of finely ground coffee to the chamber, then add 220 grams of hot water.
3. Stir and Plunge: Stir the coffee and water for 10 seconds, then insert the plunger and press down gently for about 30 seconds.
4. Enjoy: You'll have a concentrated shot of coffee that you can drink as is or add hot water to for an Americano-style drink.

Whether you prefer the control of the pour-over, the richness of the French press, or the speed of the AeroPress, there's a brewing method out there for you. So go ahead, experiment, and find your perfect cup.

2. The French Press: For a Rich and Full-Bodied Brew

The French press is a classic for a reason. This immersion-style brewer steeps the coffee grounds directly in water, resulting in a rich, full-bodied, and robust cup of coffee. It's a simple and forgiving method that's perfect for those who enjoy a bold and flavorful brew.

What You'll Need:

• French press
• Coarsely ground coffee
• Hot water (200°F/93°C)
• Digital scale
• Timer

How to Brew:

1. Preheat Your Press: Fill your French press with hot water to preheat it, then discard the water.
2. Add Coffee and Water: Add 56 grams of coarsely ground coffee to the press, then add 900 grams of hot water.
3. Steep: Let the coffee steep for 4 minutes.
4. Plunge: Slowly and evenly press the plunger all the way down.
5. Serve: Serve the coffee immediately to prevent it from becoming bitter.

1. The Pour-Over: For the Coffee Connoisseur

The pour-over method is the go-to for coffee aficionados who want complete control over their brew. By manually pouring water over the coffee grounds, you can control the water temperature, flow rate, and extraction time, resulting in a clean, nuanced, and flavorful cup that highlights the unique characteristics of your beans.

What You'll Need:

• Pour-over brewer (like a Hario V60 or Kalita Wave)
• Paper filter
• Gooseneck kettle
• Digital scale
• Burr grinder
• Your favorite whole bean coffee
• Mug or carafe

How to Brew:

1. Heat Your Water: Heat your filtered water to 200°F (93°C).
2. Rinse the Filter: Place the filter in your pour-over brewer and rinse it with hot water to remove any paper taste and preheat the brewer. Discard the rinse water.
3. Measure and Grind: Measure out 20 grams of coffee and grind it to a medium-fine consistency.
4. The Bloom: Add the coffee to the filter, place it on your mug and scale, and pour 60 grams of water over the grounds. Let it "bloom" for 45 seconds to release CO2.
5. The Pour: Continue pouring the water in a slow, circular motion until you reach a total of 340 grams of water. The entire process should take about 3 minutes.
6. Enjoy: Let the coffee drain, and then enjoy your perfectly crafted cup.

Now for a Little Explainer

From this point forward, 'tis I who write (/ dictate) and not some crummy Ai.

Why did I do it? you may never fully know. Just kidding. It is a proof of concept. I wanted to to see if i could get Ai running in the terminal and able to access the file structure of my computer to do its work.

The cool part is that it works (as you plainly see witthhe post above).

The not-so-cool part is that Ai robs us of the opportunity to create. Because this blog post was nothing more than just a proof of concept I didn't really care necessarily that it should have anything of personality baked into the methods presented.

I can understand the allure of something seen as a time saver but at the end of the day just like there is absolutely no life present in the blog post above. I can't get behind anybody using AI to completely take control of the writing process.

That said, I don't think the output was terrible, but there's just really no humanity in it whatsoever. Tell me if you think otherwise.

It may sound a bit meta at this point of development in regard to this website (after all, it started as a way to share my current learning in the tech field), but I think this is a useful stratagem in enabling others to do the same. My aim is to pull back the curtain and show how the sausage is made, so to say. We will focus on the method and format I try to use when writing here such that you may be able to do the same.

What are we doing?

In theory, in taking pen to paper, or in this case, keyboard to screen, we are passing a torch, conveying a skill gained to a learner come behind us. Keeping in mind that acquiring a singular skill (learning) plays part of a grander mosaic called mastery, we the teachers all too often overdo it when imparting what should be a nugget-sized soundbite. We (myself very much so included) forget that education is i+1, not plus everything-all-at-once.

The curse of knowledge erases our collective memories of the burden learning places on students. We bite off too much for them to chew, swallow, and digest.

What should we be doing?

Well, if you have known me for any time at all, you know that I have a propensity to be a touch (possibly understated for effect) verbose in my output. We need to keep it concise.

Concise does not mean shallow. It means cutting the fluff, giving one idea room to breathe before introducing the next. Think of it like serving a tasting menu instead of dumping the entire pantry on the table. Each dish has a purpose, a sequence, and a chance to be appreciated before the next one arrives.

That’s where writing to teach shines. When we write with the learner in mind, we are forced to distill, to prioritize, to make trade-offs. It is an act of empathy as much as it is an act of explanation. And the funny thing? You end up learning more yourself. The act of teaching locks in the lesson for the teacher better than almost anything else.

So, what do we do with this?

We write like someone will actually try to follow our steps (because they might).
We trim the fat, we break things down, and we accept that we will never say everything there is to say in a single post—nor should we.

Writing to teach is not just about sharing what we know. It’s about making the path a little clearer for whoever comes next. And if we do it right, we leave them just enough light to take one more step. No more, no less.

If you’ve learned something recently, try writing it down—not for yourself, but for the version of you who didn’t know it yet. Then hit publish. Someone out there is just one well-written paragraph away from their own aha moment. Who knows, you might be the one holding the torch.

I'm one of the lucky ones. This may not come as a surprise to you, but in the face of the statistics pointing to an ever-decreasing demand to attend and complete a bachelor's course, somehow I managed to pass that milestone, land a job in my incredibly niche chosen field (teaching German), and still somehow move on.

A quick search shows the following statistics:

• Only 46% of college graduates work in jobs related to their field of study.
• 52% of college graduates are underemployed, meaning they work in jobs that typically do not require a bachelor's degree.

Crazy to think that I could be the lead administrator for any online technology education platform, let alone the NetworkChuck Academy. Let me unpack that for just second. The skills I cultivated while building this very website enabled me to meet the meet the challenges (I'll save you the details) of my current position head on. I work for and alongside the guy whose YouTube videos got me interested in technology in the first place. None of the training / learning path went to waste; my tool pouch is full, and I use every one of my tools to accomplish my day's work.

While puttering with my fledgling website back in (maybe early) 2020 while working for DYB in Seoul, I would never have dreamed that I would someday be in the position I currently occupy. It's not without difficulty that I trace back the steps that lead to this moment. I can't help but remember back to the nostalgic refrains echoed by my old boss / supervisor, Michael Meyers (no, not that one—the other one you didn't know from the movies), quoting the opening lines from that old Talking Heads song.

I ask myself, "How did I get so lucky?" I may not be teaching German any more, but once the shine wore off that apple I found something else that tickled and got trained up in it. Somehow, despite the odds, that second education also ended up being put to use.

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

If there is one takeaway for me, it's to never stop learning—it won't be in vain (edited; credit goes to Kevin Kim for catching my sloppiness). You may not land where you had envisioned your future self to be 10 years from now, but in pursuing what you love, you'll be in the right ballpark. Maybe it sounds a bit hedonistic, but if learning something pleases you to, you should probably do it. You never know where it might take you or when you will look back wondering how you got where you are.

There's been so much to occupy my time and thoughts since arriving in Colorado, but occasionally I have a moment to look back at photos of daily life in Seoul. This is the view from across the Han at night. I may have shot many dozens of similar photos but they never came out as good as this--likely owing to sweat smudges smeared across an iPhone lens as I huffed homeward.

I can't help missing it although it seems like a lifetime has passed. They say the passage of time heals, but my heart hungers for the frenetic pace and bustling traditional market where we used to get 떡볶이 and 순대 간. Life is good here--good but very, very different.

It came as a shock having landed how easily overheard and understood chatter of passersby floated into the background consciousness; the strain to grasp at a potential third of Korean speech directed at me just hours before still stands in stark contrast.

Seoul never really belonged to me nor I to Seoul, but it was home. On rare rainy days such as this, I remember.

Well, if you’ve been wondering where I’ve gone or why I haven’t been updating the site since back in August, this post is for you.

It‘s no secret that I usually have a lot to say, and maybe because of that, I have found it much easier to say nothing than to stay tight-lipped and risk accidentally alluding to something that may as well be the same as spilling the beans.

Here is The Update…

I have been working with a company called The Network Chuck Academy .

For anyone out there unfamiliar with that particular website I linked above, that’s ok. This is an up-and-coming online training platform focused on just about anything tech related. I have been assigned the CCNA portion and have been having a blast making study materials for the course. I never would have thought that my training as a curriculum writer in Korea would have such direct intersection with anything remotely related to network engineering, but here we are.

There isn‘t much more that I can really say at this point, but I really wanted to break the silence.

I plan to be coming back here with more updates and material in the coming months, now that I know I am not stepping on any toes by keeping this website while also working for a company that could otherwise take the work I do here as direct competition or undermining of their mission.

Last Thoughts

I wish only that I had had while in Korea exposure to the writing and versioning tools I am using now at NCA. If you’re curious, go check out Notion and Clickup. I don’t love Clickup, but it is super helpful in tracking the progress of jobs; Notion is awesome for anyone who uses markdown. And if you don‘t yet, what are you waiting for? It’s far easier than clicking through a billion dropdowns in Microsoft Word with the added bonus of being essentially plain text that your designer will love you for using.

If you are anything at all like me, and I bet you are, because it's kinda human to be as lazy as possible conserve energy for more cognitively demanding tasks, you likely will benefit from removing any friction that slows you down from starting anything involving lab work on your server. Don't feel bad, the smartest people I know sometimes have the hardest time getting started. But once you get moving, you start to fly.

Today, I will show you how I automate the startup process of my EVE-NG server.

First, a little as to why I bothered to set this up. I recently migrated my EVE-NG server out of Google's Cloud in favor of an on-prem server. I managed to pick up a retired Dell R720 for a reasonable price. But whatever it saved me in upfront costs would be eclipsed by the electric bill soon enough if I just let it run around the clock. Therefore, I find it much more practical to start up the lab and load startup configurations before doing my lab work.

So, What's the Problem?

Server-grade hardware goes through way more system checks before loading the operating system. The short version is that it takes anywhere between ten and twenty minutes conservatively before I can even begin my studies. Those 20 minutes are not completely unattended, thus I kinda have to babysit the server and make sure everything loads and we are ready to rock.

Wouldn't it be great is there were a way to set a schedule such that the server basically had its own alarm clock? I could wake up and brew my favorite coffee, and by the time I was half-way through the first cup, EVE-NG would be ready and waiting for me. Speaking of...

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

Now with that out of the way...

There is a short list of things that we will need to accomplish. Before getting started, this video assumes you have a server that runs 24/7 to function as the "alarm clock" for your (much-more-expensive-to-operate) EVE-NG server. You could use a Raspberry Pi, but I will use my Proxmox server.

The 4-Part Secret Sauce

• SSH keys — for password-less authentication
• An alias 
• A Powerup script
• A cronjob

If that doesn't make sense, watch this walkthrough. I also have my scripts available on my GitHub. Feel free to modify them as you see fit, and let me know how you get on with it.

Cheers!

#automateallthethings

Hi. Today I was running some Ethernet cable when it dawned on me that there may be someone out there who finds use in a brief tutorial on this mundane task. I'm going to show you how to terminate RJ45 connectors.

Unfortunately, I didn't have a decent tripod to hand, so please excuse the awkward angle. Also, I left out the dragging of the cable through the crawlspace, as there really isn't enough light down there to make decent content; you will just have to use your imagination, I guess.

Why Cat6a?

On the end of the cable that I demonstrate, you will see a passthrough style RJ45 connector crimped to a Cat6a cable. A million, million years ago when I did my first-ever Ethernet cable run (I believe I was in 6th grade and fastEthernet at 100mpbs wasn't even a thing yet). That was Cat5e and is surprisingly still very capable for short runs. But as with all things tech, eventually every specification meets its end. If this house had had existing Cat5e, there may have been a bit more to deliberate, however, as there was no wired infrastructure when I moved in here, the choice to use Cat6a was a no-brainer.

Not only does Cat6a support higher bandwidth (10Gbps up to 100 m), it also allows for higher voltages for power over Ethernet (PoE) devices, such as CCTV cameras and even monitors—there is even a way to use the Ethernet cabling to provide power to a laptop or PC.

In a word, deploying Cat6a future-proofs my home network for whatever comes down the line. I can rest well, knowing that this will likely be the last set of Ethernet cables this house will need in my lifetime. Whether or not I ever take full advantage will tell in time. But you may not have access to the same quality of cable at the low price I found it. Really when it came to cable selection, the extra $6 was worth the cost for me.

The Demo

That said, here is a video of the crimping process. It's pretty straightforward. But keep in mind there are two different wiring standards. I use T568B. You can use whatever you like, but the standard must match on both ends of the cable.

If you made it this far and like what you see

• please consider supporting us
• share this content wherever appropriate and applicable
• jump in the comments and give us a shout out
• and subscribe for more

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

The Tools

The Other Side

I apologize for the lack of video for this one, but the background noise was pretty bad. Perhaps, if there is enough interest in the future, I will make a quick tutorial on crimping keystone jacks—although that's pretty straightforward.

The Final

Here's how it looks, now that it's all finished up.

And now that everything is done, and the pain has been touched up, I hope you have a better idea of how to terminate Ethernet cabling in your own home.

I've been holding off on this post for a while. Mostly I needed get my house in order. There has been a lot of change in the past week or so and yet a lot has stayed the same; I'm still learning and growing in my pursuit of network engineering; I'm still looking forward to making my my mark on the world of networking; I'm still waiting on the TS clearance before starting my new position.

Everything has happened at the exact time it was meant to

This journey has been one of setting one foot down before knowing where it was going to land and praying there would be solid foundation to plant myself on before pushing off onto the next footfall. If it feels as though I am being vague, I am merely sparing you the details. I know that things are working out--even if it is slower than the timeframe I have been hoping for.

In the meantime, I am just glad to be surrounded with the beauty of nature and the closeness of friends. I leave you with a mild photo-dump. I will omit thechallanges and include just the good bits:

I may be a place I’m not necessarily supposed to be. More to follow; stay tuned…

Here's the Rest...so...EDITED

If you've ever looked up to the skyline here in Seoul, you've no doubt seen loads of items similar to the one pictured in my rooftop photoshop as the image for this post. Maybe you've seen them and not thought much about them, or perhaps you have seen them and not really noticed them.

It's also possible you notice them no matter how poorly disguised as pine trees they tend to be. Whatever the camouflage used to blend these objects into their surroundings, my eyes cannot help giving them extra attention.

So, what are they?

What you are looking at there is a classic example of cellular antennas. The technology behind them is not terribly dissimilar from your WiFi at home or in the café you are sitting in sipping your cup of Joe right now. The main difference is that these big boys tend to use a different means of transmission.

Microwaves...not just for Hot Pockets

It is helpful to think of the energy emitted from and received by (not just) these cellular antennas as part of the electromagnetic spectrum—because that's what it is. I've heard it explained as infra-infrared light. But where I am not a physicist, nor will I attempt to explain it that way, I will say that framing in radio waves with comparison to the energy of light is somewhat helpful for me.

As you can see, microwaves are just below the infrared portion of the spectrum, followed by radio waves with slightly less energy. That means the cellular data is beamed from one tower to the next with waves at only a slightly higher frequency than those used to send music to your FM receiver—it's high-power radio!

Fear Not

If you are worried about the emissions from these antennas, let me put your mind at ease. There isn't enough wattage available to the antennas to do much damage to your leftovers. There are differences from country to country and carrier to carrier, but these are some of the estimates I've come across:

20 Watt for GSM 900 and LTE 800
10 Watt for GSM 1800 & UMTS
6 Watt for WLAN & LTE

Compare that with the common household popcorn burner at 1200w, and you will see why I am not really concerned to get so close for the photo opp. Furthermore, cellular data is not broadcast transmission and, as a result, ramps up and down its wattage requirement based on load.

So why am I telling you all this?

I'll tell you after this coffee break...

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

Glad you're back. I hope your brew was fulfilling. Mine was a locally roasted Kenyan blend from a roastery by the name of Sanchez. I'll see if I can get a photo of the shop the next time I pass by on my bike. Now back to the story.

There is just something to be said about the appreciation of the hard work that has been done in connecting us all. Something of a sense of pride as a network engineer wells up within me when I see the hardware that represents the planning and deployment that makes our modern mobile connections possible.

It blows my mind to know how rigorous the certification for CCNA is (after having passed it over a year ago), and yet how little the certification concerns itself with wireless technologies. As a matter of fact, there is a whole CCNP (Cisco Certified Network Professional) specialization that scratches the surface—Designing Cisco Enterprise Wireless Networks (300-425 ENWLSD)—if you are interested.

There is still a lot about wireless technology out there for me to learn, but the more and more I dive down this rabbit hole, the more and more intriguing I find it. I am not pledging to certify in wireless design, but this post is mostly my way of announcing that I am focusing on wireless design for the foreseeable future.

If you see someone on a rooftop taking pictures of wireless equipment installed by your local ISP, don't be shocked to find me behind the camera. And keep your eyes open for more of these towers of connection; once you start to recognize cellular radios and antennas, you start noticing them everywhere.

RADIUS (Remote Authentication Dial-In User Service) is a protocol used to manage and secure network access. In simple terms, it's a system that helps ensure that only authorized users can connect to a network. Think of is as a server that network devices query to determine who is allowed to login and what privileges that user should have.

Why would I use RADIUS?

For those of you on the fence about implementing RADIUS authentication in your enterprise network, here is a short list of the benefits of using RADIUS:

1. Centralized Authentication: Simplifies management by storing user credentials on a central server.
2. Improved Security: Utilizes encrypted communication for secure credential transmission.
3. Scalability: Handles high volumes of authentication requests for large networks.
4. Granular Access Control: Allows detailed access policies based on user roles and attributes.
5. Accounting and Auditing: Logs user activity for monitoring and billing purposes.
6. Ease of Integration: Compatible with various networking devices and services.
7. Enhanced User Management: Simplifies adding, modifying, or removing user accounts centrally.
8. Reduced Administrative Overhead: Saves time and reduces configuration errors.
9. Policy Enforcement: Enforces security policies consistently across the network.

Using RADIUS for authentication enhances security, efficiency, and manageability of network operations.

How can I set it up on my Edgerouter?

You're in luck, I have a tutorial on setting RADIUS up. Unfortunately, I couldn't find any other tutorials out there, so you are stuck with mine for now. If you need any further help, or if there was anything in my explanation that was unclear, toss your questions in the comments below. Before getting to the configuration video, here is a friendly reminder to get your coffee ready before diving into the configuration on your own.

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

As always, I hope this helps—Happy networking!

The debate over AI is far from settled. Competing views on appropriate use of generative AI chatbots range from allowing full, unfettered implementation to blocking their use altogether. Some advocate age restrictions. I wrote on the topic of AI in this post and have not been persuaded into a changed stance as of yet—frankly, there has been little noteworthy informed conversation on the topic in my circles. Still, I wanted to revisit the topic and frame the dialogue through the lens of education. Nothing compares to the potential for expediting the learning process the way that AI does. When leveraged wisely, AI chatbots hot-knife-butter learners through topics that take much longer to accomplish in a more traditional learning situation.

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

One undeniable advantage of AI chatbots is their ability to foster critical thinking skills. By engaging with these virtual assistants, we are encouraged to ask concise, probing questions and go deeper into subjects than we might in a traditional classroom setting. This process not only enhances understanding of the material but also cultivates a habit of inquiry that will serve well in all pursuits. Practice posing the right types of questions, and you will start to get better and better results—kinda a garbage-in-garbage-out situation—that improves through active engagement. If doing is learning, then the act of asking questions to a chatbot is honing the skill of inquisition, which in turn, unshackles the brain to ask more creative questions.

However, the vast sea of information accessible through AI chatbots also presents a potential pitfall. With so much knowledge at our fingertips, we may find ourselves easily distracted or led astray from the intended learning objectives. Without proper guidance and structure, the allure of endless information could hinder rather than help the learning process. It’s the problem of ignorance; we don’t know what we don’t know. 

At best, AI chatbots mirror our language and provide a believable response. It’s not entirely a blind-leading-blind situation, but we should be careful not to crown king the cyclops only because we ourselves are cannot see. My prescription? Seek professional help. If you really want to or need to know, ask someone who does—or get a book and put in the work to really learn.

There's more to the issue of accuracy and reliability. While AI chatbots can provide instant answers to a wide range of queries, raising the risk of encountering misinformation, we can develop discernment skills to distinguish between credible sources and dubious ones, lest we inadvertently internalize inaccuracies. Learning takes on a new degree of completion when we sit in the judge’s seat and weigh the validity of whatever facts or arguments we encounter. AI can and should be instrumental in cultivating a healthy skepticism. At least according to Bloom, there is only one more step to complete mastery.

On a more positive note, AI chatbots can also inspire creativity. They can expose diverse perspectives and ideas without bias. These virtual assistants may not directly promote ideation, but their output can become the seeds of great epiphany that would otherwise have been left unplanted; they are a wildcard in the brainstorming process. We still need to think outside the…noodle and approach problems from new Blickwinkeln. This injection of creativity is the pinnacle of the learning experience and the very essence of our humanity.

Summing it all up would be an exercise in futility, as I assume the debate will smolder on for quite some time. In the meantime, I see no reason to bar anyone from exploring the possibilities that chatbots offer. If anything, I see a sound reason in exposing even the youngest of learners to AI, so long as it is done responsibly. Whatever that may look like, I encourage you to explore it out loud in the comments below. #commentifyourenotabot

Ah, spring! The season of renewal, blooming flowers, and... allergy-itchy skin and eyes? Yes, you read that right. I find myself battling not with the complexities of network configurations but with the irritating discomfort of dry skin and a bit of python code. At least pythons don’t bite, am I right? But hey, I'm not one to let a little itchiness get me down, especially when the trees budding and the flowers—ah-ah-ah-choo!

As I sit at my keyboard, scratching away at my arms and neck (sorry for any disturbing visuals that statement conjures), I can't help my excitement for the changing weather outside. The sun shining, birds chirping—it's a welcomed change from the dreariness of winter. When I’m honest, winter is overrated; it’s great at first, but it wears on my spirits by mid-January, beginning of February at the latest. I never really have the same spark in winter that warmer weather ushers in. 

Anyway, despite the uplift spring brings, there's a technical challenge that's been nagging me to the core. You see, I've been working on a custom Python script to handle the synchronization of .ssh/config and /etc/hosts configurations across multiple devices. Sounds fancy, right? Well, it's been a bit of a bumpy ride, and it really shouldn’t be. After all, I am basically just building a very localized DNS listing.

I thought I had everything figured out—writing lines of code, testing, rewriting, retesting, scrapping, reimagining—but alas, it seems like there's always something that doesn't quite work as expected. Maybe it's a syntax error here, a logic flaw there, or perhaps just the universe reminding me that perfection is a myth. Or maybe it is spring kicking me out the door to go enjoy the 개나리 (Forsythia, pictured above) that are barely blooming. Regardless, I refuse to let this setback wet-blanket my enthusiasm.

So here I am, scratching my head both figuratively and literally, determined to make this code confirm to my will. And while I may not have all the answers just yet, I'm hopeful that with a little persistence and perhaps some input from fellow tech enthusiasts, I'll get there—wherever there happens to be.

Which brings me to you, dear reader. If you happen to be a seasoned Pythonista or a Linux expert, I welcome your insights and suggestions. Heck—I’d even settle for a few words of encouragement. After all, isn't that what community is all about? Sharing knowledge, lending a helping hand, and celebrating victories—no matter how small they may seem.

So why not embrace the springtime challenges with optimism and dogged determination (not to be confused with dogwood irritation).

Together, let's keep coding, keep exploring, and keep scratching that itch (figuratively, of course). And who knows, maybe we'll sort this code together sooner than we think.

Here is a few revisions of the code, so you can kinda get an idea of what I’m working with.

I started with python script to just create a simple sqlite3 database.

import sqlite3
Connect to the database
conn = sqlite3.connect('hosts.db')
c = conn.cursor()
Create table
c.execute('''CREATE TABLE IF NOT EXISTS hosts
             (id INTEGER PRIMARY KEY,
             hostname TEXT NOT NULL,
             ip_address TEXT NOT NULL,
             username TEXT,
             identity_file TEXT)''')
Commit changes and close connection
conn.commit()
conn.close()

And here is a somewhat working bit of code:

import sqlite3
import shutil
import os
from datetime import datetime

# Function to create backups of config and hosts files
def backup_files():
    now = datetime.now().strftime('%Y%m%d%H%M%S')
    shutil.copyfile('.ssh/config', f'.ssh/config.{now}')
    shutil.copyfile('/etc/hosts', f'/etc/hosts.{now}')

# Function to handle getting hosts from the database
def get_hosts():
    conn = sqlite3.connect('hosts.db')
    c = conn.cursor()
    c.execute('SELECT * FROM hosts')
    hosts = c.fetchall()
    conn.close()
    return hosts

# Function to handle adding hosts to the database
def add_host(hostname, ip_address, username=None, identity_file=None):
    conn = sqlite3.connect('hosts.db')
    c = conn.cursor()
    c.execute('INSERT INTO hosts (hostname, ip_address, username, identity_file) VALUES (?, ?, ?, ?)',
              (hostname, ip_address, username, identity_file))
    conn.commit()
    conn.close()

# Function to handle editing hosts in the database
def edit_host(hostname, new_ip_address=None, new_username=None, new_identity_file=None):
    conn = sqlite3.connect('hosts.db')
    c = conn.cursor()
    if new_ip_address:
        c.execute('UPDATE hosts SET ip_address = ? WHERE hostname = ?', (new_ip_address, hostname))
    if new_username:
        c.execute('UPDATE hosts SET username = ? WHERE hostname = ?', (new_username, hostname))
    if new_identity_file:
        c.execute('UPDATE hosts SET identity_file = ? WHERE hostname = ?', (new_identity_file, hostname))
    conn.commit()
    conn.close()

# Function to handle deleting hosts from the database
def delete_host(hostname):
    conn = sqlite3.connect('hosts.db')
    c = conn.cursor()
    c.execute('DELETE FROM hosts WHERE hostname = ?', (hostname,))
    conn.commit()
    conn.close()

# Main function
def main():
    backup_files()
    action = input("Would you like to: Get (1), Add (2), Edit (3), or Delete (4) hosts? ")
    
    if action == '1':
        hosts = get_hosts()
        for host in hosts:
            print(host)
    elif action == '2':
        hostname = input("Enter hostname: ")
        ip_address = input("Enter IP address: ")
        username = input("Enter username (optional): ")
        identity_file = input("Enter identity file (optional): ")
        add_host(hostname, ip_address, username, identity_file)
    elif action == '3':
        hostname = input("Enter hostname to edit: ")
        new_ip_address = input("Enter new IP address (leave blank to keep current): ")
        new_username = input("Enter new username (leave blank to keep current): ")
        new_identity_file = input("Enter new identity file (leave blank to keep current): ")
        edit_host(hostname, new_ip_address, new_username, new_identity_file)
    elif action == '4':
        hostname = input("Enter hostname to delete: ")
        delete_host(hostname)
    else:
        print("Invalid option")

if __name__ == "__main__":
    main()

If you followed all that, you will see that at this point in the code arch, we have a programmatically-generated database and a basic user interface to input hostnames, ip addresses, usernames, and identityfiles (useful for pointing to the location of public-keys for secure ssh connections). The script successfully generates a datetime stamped backup of the .ssh/config and /etc/hosts files.

Where I am having trouble is in writing out to the appropriate files (and in the correct format) the variables that users input into the database. In further code revisions, I have tried (and succeeded at) parsing the .ssh/config and etc/hosts files into json key-value pairs, but I haven’t been able to connect all the dots before getting the itch—er, uh, urge—to get outside. 

And with that, I will leave you with this: Do you even think this is a problem worth solving? Would you just run pihole or bind9 as a prebuilt DNS? What exactly is the problem I have set out to resolve (bonus points if you can tell me, but I’ll give you a hint: VPN)?

I’m going out to play—laters—and happy networking!

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee

Well, it’s March again, and that means at least one thing: it’s time to revisit my phone situation. For anyone who doesn’t know, life abroad comes with a slew of challenges, not least among them being telephony. 

It was two years ago when I came across an enterprise solution that knocked off my socks. I was introduced to the world of PBX, SIP, and SPTN. If all those acronyms are too much, you are in for a read. The shortest version of the story is that through the magic of the aforementioned alphabet soup, phone calls and text messages using a local US-based (but not necessarily, as we will discover later) phone number are simple and cheap. Say goodbye to your calling cards, if you are old enough to remember those things.

Let’s define a few things:

PBX

A PBX is like a private telephone network used within a company or organization. It allows internal communication between employees as well as connection to external phone lines. Think of it as a sophisticated telephone system that manages calls within a business, offering features such as call forwarding, voicemail, and conferencing. Some will even allow you to send and receive text messages and host video chats. 

SIP

SIP is a protocol that enables voice and video calls over the internet and is often associated with VoIP (Voice over Internet Protocol) services. SIP allows different devices and networks to communicate seamlessly, enabling services like internet telephony and video conferencing. It is responsible for setting up, maintaining, and tearing down connections between participants on a call. SIP sounds complicated; I am over complicating it in my definition in efforts to feel smart.

One of the neatest things about SIP trunks is the availability to bridge between your PBX and the PSTN of just about any given country. You just find a SIP provider with favorable rates in the calling region of your choice, subscribe (get a phone number) and you are good to go.

PSTN

PSTN refers to the traditional telephone system that most people are familiar with. It's the global network of copper wires, fiber-optic cables, cellular networks, and satellites that enable landline and mobile phone communication. In essence, it's the backbone of global telephone communication, connecting callers worldwide.

*note* Depending on where in the world you are reading this post from, your country may already be on the path to phasing out the PSTN entirely in favor of VoIP (voice over internet protocol) technology. As PSTN infrastructure ages, there are fewer and fewer professionals around to service and maintain the analogue phone network. Moreover there has been an industry-wide push to see adoption of VoIP technology slated globally for as soon as 2030 and earlier in some regions (such as the UK, 2025).

What does that mean to you as an end user? Likely nothing. Calls  and texts will still work as before—they will just connect over the internet instead of the PSTN. It may translate into savings, if you are the one setting up and maintaining the services, though. Then again, it may not. Fear not; as VoIP will replace PSTN, so also did PSTN supplant telegraphs.

So, now that we have all that out of the way,

let’s get back to my case of these technologies and the motivation for writing this particular piece in the first place.

I imagine you either want to get ahead of the curve and set up your own VoIP stack,  or you are just curious what has been keeping my occupied of late. To be honest, tuning my setup hasn’t been even remotely as time consuming as writing up this post to accompany it. I was left speechless at the speed of deployment after said was done, and a brand-spanking-nagelneu PBX was successfully up and running in the cloud. Here is a brief overview of what to do. Don’t worry, you‘ll be home in time for supper. This won’t take long at all.

Before you get started, you will need a few things: first, you will need a PBX. I use 3CX, because it has great cross-platform support. Second, you need a SIP provider. Telnyx is my provider of choice, because they allow a pay-as-you-go billing model. And last, you will need a place to run your PBX server.

That said, before you really get started, if you are going to follow along, you will need to set up an account with 3CX and AWS. I will not be going through the setup of the SIP trunk, because you may find a better deal in your region with a different provider. Where I could give you a general overview of the SIP trunk setup, it’s best to follow the documentation from your provider.

There are a number of different locations to run a PBX: you can pay a company to host it for you, you can host it on premise, or you can spin it up in a cloud provider’s environment  (after all, cloud is little more than just someone else’s computer that you pay for the privilege of using).

I chose Amazon Web Services (AWS) as my cloud provider. It came down to the sheer quantity of relative, in-depth documentation. That documentation was well-needed when I first set up the PBX. And joining AWS meant I got a year’s worth of free-tier resources to tinker with for free. Luckily enough for me, the PBX is so light on resources, I was able to run it non-stop for about $0.80 per month.

Here is link to a tutorial on how the setup process looked when I first set things up.

That was a while ago—2 years. And a lot has happened in the world of computers since then. If you took the time to watch even half of Chuck’s tutorial, your head may be rightly spinning. But 3CX have made cloud deployment so much easier since then. I think I had the whole thing set up in about 15 or 20 clicks—you count and see and see how far off my click count was.

Happy networking!  Happy calling!

If you made it this far and like what you see

• please consider supporting us
• share this content wherever appropriate and applicable
• jump in the comments and give us a shout out
• and subscribe for more

Keep us brewing!

Nothing good in IT ever happened without coffee.Thanks for your support!

Buy us a coffee