Good work from a friend ...

Its fully Confirmed now That Anthropic is adding Ethiopia to Supported Countries List!!!

Its rolling from Claude code first!

No more vpn blah blah 😊

After dinner let us not go to bed but to prayer, or we may become more irrational than the irrational beasts.

St. John Chrysostom

Lets pray before we sleep 🙏, one አቡነ ዘበሰማያት would suffice

In all you do, remember the end of your life, and then you will never sin
Sir 7:36

Let us start our day remembering this

We reached 300 🎉 hoorayy ....

Thanks @all ... and @byte_philosopher almost ~ 45 people joined because of u ...

Man-in-the-Middle attack happens when a hacker secretly intercepts communication between two parties without them knowing.

You think you're talking directly to someone…
But the attacker is sitting in between, watching and sometimes changing everything.

https://github.com/jajos12/basic_red_teaming

back then just for learning purposes I was experimenting with red teaming attacks ... u can see the details in the repo and one of them was ARP spoofing ... it is a basic MiTM attack ... and it goes as follows:

ARP (Address Resolution Protocol) bridges Layer 2 (MAC addresses) and Layer 3 (IP addresses) of the network stack. When your computer wants to send a packet to 192.168.1.1, it needs to know the MAC address of that IP on the local network. It broadcasts:
"Who has 192.168.1.1? Tell 192.168.1.50"
The device at 192.168.1.1 responds:
"192.168.1.1 is at AA:BB:CC:DD:EE:FF"
Your OS stores this in the ARP cache — a table mapping IP addresses to MAC addresses. All subsequent packets to 192.168.1.1 are sent to MAC AA:BB:CC:DD:EE:FF at the Ethernet layer.

The Fatal Flaw
ARP has zero authentication. The protocol was designed in 1982 (RFC 826) when networks were trusted environments. There is:

❌ No verification that the sender is who they claim to be
❌ No cryptographic signing of ARP messages
❌ No way to distinguish legitimate replies from forged ones
❌ Devices accept unsolicited ARP replies (no request needed!)
This means any device on the network can lie about who owns any IP address.

There are about 10 attacks described there ... and disclaimer it is for learning purposes only ❗️😁 I mean it ...

I intentionally left the 7 uncommited but u can experiment with ARP, DNS spoofing and credential sniffing ....

If u have found something useful don't forget to give it ⭐️ https://github.com/jajos12/basic_red_teaming

https://youtu.be/xmkSf5IS-zw?si=Cu83rwaIZB9SzovV

Haven't seen it but it seems a different format ... like with technical rigor I will definitely see it ... those of u comfortable with the basics should check it out ...

This is so shocking

@byte_philosopher

I enjoyed A2SV(I was Generation 6) - but when I was leaving it - u know it was a shock also for me ... bcha Emre tchilaleh ... but u are surrounded with ignorant ... bcha mnm albel

Claude think I am a robot because I am from Ethiopia .. 😂 it looped 5 times asking me again and again lol ...

VPN the life savior be it "app unavailable" or "u are robot" 😂

መልካም ሰንበት 💛

እኔ ተኛሁ አንቀላፋሁም፤ እግዚአብሔርም ደግፎኛልና ነቃሁ።
ከሚከብቡኝ ከአእላፍ ሕዝብ አልፈራም።

መዝ 3 :5-6

@byte_philosopher

https://youtu.be/tpIctyqH29Q?si=eKoLPExAXx6aiHF4

This is an insane playlist for anyone from scratch to understand how computers work in macro level ... and u know literally it is for everyone ...

The jealous person would prefer to suffer innumerable troubles rather than to see his neighbor in good repute, even if the cause of the good repute were to benefit him also. What could be more wretched than such a person?

St. John Chrysostom

https://www.coursera.org/learn/build-a-computer

This is also very valuable material ... u will build ur own computer ...

I will take it ... I think my financial aid have expired(after 6 months it expires) and I should use another acc to apply ... bcha It is insane like it is project based betam ena at every module they give like ~ 2 hrs for lecture and ~ 3 - 6 hrs for implementing the things learnt

now someone's gonna say ... "why even bother learning this stuff when ai is just gonna surpass us anyway"

bro. if we don't understand how things work at a fundamental level, we don't have a long future in this field. ai doesn't make deep knowledge useless — it makes it more valuable

like gpu design, computer architecture ... these things directly affect how ai is built and how far it can go. the people who will actually push things forward aren't just the ones using the tools, they're the ones who understand what's underneath well enough to rethink the tools entirely

we need to go beyond surface level and start thinking like researchers. and that starts with actually knowing how things work

Me talking this having OS exam tomorrow 😭

have a nice day ለማንኛውም 😁

was studying page replacement algorithms and had a thought ...

quick context first: your computer's RAM is limited. the OS creates an illusion that every program has way more memory than physically exists — it secretly stores some of it on disk. but when a program needs a page that's currently on disk, the OS has to bring it in ... and if RAM is full, it has to kick something out first

which page do you kick out? that's the page replacement problem

the classic algorithms (LRU, CLOCK ...) are hand-crafted heuristics. I bet many of u know abt LRU - it evicts the least recently used page basically ...

the theoretically perfect one — OPT — evicts whichever page won't be needed for the longest time. but it requires knowing the future lol

so I was thinking ... why not RL?

— state: recent page access history
— action: which page to evict on a fault
— reward: +1 hit, -1 fault

RL could learn to approximate OPT from patterns in the workload instead of needing oracle knowledge

turns out this is a real research area called learning-augmented algorithms — you take something like LRU and plug in an ML predictor on top. good predictor → approach OPT, bad predictor → fall back to LRU gracefully

OS exam tomorrow and my brain is doing research ideation 😭