Tag: technology

What’s the Endgame with AI

~ Veselin ~ 13 Comments

AI can write code. It can’t do 100% of the work, can’t even do 50%, but it can do a lot and is improving. It doesn’t get tired and doesn’t freak out when facing a new codebase.

It can write blog posts. Maybe not good posts but some posts that can fool some readers. It can also generate quantity that cannot be matched by humans.

AI can respond to support requests. Maybe not the greatest support on the planet but enough to be used by all Bulgarian telecom operators. It might be bad but it is also fast.

Given that 10 years ago, none of that was possible, if we plot a chart where 10 years ago we had nearly zero AI, and today we have some AI replacing humans, where does the chart go 10 years from now? Is AI becoming omniscient?

I recently read a 1962 book where one AI had the ambition to eliminate the entire human creativity (Gordon Dickson’s Necromancer). Gordon Dickson wasn’t far of from what LLM is doing at the moment. It’s hard to predict how far it will go without imagining some things it could possibly start doing and doesn’t do right now.

Here’s some AI engineering milestones to watch for.

1. AI, play me 5 new seasons of Wednesday

We should be close to that. Maybe a bit expensive with the tokens but what’s really missing to make it possible? It doesn’t even need a robotic body.

Speaking of bodies, giving it access to a printer or some other tools opens a Pandora box of possibilities.

2. AI, make me a sandwich

Why not? Building that may not even require AI. Most of the tech for it is available, perhaps some software and hardware is missing here and there but we can imagine seeing startups doing cooking bots in the nearby future. Cleaning and refilling the food toner would be very interesting challenges.

3. AI, make me a car

Okay, this is a tougher one. Lots of patents will be violated. AI doesn’t seem to care right now, and I’m sure there will be ways to circumvent intellectual property. Would that ever be possible? It should be. The AI may need access to some machinery but nothing that doesn’t already exist.

4. AI, make me a nuke

An AI capable of building cars would have no trouble producing weapons, and particularly copying and modifying existing weapon systems. I’m sure it will be used for weapons long before it’s used for cars. But what if this capability becomes available to individuals, not state actors?

And last but not least,

5. AI, print me some cash

The primary reason for not going down this ladder would be that the blueprints are protected, not that it can’t be used that way if it’s trained that way.

Overall, I think the development of AI presents bigger problems than humans becoming redundant, administrative bloat, and UBI. While we already observe a decline in all kinds of human activities that are being automated and made mediocre by AI, humans won’t stop trying to use it elsewhere. I see lots of room for changes that can damage the existing societal order.

Think Wrong, Move Fast and Break Things

~ Veselin ~ 6 Comments

I’m currently reading Careless People by Sarah Wynn-Williams. I’m less than halfway through, but it already feels like this book deserves more than one post. So far, it doesn’t paint Mark Zuckerberg and Sheryl Sandberg as supervillains, and I’m getting a glimpse into Facebook’s early culture.

One of the big ideas from Facebook’s early years was Move Fast and Break Things. This mantra has been both confirmed working and disproved many times – often by engineers like me, who’ve lived through its successes and catastrophic failures. .

Moving Fast and Breaking Things Works

It works because the software industry can be like a gator-infested pool. When a new idea drops like a piece of meat in the pool, everyone jumps on it. The biggest reward goes to the fastest gator that ships first and markets well. There’s often no time to make things well.

Facebook won the social network race in large parts of the world. Twitter and a few others got the leftovers. But this principle applies beyond tech giants – down to much smaller scales. It’s a form of the Pareto principle: 80% of the outcomes stem from 20% of the causes. If you can roughly identify the 20% and validate an idea quickly, you’ve already won even if it doesn’t work. You saved the effort for something that may work.

On an individual level, it also feels like it works. You get a task, you ship something quickly – it shows up in your weekly update, your team’s update, maybe even the leadership sees it. You’re productive, visible, and valuable.

But It Also Doesn’t Work

Once an idea is validated, it gains users, traction, and revenue. A bug that shows up once in 1000 runs might never happen with 10 users/day. Once you have a million users, it happens 1000 times a day. Also, one broken user profile may be easily fixable but a million? Not so much.

Zuckerberg himself cited this kind of thinking when Facebook moved away from the motto around 2014. You can’t keep patching the same issues over and over at scale. Stability becomes a requirement.

From an individual contributor point of view, it looks that profitable ideas attract many layers of heavily invested people – technical, marketing, finance, data, legal, executive, investors. And when something breaks, you’re not just dealing with bugs. You’re affecting dashboards, KPIs, morale, and your own job security. Blame becomes easier to assign. 10 of these people will know how things work and won’t blame you but the eleventh may have a bad day and push the button.

How to make a difference?

In early-stage product development or during moments of intense change, moving fast and breaking things can be the right move. But in mature projects, where uptime matters and stakeholders are many, the priority shifts. It’s more about stability, reliability, and trust.

Ultimately, Mark Zuckerberg hung that motto on Facebook’s wall – and eventually took it down. He may put it back up if he recognizes a need for it. Recognizing the moment is a key part of leadership.

Headphones for Calls – SteelSeries Arctis Nova 7x/7p

~ Veselin ~ 8 Comments

My one of my old JBLs broke and I needed a new headset for work. After a careful research, I ordered Arctis Nova 7x. The 7x is the XBOX version of the headset and differs from the other 7 by one additional button, which is irrelevant to me. I liked the green/black color.

So I got them yesterday and I decided to make some recordings and see if I can actually use them. The main issues I have during calls are two:

  • Can people hear and understand me loud and clear
  • Can I hear them. I’ve had headphones that were too quiet

As it turns out, Arctis Nova 7 produces a different result depending on what kind of connectivity I use. I tried the recommended 2.4GHz Dongle and a 3.5mm TRRS. They also support Bluetooth but I’m not enthusiastic about it. It’s supposed to come with more latency and connection woes.

Noisy environment

I went to the local Starbucks, where people chat, and the music is loud. The sound was sharp and clear. It can be so loud that it can be harmful, which is something I’ve not recently experienced with other headphones. It’s good and I don’t think I’ll have issues hearing people.

However, the mic captures too much background noise. I would need to rely on additional software removing that noise. I didn’t have that problem with the previous headphones. People speaking nearby would be mostly filtered out.

Quiet environment

The sound is great, particularly when connected with a cable.

Using the 2.4GHz dongle produces much louder recording. Unfortunately, all I can hear from it is SSssssSSS. Do I really talk like that? It’s not great.

When connected with the cable, the sound is a bit more muted and I find it tolerable, although chair movements and breathing can still be heard.

Overall, I think I can rely on them when doing calls from home but not when working from the outside. I’ll use them for a week or two and gather feedback. I feel like the microphone can be a source of discomfort during calls, mainly for the people who need to understand what I say. The cable would be the way to go for me.

Comfort and controls

  • The button placement is well picked and I can’t roll the volume or mute by accident.
  • The microphone pulls out and is great when hidden. However, I have my doubts about how long this pulling in and out can last before something breaks
  • The headphones feel very soft and comfortable when used with glasses.
  • The 2.4GHz dongle is very wide. When I plug it in, thy cover the next USB-C port and I can’t charge my computer. I think the dongle is not something I’ll be interested using. I may give Bluetooth a chance but not that thing.

Am I happy

Not quite. I can have a better experience listening to music compared to my 3 pairs of mildly broken JBLs but the microphone is a source of anxiety. I’ll give them a try and will post an update in a month or two.

UPDATE:

  • The headphones have a green light that blinks and is visible at night, feels like I’m on an airplane at night
  • They’re very good for music and movies, and I actually use the dongle for that
  • They seem to turn off when they consider are in idle, even if you’re actually using them
  • When the dongle is plugged, the audio is directed to it so switching between headphones and computer audio means clicking a navigation somewhere
  • Mic also needs to be manually changed (did I mention I love wired headphones? Everything is so simple with them)

UPDATE 2:

  • I got used to everything and kind of like them far more than the JBLs. The sound is better, the buttons and rolls just never click themselves, except the maybe mute button, which can be pressed in the backpack.

Charging Cables

~ Veselin ~ 14 Comments

I’ve not cracked the charging issue yet. My kids run through cables very quickly because they charge their numerous devices while playing games. They twist them, bend them, and use lots of power. We also need to maintain a collection of chargers with cables at the grandparents where they’re used by the niblings as well. The devices and chargers vary greatly and I’m always in the look for improvements.

10 days ago I bought a beautifully looking Energizer cable. 2m long, 100w, looked very, very solid. It stopped working after charging my computer and phone 5-6-ish times each with an Apple’s 45-watt charger.

I was sure something in that cable was fake and decided to slice and dice it to see proof of deception. Nope. You can pull a car with it. But you can’t charge a phone.

Took me significant effort with the pliers to cut through the outside. So I guess, just another brand that wasn’t meant to be.

So what somewhat worked:

  • The Apple’s Lightning and USB-C cables tend to last months to a year with our usage. The downside is that they’re bland, look dirty, subject to dry rot and twisting. I’ve stopped buying these as they are the highest price, lowest joy level, and medium durability for our use case.
  • Vivanco’s 1m cable is my personal favorite for a durable lightning cable. It doesn’t twist as much and is thicker. My kids have destroyed some of these while playing Minecraft but I’ve personally never broken one. It’s short and ugly, so it’s far from ideal.
  • XMart’s cables with textile braid look great, probably best in term of looks from what we use. They are also 1.2m long, which is better. I have an USB-C braided cable at one of my work spots, working with a 100-watt charger, and it’s fine after a few years of use. However, those cables with multiple ends my family likes? Not so much. I don’t know why we keep buying them.
  • Cellularline’s braided USB-C cable would be my top choice. The one I got is 2.5 meters long, about 1 meter longer than it should be, and just works. However, it’s never been put through the real Minecraft test because it’s USB-C and the kids use Lightning.

UPDATE (2025-01-25) Cellularline’s Lightning cable I bought to my kids failed after less than 10 days of use using a 65 watt Cellularline charger: Not ideal.

What do you use?

Zuck’s Moderation Changes And My Blog

~ Veselin ~ 26 Comments

Mark Zuckerberg announced that Meta is changing its moderation policy. The announcement was posted on Threads in 6 points and struck me as unusual. I went over these points several times and used ChatGPT to additionally separate the meaning from the PR. Came up to the conclusion that three things were announced and wrapped in zuckspeak:

  • Meta will cut moderation
  • Meta will relocate teams out of California and Texas
  • Meta will support a future initiative by president Trump

From the point of view of a personal blogger who actively uses Facebook, Threads, and formerly Instagram, Meta is my second-biggest source of traffic. I’m interested in them being successful and helping me succeed as well. Unfortunately, I’ve experienced problems with their services and think the problems are not going anywhere with these changes.

The problems that I’ve experienced with Facebook:

  • Publicize is no longer possible to my profile. I need to manually add my blog posts. This is likely to keep the traffic within the platform, which is something OpenAI, Google, and X also do
  • AI moderation bans/removes my blog posts, for example it banned this scary post and this photo of a bridge in my neighborhood
  • The moderation decisions are not reversible. Didn’t get a response to any appeal
  • Algorithmic deprioritization of manually shared posts if they contain external links
  • Propaganda/triggering articles reach my attention due to insufficient filtering of fake news
  • The response to any report I filed against hate speech was that that it doesn’t violate their guidelines (but my fitness achievements violated them)

So whatever moderation they do, it likely hides humans behind multiple layers of automation that leave bloggers vulnerable to frequent unfair treatment.

The problems that Facebook likely needs to solve and are lurking in the 6 changes:

  • Many parents from my generation consider Facebook and Instagram as not appropriate for children. We are raising a generation that’s banned from using these services and lives on Youtube, TikTok, and a few messenger apps. Some of these kids are already entering the work force and not spending any time or money on Meta.
  • Some social networks thrive because they allow X-rated content (Reddit, X)

My expectation for changes based on these observations

  • I think Zuckerberg expresses support to the new administration with a hope to get TikTok banned and Meta not banned, which can direct some young users to Instagram
  • I think they’ll let X-rated content slip in on all platforms to increase engagement, following similar moves by other networks, and label that a battle for the freedom of speech.

I don’t expect that any personal blogger will benefit from the changes unless they’re involved in whistleblowing/political criticism. Meanwhile, bloggers like me will continue to be banned when posting about outdoor walking.