Technology & Public Life on Arda Akgül

The Blockchain Problem of Bitcoin

Wed, 11 Mar 2026 00:00:00 +0000

I already invest in Bitcoin, so this is not a hit piece. If anything, it is the opposite. I think the strongest way to stay bullish on Bitcoin is to be honest about what it does badly. And for me, the biggest weakness is still the same one it has had for years: the blockchain itself is not a very efficient place to scale global money.

Bitcoin solved one problem brilliantly

The original achievement still matters. In the white paper , Satoshi solved the double-spending problem without relying on a central authority. That is not a small thing. Bitcoin created digital scarcity and a settlement system that can survive without a state or a corporation sitting in the middle.

That is why I still take it seriously. Bitcoin is not just another fintech app. It is a monetary system with a different trust model.

But the base layer is cramped by design

The problem begins the moment people try to confuse monetary importance with raw transaction efficiency.

Bitcoin’s base layer was never built to process the entire world’s daily retail activity directly. The Bitcoin developer documentation still reflects the basic constraint: block space is scarce, blocks arrive roughly every ten minutes, and under classic consensus rules a serialized block was limited to 1 MB. Even with later efficiency improvements like SegWit, the underlying design still forces competition for a finite settlement surface.

I actually think this scarcity is part of Bitcoin’s strength. It makes the chain expensive to attack and hard to bloat irresponsibly. But it also means the network becomes awkward the moment too many people want to use the base layer at once.

Fees are not a bug. They are also a problem

Bitcoiners often say high fees are healthy because block space is valuable. I understand that argument. In a narrow security sense, it is true.

But if ordinary users cannot move funds cheaply during periods of congestion, then Bitcoin’s social usability weakens. In bull markets, mempools fill, fees spike, and small transactions become irrational. At exactly the moment interest rises, the user experience often gets worse. That is not fatal for Bitcoin as a settlement asset, but it is a real problem for Bitcoin as everyday money.

Layer 2 helps, but it also proves the point

This is why the Lightning Network matters so much. Lightning promises instant, low-cost payments by moving activity off-chain and using the base layer as a security anchor. I am not dismissing that. In fact, I think Lightning is one of the most important reasons to stay constructive on Bitcoin’s future.

But Lightning also proves the core point: the blockchain by itself is not enough.

Once a system needs payment channels, liquidity routing, wallet abstraction, and extra coordination layers to feel consumer-friendly, you are no longer defending the base chain as a universal transaction rail. You are defending it as a settlement layer beneath a stack. That is a much more realistic framing, and I think Bitcoin would benefit if more people admitted it openly.

There is also an efficiency tradeoff

Bitcoin’s proof-of-work design gives it seriousness, neutrality, and costliness. I do not think those things are fake. But proof of work is still resource-intensive by nature. Even if you believe the energy criticism is often exaggerated or politically selective, the system undeniably converts real-world energy into monetary security.

I am not morally scandalized by that. Modern finance also has hidden physical costs everywhere. Still, if we are comparing architectures, Bitcoin is not an elegant high-throughput database. It is a deliberately expensive machine for producing irreversible monetary settlement. That is powerful, but it is not efficient in the everyday engineering sense.

Why am I still bullish?

Because I do not think Bitcoin needs to win every argument to remain important.

For me, Bitcoin’s strongest use case is not “Visa on-chain.” It is being the hardest, most credible, non-state monetary asset in the digital world. If that thesis holds, then the base layer only needs to do one thing exceptionally well: secure final settlement for a scarce asset people trust.

Everything else can be built around it, though not effortlessly. That is where I stay constructive. I do think Bitcoin can scale socially through layers, better wallet design, and more mature infrastructure. I just do not think the blockchain itself should be romanticized as if it has no tradeoffs.

That is the critical friend’s view I keep coming back to. Bitcoin is brilliant. Bitcoin is also clunky. The same design that makes it credible makes it hard to scale smoothly. I am still bullish because I think the monetary breakthrough is bigger than the engineering frustration. But the frustration is real, and pretending otherwise only makes the case weaker.

How Did Anna's Archive Scrape Spotify's Library?

Sat, 31 Jan 2026 00:00:00 +0000

When I first saw the claim that Anna’s Archive had effectively mirrored Spotify’s library, my reaction was not only amazement. It was recognition. This felt like one more episode in a very internet-native pattern: if a platform centralizes cultural access at global scale, somebody will eventually try to extract that access back out into an archive.

The metadata layer was the easy part

The least mysterious part of this story is the metadata. Spotify has long exposed huge amounts of track, album, artist, playlist, and identifier data through its developer ecosystem and platform structure. Even without downloading actual audio files, a determined scraper can map an enormous portion of the catalog by enumerating IDs, following artist-to-album links, tracking playlist relationships, and normalizing releases across regions and editions.

That is why I do not think the real technical trick was “finding the songs.” The real trick was scaling the crawl, cleaning the metadata, and linking it into a usable archive structure. Anyone who has worked with public web data knows this is less glamorous than hacking, but often more important. Classification is power.

The harder part was the audio

This is where the public evidence gets thinner, and I need to be careful. Spotify’s licensed audio is not openly downloadable through the normal developer API. So any claim that a full Spotify library was captured implies an additional acquisition layer beyond ordinary metadata crawling.

My reading of the available reporting is that the pipeline was probably hybrid. Public or semi-public Spotify metadata would have been used to build the map, while the underlying audio appears to have been obtained through other channels, likely a mix of capture, pre-existing scene sources, user-contributed files, or other unauthorized collection methods. The exact extraction workflow is not fully documented publicly, so that part is an inference, not a confirmed line-by-line description.

Why does this matter technically?

Because it shows the difference between access and ownership on the modern internet.

Spotify feels like an all-encompassing music library. But it is not a library in the archival sense. It is a licensed access platform. Songs appear, disappear, get region-locked, get reissued, or change in subtle ways users rarely notice. From an archivist’s perspective, that is unstable. From a platform perspective, it is normal business.

Anna’s Archive sits on the opposite side of that logic. Its whole worldview is that cultural material should be indexable, preservable, and portable even when commercial platforms prefer controlled access. I do not think you can understand the Spotify episode without understanding that philosophical clash. This was not just scraping for convenience. It was also a statement about who gets to preserve culture.

The ethics are not simple

I am sympathetic to preservation arguments. I also think people become too lazy when they assume “open access” automatically settles every ethical question.

There is still a difference between preserving metadata, preserving access to out-of-print material, and redistributing a live commercial catalog that is built on ongoing licensing agreements. Musicians, labels, publishers, and platforms all sit somewhere inside that economic chain. Once a shadow archive appears, the moral story stops being clean.

That is why I see the Spotify scrape as both technically fascinating and politically messy. On one side, it exposes how much of the modern cultural record sits inside private platforms that are not designed for long-term public memory. On the other, it raises obvious questions about consent, compensation, and the limits of archive activism.

What does this reveal about open access culture?

For me, it reveals that open access culture has matured beyond books and papers. It is now moving into subscription media ecosystems that were built on the assumption of permanent platform control.

Once that happens, scraping stops being only a technical act. It becomes an argument about whether the internet should preserve culture as a commons or rent it back to us through interfaces. I do not think this debate is going away. If anything, Spotify is exactly the kind of platform that makes it inevitable.

So how did Anna’s Archive probably do it? Not through one magic exploit. More likely through layered infrastructure: large-scale metadata crawling, aggressive normalization, cross-referencing, and some separate path to unauthorized file acquisition. The engineering story is real. But the bigger story, at least to me, is ideological. Archives are starting to challenge platforms on their own terrain.

Turkcell and Huawei's Integration on 5G

Tue, 20 Jan 2026 00:00:00 +0000

When people talk about 5G, they often talk as if it were just a faster version of 4G. I think that misses the point. 5G is not only a consumer internet upgrade. It is a strategic infrastructure layer. Once I look at it that way, the Turkcell-Huawei relationship becomes much more interesting than a normal vendor deal.

This is not a new relationship

Turkcell and Huawei did not suddenly discover each other because 5G became fashionable. They have been building together for years. Back in 2019, Turkcell announced that its network had become the first in the world to run fully on Huawei’s cloud-native core architecture. Since then, the two companies have kept deepening the relationship around 5G, artificial intelligence, energy efficiency, and what Huawei now brands as 5.5G.

At Mobile World Congress, Turkcell and Huawei have repeatedly used the partnership to present Turkey as a serious next-generation connectivity market rather than a peripheral adopter. I think that matters because telecom infrastructure is path dependent. Once a carrier builds operational familiarity, integration layers, and vendor-specific optimization into its network, switching becomes expensive, technical, and political all at once.

The technical issue is integration, not just equipment

This is why the Huawei debate is always misunderstood when it is reduced to antennas. A 5G network is not just radio equipment on towers. It includes the core network, orchestration software, edge computing potential, enterprise use cases, energy management, and long-term maintenance.

If Huawei is deeply present in those layers, then Turkcell is not simply buying hardware. It is building an ecosystem logic. That logic can be efficient. Huawei has scale, engineering depth, and a strong record in rapid telecom deployment. For an operator trying to control costs while keeping performance high, that is attractive.

From a purely technical perspective, I understand the appeal. Turkey wants higher-capacity networks, industrial connectivity, smart logistics, low-latency enterprise services, and stronger digital infrastructure. Those ambitions require partners that can actually build.

The geopolitical issue is harder

But the moment Huawei enters the story, technology becomes geopolitics.

The United States has spent years framing Huawei as a security risk and pushing allies to reduce dependency on Chinese telecom equipment. Europe has been more uneven, but the pressure is real there too. Turkey, meanwhile, is in a familiar position: a NATO member that still wants room to maneuver, buy, trade, and negotiate across multiple power centers instead of fully inheriting Washington’s technology map.

That is why I think the Turkcell-Huawei relationship says something broader about Turkey’s strategic posture. It reflects a country that still wants Western capital, Western market access, and Western security ties, but does not want to surrender every infrastructure decision to Western strategic preferences. In practice, that means balancing.

What does this mean for Turkey?

For me, the biggest question is not whether Huawei is good or bad in the abstract. The real question is whether Turkey can use external vendors without becoming structurally dependent on them.

That means Turkey needs more than a purchase strategy. It needs a capability strategy. Domestic software layers, domestic cybersecurity competence, stronger local telecom suppliers, and real institutional oversight matter more than slogans about sovereignty. If those are weak, then any imported 5G stack, Chinese, European, or otherwise, creates dependency. If those are strong, the country has more room to bargain.

This is where the Turkcell-Huawei integration becomes a revealing test. It shows that Turkey is serious about building next-generation infrastructure, but it also exposes how difficult strategic autonomy really is. You cannot claim technological independence while importing everything that makes the system function. At the same time, you cannot build a national network by pretending large foreign vendors are optional.

I think Turkey’s real path lies somewhere in between. Work with whoever can deliver. But build domestic competence aggressively enough that partnership does not turn into submission. That is the line that matters.

So when I look at Turkcell and Huawei on 5G, I do not just see telecom engineering. I see a live example of how middle powers try to modernize under great-power competition. And honestly, that is where the story becomes much more important than download speed.