[Maro]

What's the value of the in-app currency being a cryptocurrency (putting aside hype/pr), vs calling it a virtual currency, a Mysql database with a API on top?

Dropping to a technical level, if the miners are all run by Telegram, then I think it's just a pointless overhead over Mysql. On the other hand, if anybody can run miners, I don't comprehend what value that adds, what it means.

[goodside]

It makes it possible for the asset to be traded on third-party markets without the knowledge, consent, or control of the organization that issued the tokens.

It also helps you trust that movement of the asset will not be hindered by the issuer's inability to keep servers running in the face of potential DDoS attacks.

[Maro]

That's a good point. It assumes they don't control the miners, and don't demand "registration" to participate, which would void anonymity. It still sounds like they could use an existing coin though, no?

[CPLX]

Sure, but then they wouldn't get a few billion dollars of free money.

[goodside]

I'm not sure what you mean by "control the miners" — they're building this on OmiseGo, which isn't mineable.

All of the substantial ICOs require some level of KYC/AML registration for direct public sales and pre-sales. Tokens later become available to everyone else through independent exchanges, which happens quickly these days.

[mistermackle]

Where did you see that they are building on OmiseGo?

[philfrasty]

speculation at this point IMO https://www.reddit.com/r/omise_go/comments/7oif3j/telegram_i...

[45h34jh53k4j]

It will be an eth token like OMG until they launch their 'main chain and 2^92 sidechains' LOL! good luck with that

[lakechfoma]

I see this same thing happening at some corporations now trying to create blockchain as a service. If one corporation, or just a few, are the only ones participating in the chain then why not just use traditional DBs?..

[donkeyd]

Because blockchain is cool. The other arguments, however true, are not relevant for 90% I've seen it be proposed for. It's mostly about buzzword bingo in large organizations, where people want to be 'the one who proposed [insert bleeding edge technology here]'.

[rando444]

The main reason is because you can't edit a blockchain.

With a database there's always ways around editing the data if you own the entire system.

With a blockchain it's not possible to edit the data without replacing the entire chain.

[prostoalex]

If a database snapshot and a checksum are published out to a CDN on a regular interval, you can certainly run UPDATE on a row and publish a new snapshot, but those tracking the integrity of the data will detect the change.

If you are in control of write access to the blockchain, you can roll back to whatever block suits you better, and just fork it. Others might protest, but those fools have no write access, so they're stuck with your fork for the time being.

Blockchain is just a protocol for consensus-based DB replication, you don't somehow magically end up with a superior DB because of that.

[oh_sigh]

You could still use merkle trees it similar technology in your own database and publish checkpoint hashes of the DB every so often

[maksimum]

Then it seems like the data being hashed would be serializations of databases rather than sets of transactions. I.e. you would be publishing snapshots of the world rather than agreed-upon changes starting from a pre-agreed on state.

[odammit]

I’ve been curious about this. I’ve thought about it (inspired by event sourcing), but am not familiar with the Blockchain enough to know if it’s feasible from the trust/security side.

[woah]

At some point you will just be implementing your own half-assed blockchain because you "don't need a blockchain". Why?

[lozenge]

Because git is easy and proven and a blockchain is complicated.

[rlpb]

They could just publish a git repository. It'd be the same: it's not possible to edit the data without replacing the entire chain [of commits].

[zeroxfe]

It's primarily because large corporations within an industry don't trust each other (and IMO, shouldn't.) This is especially so when the corporations cross national boundaries.

[prostoalex]

That's quite a blanket statement. There's a bunch of cases where companies trust each other enough to share the data for the benefit of the industry - credit reports being one example. Other industries have bad customer lists, etc.

So let's say you do set up a meeting with Experian, TransUnion and EquiFax and now need to prepare a slide deck to convince them to migrate their existing tech (let's say a master MySQL with a bunch of slaves, shared write access for trusted entities) to blockchain.

What's your selling point? Lower costs when deployed at the datacenter? Decreased hardware costs? Greatly increased TPS? Faster lookups? Easier to build applications on top compared to the existing stack?

[fernandopj]

IMHO that's not how a disruption on that level would ever play out. If a future blockchain replace those companies/services, it will be (again, IMO) because a team decided to create a blockchain who would provide those services, and instead of trying to please the incumbents, it would be the lower tier providers who would see value on it (users, faster time to market, etc), and little by little it would compound into market disruption, to a point where those 3 incumbents would either join the train or launch a new stack. It's a scenario akin to streaming: it is a model so disruptive from what was standard, that every first mover got in trouble with hostile incumbents (Pandora comes to mind). It takes an Apple-sized company to launch an iTunes with blessings and partnerships of that level. Look how long it is taking Ripple (XRP) to effectively launch in a similar space.

[prostoalex]

I disagree on the Pandora analogy and hostile incumbents.

In corporate IT the mighty dollar rules the day. MySQL and Postgres had hostile incumbents, Linux and open source in general had hostile incumbents, commodity hardware for the data center had hostile incumbents, AWS (or Google Cloud) had hostile incumbents, Splunk, Cloudera, SalesForce and any other SAAS company had plenty of hostile incumbents.

Yet end of the day they were able to show cost savings, and either sell it at the CTO/CIO level to existing companies or enable startups with much more rigid cost base, which in big companies quickly got the CIO's attention.

There's nothing particular about blockchain tech that I can see that introduces a major game changer on the cost savings front. Launch two identical startups today, one on top of blockchain, another on top of MySQL. Will the first one have lower operating costs 1, 5, 10, 20 years down the road?

[avip]

I believe it's partly evidence-based stock ramping [1].

[1] Not sure that's the appropriate term.

[s_dev]

>What's the value of the in-app currency being a cryptocurrency

In a word -- decentralization -- is the key difference between a virtual currency described by some SQL tables and an API and one described by a blockchain.

A gov if it wants can take down a server if it is physically hosted in that country. A blockchain unless you've gargantuan CPU power is nigh on impossible to take down.

[CPLX]

It's not decentralized if one organization owns 52% of it.

[vernon99]

Not necessarily. This is actually orthogonal to decentralization question. Very much depends on their proof of stake implementation.

[maksimum]

> Very much depends on their proof of stake implementation

Exactly, if someone with >50% power can essentially double spend it makes the currency as a whole almost worthless. As a result it's very much in the creator's interest to either not have >50% or to come up with a distributed consensus algorithm where a majority party can't double spend (very hard).

[edraferi]

I agree. It's more like they're retaining control of the money supply like a central bank does. It give you the power to indirectly influence the price / exchange rate of the currency, but you're not involved with every transaction.

[hndamien]

Except that you are with Proof of Stake.

[jmcqk6]

>In a word -- decentralization...

That is the difference, but what is the value in that? You're making things much more complicated and for what - especially for an in-app currency.

[s_dev]

Redundancy -- you don't need to keep backups of a blockchain but you do if everything is centralized in a single place.

It depends on the value of the backups I guess and the total cost involved. If you're looking at store of value there needs to be backups of backups. I see room for both depending on the needs of the business. It's just a trade off -- the blockchain is inherently more stable and secure than a server for certain purposes and server is far faster and cheaper depending on how you use it.

[imustbeevil]

They can't make a cool 500 million from adding a row in their Users table for currency.

[maksimum]

I don't know if you've heard of the d2jsp forum. But the site's owner (he's essentially a 1-man operation) made a probably $10M+ by doing just that. Not sure whether this is possible (or optimal in terms of growth rate) at Telegram's scale.

[natalyarostova]

Hah, d2jsp... Haven't heard that name in a long long time.

[fastball]

Because if Telegram ever goes under as a company, your "virtual currency" doesn't disappear overnight.

[Maro]

Yes, but if the Telegram app/platform was the only place I could conduct commerce using their money, I'm still screwed, unless a "secondary" marketplace emerges where you can use it. In which case, why not use Bitcoin or Eth to begin with? (Genuine question.)

[ribosometronome]

>In which case, why not use Bitcoin

Is there anyone using bitcoin for transactions anymore? Fees are absurdly prohibitive.

[hndamien]

Crypto markets use it a lot.

[cantrip]

Just like if the US Government ceases to exist we'll still have all these dollar bills.

[shafyy]

Or if the world stopped to exist we also wouldn't need any currency. You know where I'm going. The default risk of Telegram cannot even be compared to that of the US.

[cantrip]

I honestly don't know where you're going with that argument.

My point was that if the central bank behind a currency goes away so does all of its utility, even if the currency is still physically or digitally there.

No one said anything about default. You can't default on a currency that isn't backed by anything.

[Jtsummers]

> My point was that if the central bank behind a currency goes away so does all of its utility, even if the currency is still physically or digitally there.

That's not true. Unless the bank is critical for conducting transactions, their presence is not required for a currency they produced/backed to have value.

Value in currencies is whatever people assign to it. Utility is whether people will accept it as payment. If the US government goes tango uniform next week, and I and those around me have enough cash available (let's also assume the banks and CC companies we rely on for card/digital transactions also disappear) we can use the remaining physical currency as our method of denominating our local economy, and likely would.

[fastball]

You can't send people dollar bills on Telegram... otherwise yes, that would be a great solution.

[UncleMeat]

And then nobody ever maintains the software and it still falls apart eventually.

[dpflan]

To add to your point, the article states that Telegram will try to retain a 52% stake in its economy to control it.

"Telegram will The white paper also makes clear that four percent of the supply of Grams (200 million Grams) will be reserved for Telegram’s development team with a four-year vesting period. Telegram also plans to retain “at least 52 percent” of the entire supply of the Grams cryptocurrency to protect it from speculative trading and maintain flexibility. The remaining 44 percent will be sold in both the public and private sale.

The currency will be listed on external exchanges and used inside the Telegram app."

[hndamien]

This is really the deal breaker. If they had a fair distribution function then I'd be ok adopting it. This is really one of the virtues of Bitcoin. Everybody had their chance - even after significant media attention - to help participate and bring about the change.

[olalonde]

You answered your own question. It's a borderline scam to make money. Otherwise, they would just use an existing coin (e.g. Bitcoin).

[ojr]

no one would be able to write a smart contract on top of your cryptocurrency if it is not on the same blockchain or follow certain standards cryptocurrency exchanges expect to have in a tradable digital asset.

For example in Solidity a language that targets the Ethereum Virtual Machine. You can have constructor parameters in your contracts like addressOfTokenUsedAsReward, then you provide the smart contract address ex "0x12dgsdg..." of the token you built. These contracts unlike your mysql database are immutable and the code can't be changed, even if there is a vulnerability in it.

[eitland]

Just like Filecoin and Ethereum this is a possibility to use cryptocurrencies for something smart and distributed instead of just as a settlement layer between banks.

If mining is somehow built on shuffling messages around this might be the good parts about old Skype (close to decentralized) without the problems (steals your bandwidth without asking).

That said: as with anything cryptocurrency-related I'm hugely sceptical.

[Eridrus]

Durov is basically a huckster who jumps on trends.

[659087]

It's a cash grab, just like 99% of the rest of the ICOs out there. They know there are lots of poorly informed, greedy greater fools trampling each other to give their money away right now, and plan to take full advantage of it.

[tluyben2]

You can wonder if that is good or bad. Well it is kind of bad but it is happening anyway and at least these guys won't just disappear with the pot overnight. I put some smallish amounts (like $100 per ICO) and in bunch of ICOs to see what would happen in the summer. At first all the negative stuff here on HN did seem to happen, now, more than half a year later, with some buy/sell actions at the right times, this only ICO portfolio is all black with a profit of over 30k. I took 50% to fiat and left the rest. The negativity is interesting on here: there are enough like me who know it's a gamble, who put experiments out with money they can miss if it is gone and who continue playing with profits only. Ofcourse we win only because your 'fools' lose, but we are on HN, you would think the people here would be in the former category. For now it is simply free money and where I live, legal and very low taxed. Besides some bad ICOs, which you can very easily recognize by the way, there are actual strong believers who fight for their company and reap the benefits. It is a bubble but not all is a fraud and ponzi in the blockchain space. Many are seeing it as a good way to run a real startup with real plans and products.