3 ms·

>but that demand has been minimal so far. I am mostly anti-crypto but I always thought the concept of Helium was cool. If the "price" was competitive with LTE

>but that demand has been minimal so far. I am mostly anti-crypto but I always thought the concept of Helium was cool. If the "price" was competitive with LTE-M and NB-IoT (I'm not sure if it was, maybe it is now?) what stopped people from using it?

PragmaticPulp 4y ago
I've worked on low data rate projects that would be a good fit for the Helium network. Biggest issue was uncertainty about the future of the network. The only reason people were interested in running nodes was to collect payouts, denominated in the Helium token. If the Helium token price crashes (as it has), the network will stop expanding at best and start shrinking at worst. Meanwhile, regular cellular data contracts dropped to relatively low prices for low data applications. When retail pricing starts at $1/month, you can imagine how cheap we're getting contracts for 10,000 devices locked in over multi-year contracts. Also, the Helium founders seem to have realized they squeezed everything they can out of their first project, so they switched their attention to a new 5G project to try to repeat their token-fueled success once more before crypto speculation ran out of steam. They didn't quite make it.
crote 4y ago
Registration of a hotspot and asserting its location costs the miner 50,000,000 Data Credits, plus 2x65,000 DC in transaction fees. Registration fees for receiving packages has a fixed fee of tens of millions of DC too. Meanwhile, transferring a single 24-byte package over the network costs the user 1 DC - which is equivalent to $.00001. I live in a medium-density city, and there are 50+ miners within a half-mile radius from me. Considering IoT is designed for low-volume data, there is no way packet payout will ever come remotely close to the network fees itself. The protocol was designed to reward proof-of-coverage at first, slowly switching to rewarding data forwarding. What we are seeing right now is that the data volume simply doesn't exist, and the proof-of-coverage was showing a false demand. Even if it were to see widespread use, it would still not live up to expectations. And if you are a commercial user, why would you mess around with Helium rather than just contracting one of the well-known commercial IoT parties? Even if a commercial provider is 100x as expensive you are still only paying $.001 / package.
spaceman_2020 4y ago
Most crypto concepts are similarly cool, but hijacked by speculators and fraud.
- rchaud 4y ago
  > but hijacked by speculators and fraud. only after the non-feasibility of scaling the concept drives out the sensible people.
- VectorLock 4y ago
  I've been interested in cryptocurrencies back when they were still called "hashcash" so this whole corruption of the original ethos really hits me in the feels.
palata 4y ago
I remember reading a Tor article (blog post maybe? Cannot find it anymore) about the "revenue model", and why they settle on donations only (mostly?). Back then I was interested in the Nym model (a mixnet trying to use a "proof of mix" to pay the nodes for their work). It's been a while, so take it with a grain of salt. I found it very interesting to see that Tor had actually considered paying to encourage nodes (different technical solutions), but decided against it for a philosophical reason: if you run a node for profit, chances are that you don't really care about helping the project ("you just want money"). So it's a risk to attract "bad actors". And they don't want that. I think there is a similar argument to make for all those services trying to use cryptocurrencies to attract nodes: those "bad actors" will always optimize for their profit, not for the health of your service. Seems like it applies for Helium, too.
rchaud 4y ago
If you go on the Helium subreddit, there are many posts from people saying that their Helium modems took several months to ship, and the modems themselves are completely overpriced for what they do.

PragmaticPulp 4y ago

I've worked on low data rate projects that would be a good fit for the Helium network. Biggest issue was uncertainty about the future of the network. The only reason people were interested in running nodes was to collect payouts, denominated in the Helium token. If the Helium token price crashes (as it has), the network will stop expanding at best and start shrinking at worst. Meanwhile, regular cellular data contracts dropped to relatively low prices for low data applications. When retail pricing starts at $1/month, you can imagine how cheap we're getting contracts for 10,000 devices locked in over multi-year contracts. Also, the Helium founders seem to have realized they squeezed everything they can out of their first project, so they switched their attention to a new 5G project to try to repeat their token-fueled success once more before crypto speculation ran out of steam. They didn't quite make it.

crote 4y ago

Registration of a hotspot and asserting its location costs the miner 50,000,000 Data Credits, plus 2x65,000 DC in transaction fees. Registration fees for receiving packages has a fixed fee of tens of millions of DC too. Meanwhile, transferring a single 24-byte package over the network costs the user 1 DC - which is equivalent to $.00001. I live in a medium-density city, and there are 50+ miners within a half-mile radius from me. Considering IoT is designed for low-volume data, there is no way packet payout will ever come remotely close to the network fees itself. The protocol was designed to reward proof-of-coverage at first, slowly switching to rewarding data forwarding. What we are seeing right now is that the data volume simply doesn't exist, and the proof-of-coverage was showing a false demand. Even if it were to see widespread use, it would still not live up to expectations. And if you are a commercial user, why would you mess around with Helium rather than just contracting one of the well-known commercial IoT parties? Even if a commercial provider is 100x as expensive you are still only paying $.001 / package.

spaceman_2020 4y ago

Most crypto concepts are similarly cool, but hijacked by speculators and fraud.

rchaud 4y ago
> but hijacked by speculators and fraud. only after the non-feasibility of scaling the concept drives out the sensible people.
VectorLock 4y ago
I've been interested in cryptocurrencies back when they were still called "hashcash" so this whole corruption of the original ethos really hits me in the feels.

palata 4y ago

I remember reading a Tor article (blog post maybe? Cannot find it anymore) about the "revenue model", and why they settle on donations only (mostly?). Back then I was interested in the Nym model (a mixnet trying to use a "proof of mix" to pay the nodes for their work). It's been a while, so take it with a grain of salt. I found it very interesting to see that Tor had actually considered paying to encourage nodes (different technical solutions), but decided against it for a philosophical reason: if you run a node for profit, chances are that you don't really care about helping the project ("you just want money"). So it's a risk to attract "bad actors". And they don't want that. I think there is a similar argument to make for all those services trying to use cryptocurrencies to attract nodes: those "bad actors" will always optimize for their profit, not for the health of your service. Seems like it applies for Helium, too.

rchaud 4y ago

If you go on the Helium subreddit, there are many posts from people saying that their Helium modems took several months to ship, and the modems themselves are completely overpriced for what they do.