My Thoughts About GOLEM & Its Future In Cloud-Computing

in #cryptocurrency7 years ago (edited)

When Golem's ICO was launched a few months ago, their target of 820,000 Ethers (~9 million USD) was met within 30 minutes. Unfortunately, I missed the window and have been eyeing on this project since its first appearance as a proposal back in Ethereum's DAO days, before the unfortunate exploit.

As Poloniex have added GOLEM (GNT) recently and turning its market-cap into ~23 million USD at time of writing, I've scooped up some although I think it's still quite overpriced for not much to show yet. But holding onto long-term, as usual. While GOLEM's version of a decentralised internet is still a huge technical hurdle, I personally like their team and their futurist approach to branding / marketing - that's what really matters to me as it will attract minds and investments.

Here's what I love about the project:-

  • Amazon's AWS business is turning into a multi-billion, perhaps hitting trillions in the near future, so GOLEM has a great chance to take part with the growing needs of such services.

  • Simply put, any projects that require massive computations could tap into GOLEM's network to get stuff done. For example, if PIXAR ever needs to render their 3D stuff faster without all the additional hardware, computational power can be rented from the network. This will go far beyond computer graphics - think about science, machine-learning, live market predictions, and microservices.

  • If GOLEM succeeds, anyone will be able to auto-assign their computing devices to rent itself out to the GOLEM network. This is one future promised by blockchain dreams. Anyone will be able to take part and earn out of having indie computer farms (on my side, I've a few contacts of internet cafes to buy off cheap 2nd hand computers if they're upgrading!). For casual users, perhaps it's something that could subsidise their device / electricity costs. On the flipside, both casual users and professionals can also tap into renting more power if they so desire (more below).

  • Future consumer devices may only come with screens and connectivity modules. Just connect to something like the GOLEM network and computations will mainly be performed by participating computers in the decentralised network. No need to upgrade hardware and software on user-side. Perhaps the cost of computation will be tied in as a telco subscription - who knows? Total speculation here, although cloud-based game streams (without the need of specific gaming consoles) are already on the rise.

Here are some of my doubts:-

  • I'm still trying to wrap my head around latency issues, an important question if the network is to be tapped by a market that demands almost-instantaneous computations. There's still a need to detect and validate correct computations, and these computations are not exactly as simple and rigid, predictable computations. These computations are wide-ranging and it's just about anything really.

  • Earning power of network participants may not be lucrative at all, which may be a good thing (?). And consider this: if the network is ordering near-instantaneous computational results, there's bound to be a need for massive redundancies. Perhaps the network may require something like 50 worldwide participants calculating and sending out the same stuff just to manage any latencies, disconnections, errors, etc. This is in the realm of smartly-automating delegation of processes in the network. Let's hope they'll be able to sort out this wizardry, or else a centralised solution may be better to quickly secure computational tasks and results.

Conclusion

All in all, I'm quite bullish on this project, although I will admit my naivety in believing that any technical issues can be solved. I remain hopeful for the future of blockchains, a future where anyone is self-enabled by technology to conduct their own business without great, resource-intensive intermediaries. Renting out idle, general computing power is a step above crypto-mining and in my opinion, should be on your radar. It's actually like having a general-utility bot working for us.

Do you have thoughts and criticisms about the project? Let's hear it out in the comments below.

Check out their website: https://golem.network
Whitepaper here: https://golem.network/doc/Golemwhitepaper.pdf


Disclaimer: I have invested in GOLEM, post-ICO. This is just my personal opinion on the project. All images screencapped from https://golem.network

Sort:  

Thank @kevinwong you for this Informative article.
I've actually written a post about this:

https://steemit.com/cryptocurrency/@pps/golem-cryptocurrency-to-solve-computation-power-tasks

Perhaps an interesting read.

just wondering, how did u discover this old post?

Do you know when Golem will go online?

There has been a flurry of projects recently with dubious ICO takings.

The notion of raising 9 million usd in ETH in 30 minutes is alarm bells.

Large holders of ETH create these ICO's that they can buy up themselves. Then, they have control of an overwhelming amount of the coins that give them options to make really easy money through moving the markets around at will.

Regardless of the merits of their proposition, the pre-sale has all but tarnished my expectation of this project. Ultimately I think it is just a medium to transfer wealth back to them with a pretty looking coin, and plausible (but unproven) narrative.

Thanks for voicing your concerns over the project. Indeed a good point to take notice :)

Anyone will be able to take part and earn out of having indie computer farms (on my side, I've a few contacts of internet cafes to buy off cheap 2nd hand computers if they're upgrading!).

If it's in any way popular, it won't be worth running old computers as their efficiency will not be good enough. There will be a cut off in terms of power, below which the return will be too small to consider, but the real decider will be efficiency.

So basically how much energy (and thus cash money cost) do your clock cycles cost? The 2nd hand computers might not cut it.

I think it's really misleading of some of these companies that describe it in terms of using your "unused" processing power. That so-called "unused" processing power is just unused capacity. To actually use it costs money, it's not for free as implied.

I once bought 42 computers for less than 350 USD, last year. Might be more than trouble's worth setting it up, but the actual use are the other parts, other than the cpu..

So basically how much energy (and thus cash money cost) do your clock cycles cost? The 2nd hand computers might not cut it.

Agreed on this point - thanks for the reminder, this should be my 3rd doubt about the project.

Wow, that is pretty cheap! 😮 Nice find.

We see it with Bitcoin, it's really hard to get a foot in the door. I think these projects are probably more suited to entrepreneurs who are going to put in a significant investment, than your average computer user who doesn't use their computer that much or that intensely (in terms of processing).

The thing about the volunteer projects like SETI, Cloud Folding, etc. is that these are really donations to the project. I can't see the same thing working when applied in a for-profit venture.

I'd love to be proven wrong though, these things are really inspiring as a concept. When MaidSafe gets off the ground soon that might work better because unlike Bitcoin mining and GOLEM type systems, it's storage based instead of processing based, though obviously still uses some processing power. From their own forum though it appears to be the same, medium to large scale operations will really only be able to be competitive.

Prove me wrong kids! Prove me wrong.

Is there enough personally identifiable information of those in control to hold them responsible if they abscond with the money?

When you're talking about this amount of money, call me crazy but I sort of want the long arm of the law to be available.

That's a very pragmatic and prudent approach. But it could also come into way of creative freedom, so I'm alright whichever way, as long as the community around such projects are also able to work on it.

Makes sense. In that case it's basically activist investing; trying to fund projects you believe in while hoping to make a return at the same time.

I've just been burned enough by "anons" to be wary. Not sure if you were in the scene back then, but I lost a few BTC to TradeFortress back in the day. That's actually one of the factors that went into me selling out of cryptos for several years.

Oh I did mine some BTCs back in 2010 (or was it 2009) then I completely forgot about cryptos until ethereum came along, so yup, kinda missed a lot of the early day shenanigans. But i dont think such attempts are going away..

Yeah, I'm pretty wary of anonymous players at this point. I think that Bitcoin is the future and that regulation still has some part to play in that future.

Gridcoin has been rewarding general computing power (harnessed through BOINC network) for years. BOINC has over 4 million registered users, with 15 million hosts.
While Golem's distribution was based entirely on an ICO, Gridcoin is minted through BOINC computations (which can be virtually any kind of distributed computing process) + POS rewards.

I looked at Gridcoin and I am afraid it is not for the majority of people who not very IT literate.

Damn, there are so many coins xD

Golem sounds interesting. I still can't fully wrap my head around how it works but if it does what it plans to that's pretty cool. I could rent some of my computer power when I'm sleeping or at work. I should look a little deeper.

I bought about 3,000GNT when it was about @05 higher than it is now. sniff sniff. and I think it will drop down quite a bit before it settles much lower. I am hoping for a brief spike.

I think @honeybee @personz 's concerns are pretty legit. There's quite a lot of uncertainties about the project, pretty much like so many cryptos lol. Check out their comments :)

This will change the world in so many different areas! Exciting discovery. Now, all I need is a 3d printer to make my own Bronze Golem.
3dp_freeformmetalprinting_printing.gif

3d-printing is still ways off.. need something fast and able to handle a variety of materials. But then again, just technical issues I guess :)

in b4 next ICO for a 3rd printing blockchainized project. :P

There was one proposal in that nature too, back in the DAO days. Haha

3d printers are getting faster and larger. This link below is from 2014 and is amazing
https://www.theguardian.com/technology/2014/jan/29/3d-printing-limbs-cars-selfies

I got some of these and Elastic too. Hopefully at least one of them gets it right.

Blockchains, powered by dreams :p

Coin Marketplace

STEEM 0.28
TRX 0.13
JST 0.032
BTC 60918.56
ETH 2919.11
USDT 1.00
SBD 3.56