Is filecoin a $257 Million bubble or a pyramid scheme?

[synctext]

A critical analysis of a $257 million dollar ecosystem.

Create a survey paper within 7 weeks. Possible points to cover:

• Introduction
• 15 years of documented failure
  • first failed attempt in 2000, mojonation documented in a paper
  • Tor community various dead attempts and 2014 proposal
  • our horrible slow progress within Tribler after 12 years of work.
• analysis of launch publicity and social media discussions.
• Is the design technically feasible
  • brief technical overview
  • proof-of-space time properties
  • Analysis of Sybil attack vulnerabilities, DDoS, transactions spamming, and attacks in general.
  • Is the market part under-specified? Both in-chain storage and especially the off-chain retrieval market.
  • critical IPFS dependancy
  • the zero lines of code problem
• Is the design economically feasible?
  • is there a market balance?
  • will it be stable?
  • how will supply or demand disruption influence price?
• investors exclusion
• Is this a pyramid scheme?
  Already a reward for early investors.
  Use the scientific 1975 rule, "A Probability Model of a Pyramid Scheme__1975.pdf".
• Is it a bubble?
  • is there a realistic valuation for initial investors?
  • would a fork make all early investment worthless?

Disclaimer: we are not neutral parties. We developed our own Bandwidth-as-a-Currency vision many years ago. In August 2007 we released our first prototype together with Harvard University.

[mjuchli]

My response to an interview of Juan Benet:
https://www.reddit.com/r/filecoin/comments/71egv1/filecoin_vs_airbnb_and_a_simple_calculation/

[synctext]

'''
This in fact means, the $257'000'000 launched at the Filecoin ICO require 95'185'185'185 Gigabyte (95'185 Petabyte) of storage to be offered by storage miners. 
'''
nice methodology. Would be great to generate multiple such compares.

[mjuchli]

Almost done with Section 3 (ICO analysis) -- after studying legal documents for quiet a while. Here some serious facts about the Simple Agreement for Future Token (SAFT):

Reasoning about the worst possible scenario:

"...SAFTs will be used to fund the Company’s development of a decentralized storage network that enables entities to earn Filecoin (the “Filecoin Network”)."

Funds can be invested in whatever development they find to be appropriate.

“Dissolution Event” means (i) a voluntary termination of operations of the Company..."
Throwing the bat is an option.

"If immediately prior to the consummation of the Dissolution Event, the assets of the Company that remain legally available for distribution to the Purchaser and all holders of all other SAFTs (the “Dissolving Purchasers”), as determined in good faith by the Company’s board of directors, are insufficient to permit the payment to the Dissolving Purchasers of their respective Returned Purchase Amounts, then the remaining assets of the Company legally available for distribution will be distributed with equal priority and pro rata among the Dissolving Purchasers in proportion to the Returned Purchase Amounts they would otherwise be entitled to receive pursuant to this Section 1(b). "

Then, only what is left right before the Dissolution Event will be payed out the investors as well as the other owners (e.g. early investors/PL people).

... and there is more.

About to tackle technical weaknesses in this week.

[mjuchli]

Draft for today's meeting: https://github.com/backender/filecoin-survey/releases/tag/0.1

[synctext]

2012 Microsoft patent from Sven Seuken, worked with him on bandwidth-as-currency when I was at Harvard from 2007 onwards
'Recent' bubble: https://en.wikipedia.org/wiki/Railway_Mania

We are not economists. First focus on spending some time trying to really break their network using standard white washing, Sybil, poisoning and spam attacks. Re-use Go expertise.

[mjuchli]

Current progress: https://github.com/backender/filecoin-survey/milestones?direction=asc&sort=due_date&state=open

[synctext]

Related project:

• https://sia.tech/
• https://storj.io/
• https://maidsafe.net/

Spammy, clickbait-storyline, but still interesting read: https://tokeneconomy.co/the-analysis-filecoin-doesnt-want-you-to-read-e60d5243f17c

[mjuchli]

This is perhaps a bit off-topic but I was reading a lot about IPFS and I think it has the potential to become huge. If you think about it, IPFS could not only replace HTTP (sorry for my comment earlier this week, IPFS is obviously not built on top of HTTP) but also extend on the idea of static file storage and introduce Publish/Subscribe functionality. Hence, implicitly provide message queuing functionality for any given content. Since the IPFS filesystem applies versioning on top of the Merkle DAG. Looking at the from another perspective, every new version of a directory/folder could essentially represent the state of a topic/channel/queue/(whatever you wanna call it). Providing new versions to clients (e.g. pushing the new file/message) wouldn't even mean that nodes have to store the last consumed state for every connected client, instead, with a heavier, pull-based client, the client node itself would know its last consumed state and thus could just as for the next state (e.g. next message). After all, this is just some brainstorming from my side on IPFS level. Probably there is a much neater way of doing so, perhaps even on a lower level provided by libp2p.

There is already a lot of discussion going on about pub/sub here: https://github.com/ipfs/notes/labels/pubsub

In terms of Filecoin, this would open the doors for decentralized pay-as-you-go streaming.

[mjuchli]

Started creating an overview of according to their specs

[synctext]

Filecoin and others compared. Plus also links to our 2007 work.

[mjuchli]

Just stumbled over this one today: https://decentralize.today/decentralized-storage-wars-storj-v-sia-v-filecoin-v-maidsafe-27dc3d37434f.
Only discusses about token distribution but contains a nice comparison.

[mjuchli]

Current state of report to discuss today:
https://github.com/backender/filecoin-survey/releases/tag/0.2

[synctext]

• Spell checker ("incentiviszed").
• Bitswap lacks a real incentive mechanism, just local counters for now: BitSwap ledgers are open to manipulation ipfs/ipfs#26
• Yet another storage coin, now with "Like-button" model : http://www.flixxo.com/assets/docs/flixxo-white-paper_v0.4.pdf
• compare all storage tokens in a 2-column table.
• Genesis block owners become rent seekers?

[synctext]

use a Celeb to grow following and market cap: https://news.ycombinator.com/item?id=15609654

[synctext]

crashed ICO https://www.bloomberg.com/news/articles/2017-11-01/shining-star-of-initial-coin-offerings-crashing-back-to-earth

[mjuchli]

Draft without Abstract/Conclustion, pre-grammar/wording correction and reference numbering not ordered yet: filecoin_survey.pdf, see also https://github.com/backender/filecoin-survey/releases/tag/0.3.

My intention is that content is evaluated and perhaps improved. After that a grammar/wording revision will follow. The final version shall be ready by the end of the week.

[mjuchli]

Today on HN: https://medium.com/@dennyk/why-and-how-the-cryptobubble-will-burst-de9bc7fc5332

[mjuchli]

Tulip mania: https://www.coindesk.com/ubs-economist-compares-bitcoin-to-tulip-mania/

[mjuchli]

New and complete draft ready: https://github.com/backender/filecoin-survey/releases/tag/0.4
filecoin_survey (4).pdf

[synctext]

impressive work. good text

'''The fact that Protocol Labs has proven enormous technical capabilities by building IPFS and libp2p leads to believe that mastering the upcoming hurdles by building Filecoin will be handled in equal elegance.'''

polish:
Therefore, in our calculations we compare the valuation of both companies against each other by opposing their resources.

[synctext]

try to make a "deliberately simple model" instead of back of the envelope calculations and expand that section. Please read SIGCOMM fluid modeling work from 13 years ago : Modeling and performance analysis of BitTorrent-like peer-to-peer networks. Try to fit and apply this methodology for airbnb, eBay, uber, and non-market leader Lyft.

brainstorm on thesis direction: proof of limit order book fair allocation. Public verifiable correct market operation, fairness, and frontrunning etc. prevention. Fancy market orders.

Direct experience: running code for predicting Bitcoin trade direction (up/down) in 15min interval with 70% precision.

[synctext]

Essential reading for the oversupply concept inside BitSwap: Managing Supply and Demand of
Bandwidth in Peer-to-Peer Communities

"academia has been struggling" tribler etc., The last author has experience with distributed storage systems. In August 2007 we launch the first operational prototype of our bandwidth-as-a-currency vision.

Section: "D. Filecoin integration", within our own work on a macroeconomic sharing model from 2012 we showed that oversupply has a high impact on system fairness and efficiency. Oversupply is a critical design consideration which needs to be taken into account for the FileCoin "data trading module". TUDelft-oversupply-paper.pdf

Solid material for an intro @backender a .tex update coming?

[mjuchli]

https://github.com/backender/filecoin-survey/releases/tag/0.5

[mjuchli]

Paper regarding "trade execution optimization": https://www.cis.upenn.edu/~mkearns/papers/rlexec.pdf

[synctext]

ToDo for remainder of year: start of master thesis with 5 weeks of prototyping the Lehman Brothers 2006 paper.

[synctext]

please upvote 👍 https://www.reddit.com/r/filecoin/comments/7gmzds/scientific_study_is_filecoin_a_257_million_ponzi/

[mjuchli]

Also here https://news.ycombinator.com/item?id=15816729

…

On Thu, Nov 30, 2017 at 4:14 PM Johan Pouwelse ***@***.***> wrote: please upvote 👍 https://www.reddit.com/r/filecoin/comments/7gmzds/scientific_study_is_filecoin_a_257_million_ponzi/ — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#3097 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA6TK1QAzDAmE382GExo6t1TqsYFN3PWks5s7sZggaJpZM4PXs-M> .

[synctext]

Can you alter the ycombinator title to:
Scientific ICO study "Is Filecoin a $257 million Ponzi scheme?" Delft Uni. [pdf]

[synctext]

Or not alter the title, it works, on frontpage! What just happened????? 🥇

[synctext]

Filecoin is estimating that they will go live with a testnet Q4 2018 or early 2019. They made impressive progress, but security and complexity concerns have not been addressed. Their codebase is close for the following stated reason:

Avoid Distractions. Having an open, public codebase generates lots of questions, issues,
pull-requests, and other interactions, all of which require significant developer time investment
in aggregate. These distractions can be very frequent for projects with a lot of attention, and
can consume large fractions of our team’s time.

Sia testnet indicates there is no demand for decentralised storage: 4.1PB offered, 191TB, 95% of capacity is just idle! A throw-away price does not seems to fix this trust problem:

On average, Sia's decentralized cloud storage costs 90% less than incumbent
cloud storage providers. Storing 1TB of files on Sia costs about $2 per month,
compared with $23 on Amazon S3.

[ghost]

Any updates on Storj coin
They released Tardigrade.io

[synctext]

The promise of 99% profit, join NOW! The crypto bubble is still expanding at a "healthy" rate:

"Ethereum is 5.5 years into a 6-year planned upgrade to greatly reduce the costs",
"after the planned upgrade later this year, high profit margins (99%+)."
From: https://ethereumcashflow.com/

IPFS update:
"build a trillions scale distributed storage industry ecology.", the dream is still alive. China pours 100 million yuan into IPFS. Company IPFSUnion claims 1.3 $billion investment.
Central data hubs in China are central to IPFS, providing more evidence for our 'preservation of centrality' law. "IPFSUnion has expertise in IDC room deployment, large-scale server cluster architecture design and operation and maintenance, software algorithm optimization and FPGA & ASIC chip development and other fields." IPFS core location: Jiangxi Big Data Industrial Park, "the machine room can accommodate 100,000-150,000 servers."