Siacoin — Update I
TL;DR. Ecology:3 (4). Technology:5 (5). Decentralization:4 (3). Valuation:5 (4). Rating:4/10 (4/10)
Six months have passed since the initial BD Ratings article on Siacoin. Let's continue to explore the project.
2018-08-06: USD 0.007989, Satoshis 112.
2019-02-06: USD 0.002117, Satoshis 62.
To start off where the first rating ended, the Siacoin community were upset by many things, including delayed Obelisk miners, allegedly secret Bitmain ASIC mining, the general Sia development pace and the core team's past decision to not fork. To counter the external ASIC threat that became more and more unattainable, a community proposal was published, pushing for forks that temporarily shut Bitmain, Innosilicon and other non-Obelisk ASICs out from the network while also introducing a new block reward/tax for Sia development funding. The community mostly seemed to support the proposal, and so did Obelisk founders and employees. Sia lead developer David Vorick was skeptical however of the temporary-aspect of the fork and instead wanted to shut the foreign ASICs out indefinitely.
On the 6th of September, Sia 1.3.4 was released, bringing bug fixes, updates to the UI, and support for streaming multiple files at the same time. The fork debate continued and on the 1st of October, the Sia developer team finally came forward with their official statement, which confirmed that these core developers were proceeding with the forking of Sia. The fork itself would indefinitely brick all non-Obelisk ASICs through a secret mechanism on the Obelisk chips.
User FaustianAGI who authored the initial community proposal came out strongly against Vorick's modified version of it. So did many others, and on 13th of October 2018, author Scott Ellis published SiaClassic - Declaration of Independence. A non-profit SiaClassic Foundation had allegedly been created for the purpose of organizing SiaClassic development. Innosilicon announced their full support for SiaClassic, that also aimed to introduce a 10% development fee. Almost exactly in parallel to this break-away declaration, FaustianAGI proclaimed the planning of yet another fork, SiaPrime. SiaPrime were to break away on the 31st of October (the same date as the Sia hard fork) and included a small premine to the initiators, as well as a systemic mining tax of 15%.
Sia finally forked successfully, at block 179 000. All large exchanges kept this fork listed as Siacoin. In December 2018, Sia was supported by Ledger, the hardware wallet company. Work on the anticipated 1.4.0 version of Sia had started, where users could finally properly back-up data and actually retrieve it should their hard drive die completely.
The Sia subreddit is somewhat active, with some good discussions on Proof of Work and cryptocurrency economics in general. Obviously fork implications have been heavily discussed as well. Blog posts and continuous updates are posted by Sia developers and team members. The number of on-chain transactions has decreased since last rating, to around 3 000 per day.
SiaPrime is still a living project, which at least for now steals some economic activity from the main Sia chain. Activity overall remains low however. SiaClassic is active as well, but the general community activity is not impressive either. The original Sia team still operates on the most important chain. BD Ratings is also of the opinion that both forks were too greedy for long term success, which most likely means that original Sia will continue to dominate.
As development proceeds, BD Ratings is still interested in seeing how the promising features of the 1.4.0 client will impact the ecology overall. The Sia product will be extremely innovative and may very well draw more economic activity to the chain in the future. Higher on-chain activity would be some evidence.
Reasons: Quite low on-chain and community activity. Multiple forks of the projects. New client versions will soon offer more utility.
The main Sia GitLab repository is active as usual, with multiple developers contributing to the project.
Sia developers can still be assumed to be highly knowledgeable in mining aspects of a public blockchain. David Vorick has continued with more excellent pieces on the subject of Proof of Work, ASICs and the shifting landscape of rented hash rate. The ASIC manufacturing is obviously a learning experience as well.
There is still uncertainty with regards to the whole mining process of Siacoin. A Public blockchain needs network stability to function properly as a value protocol, and the viability of the consensus mechanism is part of the foundation for this stability. By bricking other ASIC manufacturers' hardware, the Siacoin mining itself seems to have gotten more decentralized. But the question is if today's setup brings long term stability or not. It is too early to tell whether the Obelisk monopoly was the best answer to rising ASIC threats, and only time will give more data on that point. BD Ratings is of the opinion that the move to counter foreign ASIC machines in this case was both logical and neccesary. With that said, it is no guarantee that the resulting status quo will hold forever as there might be unintended consequences.
Reasons: Active GitLab. Very good PoW mining expertise and a good developer team in general. Volatile mining situation.
One clear example of centralization around Sia is the supposed threat a non-fork would have lead to. If the network would not have forked and Obelisk went bankrupt due to strong competition, apparently that would directly have threatened Nebulous Inc. financially as well, with the possible result being Sia core developers dropping off the project as they no longer got paid. It is easy to see how these dependencies make the network more centralized and fragile. When a large user base threatens to sue a company affiliated with most of the developers, and this in and of itself leads to fork proposals, it is clearly a sign of centralization.
As already established, BD Ratings is not against defensive forks to crush miner monopolies. Miners are hired (often with the built-in block reward) to perform a service for the network; in other words, miners exist to work for the network and not the other way around. By crushing miner monopolies, the network makes sure to reward the actual efficient miners, where efficiency non-intuitively is defined not as hash power but the contribution to decentralization of the network. But in this case, Sia core developers pushed for defensive action with the knowledge of possible bankruptcy otherwise. In other words, their own company presumably gained financially on a favorable fork decision. This is a classical conflict of interest situation they put themselves in by forming the hardware manufacturer in the first place.
Considering the centralized forking decision, it was at least good to see the developer team mitigate any plans on adding a miner tax. Nebulous Inc. already has a built-in revenue stream through the renting-fee scheme as well as hardware sales. Adding yet another income stream, from PoW mining in this case, would not look good at all, and could have opened up for yet conflicts of interest. Worth adding is that although the forking decision had characteristics of centralization, choosing to not fork would have resulted in worse centralization. It was in other words a curious dilemma, where to counter a centralization threat Sia developers had to take a decision where the result was a different kind of centralization.
One last aspect of centralization that just recently came up is the distribution of hash rate between different mining pools. The Luxor pool clearly has more than 50% of the hash rate right now, and the reason people are not freaking out more is probably because the underlying hash is rather decentralized, combined with the fact that the pool itself is supporting of the Sia developer team. Miners that point hash rate to the pool can easily switch should any situation arise. This is the main point, that there are no huge, network agnostic entities behind the Luxor hash rate. Optimally, miners should leave the pool anyway.
Reasons: Centralized fork decision. ASIC treat mitigated. Legal entities being central point of failures. Mining pool centralization (with caveat).
In the first Siacoin rating article, the issue of high inflation was brought up. Nothing has changed in this regard. By 2026, annual supply increase of Siacoin will be around 3%, just to slowly pan out. This is not a competitive number as for example Bitcoin will have an inflation as low as 2% already in 2020.
The market capitalization of all Siacoin is at time of writing USD 82M. This is much lower than 6 months ago, also compared to Bitcoin.
Reasons: Inflation characteristics do not yet lead to store of value property. Too much volatility with the consensus mechanism. Price has fallen considerably.