⚠️ RBZL, the author of SiaSetup, has been banned by Sia's unstable project management due to our criticism of the Sia/Skynet projects.
We'd suggest that you read our criticism, which includes details on how the ban came about, and avoid the Sia and Skynet projects as a result. ⚠️

Concerns about the Sia and Skynet Projects

After several years of involvement, we question the viability of either project.

Jump To A Section

Preface

About the author, this critique, and where it comes from.

Last Updated January 4, 2021
Jump To A Section

As explained on the About SiaSetup page, I (RBZL) am the sole author and maintainer of SiaSetup.info, and have been around the Sia project for several years. I found the project interesting enough in 2017 that I started this website, which grew to be the best general-purpose resource on the Sia project since that time. I have no doubt that I've put thousands of hours into SiaSetup and helping others get started with Sia or help troubleshoot issues. I'm not looking for a pat on the back - I'm simply outlining that I have spent a lot of time learning the details of the Sia project, talking with team members and other community members, and getting a good sense of where Sia is likely to be headed. And despite all that time working to improve the project, I've now been banned from the Sia/Skynet social channels, and effectively, the Sia project for posting this page of concerns - yikes! 😬

I don't hold a significant amount of Siacoins, I own no Siafunds, and I have no other real stake in the project - in fact, I lose money operating SiaSetup as donations are basically nonexistent. I did purchase an Obelisk SC1 to support the project, and more or less broke even on that endeavor. Overall, I have simply been around as a hobby, hoping that the project takes a direction which is likely to make it successful. As a result, I believe I am well-qualified to make an assessment on the viability of the project and explain the ongoing issues with the project's various components. Unfortunately, because I feel this way about the project, I will no longer update or maintain SiaSetup either, so this writeup serves as my explanation why. We'll start with a general overview of the history of the Sia project, and break down the issues with each component of the Sia ecosystem afterwards and list the problems each one faces.

Disclaimer

The information expressed below is the opinion of the author of SiaSetup, and it is based on publicly available information for Sia and Skynet, including public conversations on the Sia Discord and Sia Forums. The intention of creating this opinion is to explain why we don't feel that Sia or Skynet are viable long-term projects, and therefore to provide our justification for discontinuing updates on SiaSetup. No part of this opinion should be taken as investment, financial, or other advice for the purposes of making such decisions. Please review all available public information on Sia and Skynet before making a decision to use or not use these products.

Too Long; Didn't Read Version / Summary New

If you don't want to waste about an hour of your life reading through all of our concerns and skip a bunch of crypto drama, here's the short version for you:


A Brief History of the Sia Project

The short version of how Sia came to be, and the events since.

The Sia project was founded in 2014 by David Vorick and Luke Champine, with the creation of Nebulous Labs. The Sia whitepaper describes the basic premise of the Sia project - a decentralized storage marketplace where users can buy and sell storage. A blockchain was created from scratch and implemented in order to faciliate payments for storage using a cryptocurrency token, both to implement smart contracts which distributed payment based on the outcome of a storage contract and to make Sia more accessible worldwide as it wouldn't be tied to a centralized fiat currency. Further mechanics of the Sia product are explained throughout the rest of the SiaSetup website, if you are unfamiliar with how Sia works. Funding for the project was raised by selling Siafunds, discussed later, and further funding has been raised traditionally in exchange for equity in Nebulous Labs.

One of the distinguishing features of the Sia project is that early on, it was one of the few projects in the crypto space which was actually functional. You could acquire some Siacoins, download Sia, and upload your data to the Sia network, where it landed on hard drives around the world operated by people also running Sia. If you wanted to sell your storage space, you could acquire some Siacoins for collateral and add your storage space, and you'd receive renter data and eventally Siacoins. Many other crypto projects throughout the late 2010s were scams in comparison, and simply never materialized despite raising millions of dollars in ICOs. While having a functional product might be a low bar, Sia was indeed a cool concept at the time for no other reason than the fact that it worked and it was indeed totally decentralized.

In 2017, when Bitcoin and other altcoin projects were gaining a lot of attention, Sia also gained attention over the summer when the coin price shot from well under $0.01 USD to around $0.10 USD. Many people discovered the project through news articles on it, though many of those articles were a result of the coin price increase. This brought an influx of new users interested in the technology, including the creator of SiaSetup. Several community members made contributions to the project, either through contributions to development of the core Sia product or through websites like this one which provided alternate support for Sia. Several community members also spent hundreds or thousands of hours helping new users get started and sort out their issues through Slack or Discord.

However, development of the core Sia product was performed at a snail's pace, and no significant features or improvements were made throughout most of 2017 and 2018. Additionally, the Sia team decided to undertake the manufacturing of an ASIC (a Siacoin mining device) called Obelisk in order to attempt to improve network security. Having no prior experience in this field, the project encountered a number of setbacks such as being over time and budget, having manufacturer issues in China, releasing a device with specifications lower than advertised, and having a competitor device hit the market almost a year before Obelisks were released. A controversial fork of the Sia blockchain to kick all ASICs other than Obelisk off the network ensued, causing community division and a number of forks to offshoot Sia (which ultimately died or went nowhere). The main Sia fork resulted in a return on investment for Obelisk buyers, as otherwise they would not have been able to compete against all other ASICs on the Sia network. It was expected that this move might destroy Sia and Nebulous - however, the Obelisk project didn't seem to have much effect on Sia, and the team continued on with future plans.

The biggest Sia feature release since 2017 has been Skynet, which is a separate component of Sia that allows content stored on Sia to be served through a web server called a Skynet Portal. It was released around Feburary 2020, and can be read about in our Skynet section. Since Skynet, a number of changes have been proposed and are slated to go into effect, including the creation of a nonprofit entity called the Sia Foundation to shift core Sia development off of Nebulous Labs and to move co-founder Luke to the Foundation. The Foundation would maintain the core Sia product, and Nebulous Labs would rebrand to Skynet Labs and focus on Skynet development. The Foundation would be funded by a block subsidy from mining rewards, while Skynet Labs would focus on monetizing Skynet to users.

Today, the focus appears to be the development of Skynet and attempting to shift users and developers to Skynet instead of the basic Sia product. However, after hundreds of additional hours of involvement and discussion on the plan for the future of both Sia and Skynet over the last several months, we are convinced that neither of these products will be viable under the current business and development strategies. Our detailed analysis on each area is explained below.


Viability Issues with Sia

The Sia project has remained in a half-finished state for years.

Sia is essentially what you'd get if you combined a torrent client with encryption, sharding and a blockchain to pay people for storage in crypto. It's a new product, but not necessarily a novel one, and whether or not it gains adoption remains to be seen but so far the answer seems to be "no". There have been a number of issues with Sia which have been reported by community members repeatedly, such as issues with hosting, accounting for funds, or general wallet usability, which have either never been addressed, are always promised to be worked on later but have not been touched after years, or are totally breaking issues for the product. Specific examples are below.

Sia Isn't Easy to Use, and the UI is All But Abandoned

Sia is a sensitive piece of software. Users are discouraged from running it on Windows or using the Sia-UI, though it seems a good chunk of users would prefer to run it that way considering that as of November 2020 there have been 938k downloads of the UI against 135k downloads of the command line version. Sia also tends to be more stable on Linux as a command line app, but this is not friendly to many non-technical users. This presents a high barrier to entry for the average user. While there are some fundamental architectural differences between operating systems, such as how Windows allocates new sparse files which affects how Sia operates in some tasks, overall there shouldn't be any reason for significant instability between operating systems so long as these differences are taken into account. Additionally, Sia is a memory and resource hog when performing several tasks such as synchronizing with the network, completing file contracts, or other background tasks, making a system nearly unresponsive and showing a lack of consideration for the fact that a user may be running Sia alongside other apps.

Equally importantly, the Sia-UI (as in the graphical user interface version of Sia) is majorly broken. It does not show what is happening in the background most of the time, so users are left wondering if their wallet has any funds after they load a seed because the wallet scan continues in the background and is not yet complete - but there is no indication of it. It is not made clear that Sia needs to be fully synchronized before any transactions can be made or will appear. There is no Skynet support for anything in the UI, despite Skynet being the new and current arm of the Sia project. It is not made clear that your seed is your default wallet password when you create a new wallet or import a seed to restore it, confusing users which have set a password elsewhere before. The UI Terminal doesn't work properly - it doesn't accept any path with spaces, even if the path is wrapped in quotes, and output is improperly formatted and appears jumbled. Transaction types are not well explained, such as "Setup" transactions which combine outputs into a larger balance to make spending coins in the future easier but confusing users as to why they see several 0.09 SC transactions coming out of their wallet. There is no significant help built into the UI, nor is there significant documentation on it elsewhere (further discussed below). Various bugs and states can cause the UI to not open, open to a white screen, or open to an error message which gives no clear indication of how to resolve it.

These are only a few of the UI issues which have persisted for years. There appears to be no plan to improve any of these issues, despite years of complaints and a constant stream of users asking for help in Sia's Discord channel over issues which waste community and support staff time and resources when they could simply be resolved with some basic UI improvements. It is an embarassment to the project when users are referred to a third-party web wallet over using the official UI in most cases because the third-party option (the Sia Central Lite Wallet) works so much better than the official one, and even then unfortunately a web wallet doesn't help renters or hosts, but only people dealing with basic transactions.

Renting and Hosting Aren't Reliable and Break Often

Sia renting is complicated, unpredictable and unreliable. There's not much explanation on what your allowance is or how it works, or how to check it and see if your funds are exhausted which prevents contracts from forming or renewing. It's not clear exactly where your funds are going and what portion will be returned if unused. Contract counts can drop for no apparent reason at all, along with file health and redundancy, resulting in lost data. Sia must be opened every month or so to repair your data because hosts are unreliable and if too many disappear, your data may be lost. This also introduces additional expense if hosts keep dropping out and you need to replace them. File uploads hang or get stuck for no reason, and are not prioritized in any particular way. There are still technical limitations to how big a file must be, and individual files smaller than 40 MB are padded to that size which increases costs significantly if not zipping smaller files up (pricing issues are futher discussed in another section below). Local renter metadata must be backed up after each and every change (i.e. new upload) to Sia, so it is not truly a cloud storage solution - without your local metadata as a renter, your files are inaccessible. You can back this up to your Sia storage but it also costs you money to store that information, and again, the backups need to be done manually every single time you change anything with your Sia storage, including after opening it and letting it repair files. Forget once, and you have potentially lost data if your computer crashes or some other event occurs in the future which prevents you from accessing your original Sia installation.

Sia hosting is borderline broken. Accounting is broken, and expected revenue and contract counts climb artificially until Sia is restarted. Restarting a Sia host is dangerous - if a host doesn't shut down properly or becomes corrupted and an immediate backup of host metadata isn't available, the host is irreparably broken and all collateral and income will be lost. A bug with the Bolt database software used to keep host records also has the potential to corrupt host data in some cases, making hosting a total loss and requiring a host to start over - such a common occurrence that it's simply referred to as the "Bolt Bug". Hosts often lock up and become unresponsive when performing random tasks like processing contract renewals, performing storage proofs, or adding or removing hard drives, and this results in failed host benchmarks and potentially lost revenue and collateral. The few Sia improvements which have been made over the last few years have been focused on the renter experience, and hosting improvements are always promised but never materialize after years. Hosting is a technical endeavor which requires a lot of babysitting, and is not friendly to new or existing hosts. As with the UI, third-party tools like the SiaStats Host Monitor have stepped in to fill some of the gaps in host functionality and reporting by showing host usage stats clearly and benchmarking hosts for performance regularly - creating a reliance on a third-party tool for many renters and hosts in order to functionally use Sia. Even then, because Sia locks up regularly for many hosts, failed third-party benchmarks and failed contracts often leave hosts scratching their heads as to what went wrong.

Documentation is a Mess

Sia documentation is still lacking and scattered. While improvements have been made since 2017 and there is now at least some official documentation which is helpful in the basics of getting started, this is the original reason SiaSetup came to be, and it is still superior to official documentation. There's no official straightforward guide on how it all works, how to get started in detail as a renter or host, or of things you need to be aware of in order to not lose data or money on Sia. Any documentation which is created is soon forgotten, and left to become out of date within months. It's also not made clear anywhere just how experimental Sia still is. Community options for documentation have been attempted in the past, such as the Sia Wiki, and this option is floating around again, but historically nobody has wanted to take the time to keep documentation up to date because changes to Sia are made so suddenly and without any warning that it's nearly impossible to keep up on all of them without making it a full-time job and reading every conversation which occurs in the Sia Discord every day. This is an impractical way to maintain documentation, and is a basic thing which needs to be improved before many entities will take Sia seriously.

Hosts Have Liability Questions

Sia hosts have always had questionable liability over what they store as hosts. These concerns are likely to amplify with Skynet, where a host may hold a whole unencrypted file and serve it to a portal or public users, exposing the host as being the one that holds a file that may infringe copyright or be illegal in some other way. There is no contingency for this aside from hoping that hosts are not targeted in the future by copyright and criminal complaints, but hope is not a great plan. Hosts can also delete certain sectors of their host storage, but then they lose out on revenue for that storage as well. If hosts begin receiving many complaints, hosts are not going to want to deal with the potential liability anymore. Liability issues have generally remained untested on Sia, but with Skynet and the public nature of content being shared, we expect that this will soon change as authorities and copyright groups catch on to Skynet and how it works. We expect that these groups will operate their own portals and hosts in order to discover copyrighted or illegal content, and to start going after all public portals and Sia hosts holding or serving the content similar to how they issue mass notices to torrent users.

Lack of Adoption and Progress After 6 Years

Sia host network capacity has been stagnant since around June 2019, at around 2 PB. The number of hosts on the network has also been stagnant, with just under 300 hosts accepting new storage contracts and taking on new data. This includes after the launch of Skynet - it seems people are not interested in hosting on Sia. The target market for hosts is also unclear - it's been mentioned previously that enterprise datacenters are expected to add their spare capacity on Sia, but we find this unlikely as the amount of babysitting required and the amount of income received is not worth the trouble at scale. It also seems that home users are not terribly interested in hosting, as running the Sia client is complex, buggy, and requires dealing in Siacoins and posting collateral. Used storage and user adoption have increased slightly, but not at the rate you would expect after 6 years of work - probably due to all of the usability issues explained above.

Sia development has always been slow, but with Skynet it has essentially been halted, and there is no organization around development or release cycles. This is covered more in the section on the Sia developers below. Plans which have been announced around the Sia Foundation to take over Sia development also have very long timelines, such as a year to implement UTreeXO which would reduce the blockchain size. We wonder if hosting and renting would have a chance at being mostly bug-free by 2030 at this rate.

Conclusion

In general, since the launch of Skynet, the core Sia product and client has essentially been abandoned, with the hope of moving users to Skynet instead. There are likely several reasons for this, such as moving Sia core development to the Sia Foundation, the fact that the Sia revenue model has shown to be ineffective, and the fact that the new focus on Skynet is the latest and greatest plan for growing revenue. However, many community members have been and continue to be unhappy that not only has Sia suffered obvious issues for so long, but that now it is being left broken to focus on yet another different profit-generating endeavor instead of trying to adjust the core Sia revenue model and get things working completely on the Sia end first. The fact that this is the state of Sia after six years of full-time development is telling.


Viability Issues with Skynet

Skynet still relies on elements of the normal Internet to do what it does, but makes it more complicated.

Skynet is not much more than a web server stack using Sia for storage. However, it is extremely restrictive, as Skynet Apps ("skapps") must be designed as client-side apps to run in the browser, have no central server capabilities like MySQL, and are currently not much more than a collection of computer science fair projects which demonstrate basic concepts. Specific issues are explained below.

Portals Introduce Centralization

Skynet portals are a piece of centralization in the Sia ecosystem. The developers claim this is not so because you can access any Skynet content through any portal, and you can even run your own portal, but the primary revenue model going forward is to sell monthly subscriptions to access one single portal at premium speeds and gain access to more storage. These subscriptions will be tied to a specific portal, so in that sense you are centralized in which portal you must use to access Skynet unless you want to remain a free substandard user which is throttled or endure the expense of running your own portal. This concept seems completely counter to the complete and total decentralization mantra that the core Sia product ran on before Skynet appeared. When using a centralized portal, the portal can also see what content you access (unless it's encrypted, which on Skynet it is not by default) and portals also log IP addresses and keep a number of other metrics.

Another issue with the portal model is that if you're not interested in operating your own portal (as many average users will have no interest in doing), you pick a public portal to use or subscribe to, and then your preferred public portal goes offline, if you've conducted all of your business through that portal and uploaded items using it you've then lost control of your data. The portal is the owner of anything uploaded through it, and so you're trusting that portal to keep your content online for you. If a portal server is shut down by the government or seized or blacklisted, you're effectively hosed on anything you've committed to that portal. You could also be traced to anything you uploaded or accessed via that portal based on access logs and IP addresses. You could switch your subscription to another portal, but now you're essentially trying to stay one step ahead of the next shutdown, you're still trusting a portal to maintain your data, and you're also probably starting to feel like you're running from the law with all of these shutdowns and switching around. Finally, the portal owner could decide that you're a problematic user, just like any other centralized service, and terminate your account and delete all of your content as a result. That doesn't sound like a great user experience for a typical general Internet user to us.

Monetization Plans are Out of Touch

The general plan for Skynet going forward is that both Nebulous Labs and other private parties will run Skynet portals, which they can charge subscription fees for (explained more below). The idea is that the subscription fees will be sufficient to cover portal operating costs, but it would require somewhere around 25 users to subscribe to each portal at $5/month to reasonably cover portal costs in reality. Anything less than this, and a portal operator is running at a loss, as Skynet portals require dedicated hardware and a minimum of gigabit bandwidth to function effectively. Portal operators are also expected to run the default portal stack unmodified so that all portals have identical functionality, which limits a portal operator in customizing their portal to make it more appealing to users than any other portal. In reality, we don't expect that there are going to be a huge number of portal operators in the first place, as it is such a niche endeavor and finding 25 users willing to dedicate $5/month to your portal for eternity could be a large challenge. Currently, there are only two portals which are actually fully functional after nearly a year of Skynet existence - one run by Nebulous Labs, and another run by a community member which also does his own development on Skynet.

The monetization strategy is not well thought out for Skynet, and is therefore pretty ridiculous. The general assumption is that many people will be willing to pay a monthly subscription to access one single Skynet portal at something like $5/month, which will get them faster speeds, some storage space, and some credit for items they view which are monetized by the content's creators. Functions called registry access are another expensive component, and will be limited in number per month. The development team currently plans to sell subscriptions to their own portal, take a cut of other portal subscriptions, and take a cut of payments made through skapps (Skynet apps) to content creators. Your $5 of credit is like a prepaid balance where you have a certain amount of SC to spend during uploading to or browsing of Skynet each month. The team argues that many things we use are pay-as-you-go, like mobile data, electricity and water, but this is a step backwards for something like browsing and interacting with the Internet - we suspect few people in 2020 are not on an unlimited data plan of some sort, and electricity and water are poor comparisons because those utilities provide you with a tangible item which has a real production and transport cost. We doubt the crew would be cool with pay-as-you-go home Internet service or data caps, and we seriously doubt that a significant number of Internet users are going to pay a monthly fee for a system which provides isolated access to apps which are significantly restricted in their architecture, and which is essentially billing you in the background as you do things. It is nearly so out of touch with reality in 2020 as to defy logic that it is actually being pitched seriously.

Along the same monetization lines, the strategy for ensuring that Nebulous receives a cut of other portals' revenue is also half-baked. Payments made via the fee manager will be built into the core Sia protocol, so the team can ensure they receive their 30% cut of those payments to content creators and the like, but payments for portal subscriptions made on other portals seem to have no accounting plan. They insist that they will be able to tell if cheating is occurring based on blockchain activity and host activity, but we are not yet aware of a way to know which addresses belong to which portals or users beyond perhaps what the CIA has devised. A straightforward, foolproof system like a per-user licensing system is resisted by the team as being a viable alternative, even though thousands of Internet companies implement this strategy relatively successfully. We anticipate that this won't become an issue anyway because we don't expect a significant number of people to pay for portal subscriptions in the first place, but it is yet another example of a lack of a properly thought out plan for the future of Skynet and monetization.

Skynet Doesn't Do Anything Better. No, Really.

The development team seems to think that Skynet is going to take over a large portion of the Internet with decentralized client-side applications where users pay a monthly fee, and they say it is because Skynet is better. This is a silly claim - Skynet does not do anything better than the regular Internet, it simply complicates and restricts things in the name of having a small amount of decentralization so that you can hop between portals. Skynet still relies on Internet protocols, traditional server hardware, and traditional web server software, and gives you only client-side apps which have no capability to use any technology currently popular on the Internet today like database software, PHP or other server-side scripting, or anything of that nature. Skapps are basically static web pages with some Javascript or similar WASM languages to perform various functions client-side, and upload or download information from Skynet. The "SkyDB" system is a simple key-value store, and not a true database like a SQL server as one might expect. The architecture is extremely restrictive, and skapp developers typically need to build their applications from scratch in order for them to function on Skynet. Additionally, the team seems to think that all Skynet users will be willing to install a browser extension in order to use Skynet, or to use it more securely, despite insistence from the community that this is not likely to be the case.

Another claim as to why Skynet is better is that it offloads the infrastructure onus from skapp developers. While this is true, it doesn't create magic infrastructure that has no cost. There are still dedicated servers running to support Skynet, which are not cheap and need to be paid for somehow. Either a portal operator subsidizes this cost (along with bandwidth and storage costs on Sia) out of charity, or the cost is passed to Skynet users. The team claims that one portal should support about 1,000 users, but we find this unlikely because with the current portal usage of what we estimate to be a few hundred community members consistently using Skynet at most, all portals servers are already struggling, filling up, and needing to be rotated out - even with a load-balanced main portal consisting of several servers in the background. To help pay for infrastructure, at one point there was a proposal for the Sia Foundation to subsidize portals, but this has apparently been overturned as the community opposed Nebulous or other parties making money off portals which were operated at no cost to the operator, or at the expense of Sia miners - but even with a subsidy, there's still the issue that somebody would be paying for the portals, just not the user or the developer, and this would not scale infinitely.

Note: The team has reached out to us to say that our estimate of portal usage is inaccurate and that they have seen 80,000 unique visitors to SiaSky.net based on IP addresses with 10k users before a portal server needs to be switched out, but we suspect that the majority of these are one-off visits from a link shared or posted somewhere, or search indexers - our statement that portals are struggling under load of a few hundred community users referred to regular, active users uploading and downloading to Skynet consistently, as would be expected if Skynet became mainstream and free users continued to strain portals as they do today.

Finally, there is nothing to indicate that future Skynet apps will not become as centralized and data-driven as traditional websites and applications. The architecture of a skapp is up to the developer, and a developer could easily introduce other centralized features, advertisements, data collection, moderation and content restrictions, and other measures which can be built into any website. Skynet does not prevent any of this from happening, despite what seems to be the general sentiment right now that Skynet means everything built on it will remain free, open-source, anti-centralization and anti-data collection. We expect that if Skynet and skapps ever did truly take off, developers would exploit any means possible to monetize them, including all of the traditional means used on the Internet that we've just mentioned.

None of this is better than the extremely developed and diverse architecture of the Internet and web technologies in 2020. In comparison, Skynet is complicated, impractical, and if we get down to it, most Internet users really and honestly don't care about decentralization, so it is unnecessary. At this point we believe that it is a waste of development resources for anyone to build anything on Skynet because nearly anything built on the Internet has already been done better elsewhere, granted with some of the "evils" of centralization. Additionally, we believe that users are not going to tolerate the cost of infrastructure shifting onto them, as we explained in the section above - users would rather see ads than open their pocketbooks. We can only assume that anyone impressed by what Skynet and skapps can do after understanding the underlying architecture must be new to Javascript applications or the Internet in general.

If You Build It, They Aren't Guaranteed To Come

Assuming that everything we've said above isn't necessarily true and Skynet is a perfectly suitable new development platform which successfully shifts the infrastructure off developers and all that, and several totally amazing skapps are built which provide functionality nearly identical to other large successful platforms, there's no guarantee that anyone is going to make the switch and start using it. This problem exists even without the limitations of Skynet - a perfect example is Google Plus, the social media platform created by one of the largest tech companies in the world. They had a huge existing userbase that already had Google accounts, and they built a polished and functional product, and had a marketing budget... and Google Plus didn't end up overtaking Facebook. Why? Everybody was already on Facebook, and to leave it meant leaving everyone else you interacted with on Facebook, all of your previous posts and content, and other majorly enticing features like Farmville and Farmville 2. Building a better mousetrap on the Internet isn't enough - users need a reason to switch over to it. This is a problem that exists independent of decentralization and monetization - ask the general public to pay a monthly fee and deal with something involving cryptocurrency on top of that, and you've introduced additional complications to widespread adoption even if you manage to build a killer skapp.

Who Knows How Long Your Content Will Stay Online? Possibly Not You

Skynet portals pin and pay for files uploaded through them, so portals have storage costs for them indefinitely as long as those files are kept online. However, if portals change this in the future to only keep a certain number of files online or expire files after an amount of time, users expecting that their files or skapps would remain online are liable to lose their data as a result. A plan has also been floated to only keep files which are accessed a certain amount, which would make Skynet useless for backups, i.e. in the case of users being driven away from Sia to Skynet. The plan for all of this in the future has not been well communicated, and it is likely that either portal expenses and storage will increase to an unsustainable level if all data is kept indefinitely, or many users will be disappointed when they return later to find that their files have been removed without warning (or with a warning posted on the Skynet portal, which they may not have visited and seen in time).

There is currently no warning when uploading files via a Skynet portal that files may someday be deleted, nor is there any obvious information given on any sort of file retention policy. We expect that this will generate disappointment and frustration on behalf of both users and developers in the near future when old content is likely to be pruned. It is also counterintuitive to remove skapps which may be rarely used but useful, and which do not cost much to keep online, as they expand the Skynet skapp ecosystem and may result in dead links if posted in places like the Skynet App Store by the developer and the link is not removed once the skapp is gone. Of course, the plan is that the file retention policy will only apply to free users, so you can pay a monthly portal subscription if you don't want to lose your data - but then you might as well use any other more friendly backup client or free storage option, depending on your use case.

Portals Have Liability Issues Similar to Web Hosts

Portals have large liability risks - they are a free for all in terms of uploading data anonymously. Data on Skynet can even be uploaded directly via a Sia client to get a skylink that will work on any portal, and this uploading can be done completely anonymously and with no restrictions. This is no doubt going to result in a ton of content on Skynet which is illegal or malicious. Portal operators are going to have their work cut out for them trying to police content, as there is no way to prevent anyone from uploading content from their own Sia node so abuse can only be cleaned up afterwards, never truly prevented. The team argues that many services have this problem, but most services can perform an IP ban or ban an account to curtail abuse. This is not the case with Sia and Skynet - it is equivalent to Google giving you direct access to a hard drive to upload whatever you want, and not having any way to know who is uploading it or keep them from doing it again and again. We suspect that as a result, Skynet portals will end up blacklisted by a number of services (hosting providers, domain/DNS registrars, browsers, antivirus/malware scanners, etc) due to distributing malicious or illegal content, and in fact some of these blocks have already occurred. Any ZIP, EXE, or other select file type downloads from the main Skynet portal, SiaSky.net, are currently blocked by Google Chrome as being dangerous, even ZIP files containing the official Sia releases - which is somewhat ironic.

In order to help block malicious content, one idea which has been floated is the ability to share blacklists between portals so that items which are discovered through one portal can be shared with other portals so that portal operators do not need to duplicate work in dealing with complaints, and to help all portals abide by the law. However, a community blacklist feature is not a great solution because one bad actor operating a portal can add items to their blocklist which are not objectionable and which they simply wish to block access to. Even if not done maliciously, one portal operator may add something to the blacklist which they consider objectionable but which other portals would be fine with serving. This will create an uncontrolled patchwork of blacklist contributions with varying criteria for each blacklist item, and the potential for legitimate content to be blacklisted by all portals due to a conservative or malicious actor.

Conclusion

Skynet could have remained a component of Sia for sharing files publicly, but instead it has been branched off and pitched as some superior type of Internet when really it is more equivalent to an old GeoCities setup with more modern Javascript apps, limited client-side functions and pay-as-you-go architecture. We can only assume that this was the direction which was taken because the core Sia product has proven that it will likely never be profitable, explained below further under Siafunds. With the recent taking on of additional venture capital, we assume that investors wanted a more solid plan for monetization of the ecosystem, and monetizing Skynet instead of Sia is the best that could be produced for the time being. Any other explanation would make it seem like the team is completely out of touch with reality in terms of how anything with the above attributes has any chance to be a multi-million or billion dollar success. We don't intend to be harsh with this assessment, but making a claim of the magnitude that a product will change the Internet forever should receive appropriate scrutiny and criticism when it comes with the plan we've described above.


Viability Issues with Siacoin (SC), Siafunds (SF), and Storage Pricing

Low SC prices, no SF returns, and unpredictability in storage pricing are primary issues.

We believe there are issues with the token system and storage pricing. To be fair, we have always advocated that Siacoin price is fairly meaningless in a product where the coin is a utility token - the token is intended to be traded for storage, so in fact it is better if the price remains stable because renters and hosts know what they are going to pay or receive. However, the Siafunds mechanic has all but proven to be a failed idea, and the storage marketplace continues to be unpredictable and will be a major hurdle for any adoption.

Siacoin Price Shouldn't Matter, But It Doesn't Inspire Confidence

Siacoin prices have been stagnant below $0.01 for the entire existence of the project, with the exception of summer 2017 to summer 2018. An artificially large price pump occurred twice during that period, not associated with any announcements or other actions. In fact, any announcement of Sia news or listing on a new exchange has had zero effect on the price, suggesting that nobody is following the project, nobody finds it viable, or nobody cares. Again, we don't think price stability is a bad thing from a storage perspective, but a coin price of around $0.0025 average for the last two years solid doesn't really inspire any excitement from others in terms of the project. This is due to the ridiculously high supply of 45 billion coins and rising, but it still makes for a low market cap and what most would consider to be a small project in comparison. The fact that storage price functions can be performed with the precision of 24 decimal places of a Siacoin begs the question of why the token supply is so high.

The Potential for Price Swings Makes Buying and Selling Storage Risky

The fact that Siacoin prices are potentially subject to such wild and random fluctuation (i.e. November 2017 at $0.003 to January 2018 at $0.111, around a 30x increase) is not inspiring for using the network as a renter or host. If you're locked into a 3 month storage contract and the price increases 30x during that time, as a renter you're going to feel that you overpaid significantly for your storage, as now your host is getting 30x what they originally would have in USD. If you're a host and the price then decreases 30x as it did after the peak, you're going to feel that you basically gave your storage away - unless in any situation you price your storage in Siacoin with no thought to the USD value, which is also silly when operating with real-world costs that you need to cover. Additionally, these moving price targets mean that hosts must continuously adjust their pricing in SC if they want to maintain a certain USD income, and also try to project out 3 months and guess what the market will be at that time. Many renters and hosts will not be interested in attempting to hedge against storage and SC prices. To add further complication, some larger renting entities using Sia storage have released their own public "price targets", saying that if hosts are priced above these values they will not be selected, adding one more thing hosts need to keep a regular eye on if they want to maximize their utilization.

The storage marketplace aspect of Sia has shown that prices will shift to whatever extremes the market will bear. Sia's "target" price of $2/TB/month has been out of reach for the majority of this year, as hosts have figured out they can charge more and still be utilized. This is likely due to a number of factors, such as the small hosting market (again, fewer than 300 hosts accepting contracts) and the fact that Skynet portals are using host space fairly indiscriminately. The target price of $2/TB/month also only applies to the core Sia product, as Sia uses 3x redundancy but Skynet uses 10x. In reality, storage prices have been closer to $3.50+/TB/month and Skynet prices closer to $10/TB/month. This is also before bandwidth and contract costs. There are many other services available, including large cloud providers, which can beat these prices and include a number of better features as well.

Cryptocurrency and the General Public Don't Yet Mix

Dealing with Siacoin in the first place is a high barrier to adoption for the general population. Those involved in crypto projects tend to forget just how little the general public knows about crypto, let alone things like how the Internet works - most of them are not that technical and don't care to learn. It has and will remain a high barrier to entry for renting and hosting on Sia, as getting Siacoins is either a complicated process or an expensive one. Skynet alleviates this somewhat as users will not be required to operate in Siacoins directly, but because we don't find the Skynet model viable either, the cryptocurrency aspect of Sia remains an issue. While the team could integrate fiat payments into the core Sia client (as they have with SiaStream, a Sia client reskin focusing on home theater/media PC users), traditionally they have been opposed to any such actions in the core Sia software as to avoid any point of third-party reliance or centralization. In any event, it is clear that the strategies over the last six years haven't been working to onboard general users between the difficulty in using Sia, the fact that it's still so unstable, and the fact that getting Siacoins is one more hurdle to being able to use Sia to rent or host.

The Siafund Mechanic Appears to Have Failed

Siafunds are a token similar to a stock - 10,000 of them exist, and they pay returns based on the use of the Sia network. They collect about a 3.9% fee on all storage contracts and distribute it among the 10k Siafund holders. Siafunds were initially sold to raise capital at the beginning of the project, and since then they trade privately for between $5,000-$8,000 USD on average. However, they return about 1 SC for every 100 TB of usage on the Sia network - so today in November 2020, Siafund holders are making about 7 SC/month on their Siafunds. That's a whopping ~$0.02 or so a month! At this rate it would take them over 20,000 years to recoup a $5k USD investment. Sia storage usage would need to hit around 800 PB/month for Siafund holders to make about $250/year at current prices, which would only take 20 years to recoup costs. Nebulous holds the majority of Siafunds (about 80%) as the original plan for funding the company long-term, but after 6 years of work for $0.02/month/Siafund, it seems pretty obvious that this funding scheme has failed and/or the project didn't take off as expected.

Siafund holders typically price their Siafunds in BTC, which is irrational (as it is for many BTC-altcoin pairs) considering how much the price of Bitcoin fluctuates. It's pricing a moving target with another moving target. Typically this works to the advantage of the SF holders, because when they purchased Siafunds at 0.5 BTC years ago, Bitcoin price was much lower. Maintaining the price at 0.5 BTC now because "that's what it's always been" or "that's what I paid for it" is in reality a pretty big increase in value based on USD. If BTC was $1k in 2017 and it is $18k today in November 2018, SF holders are valuing their Siafunds at 18x what they paid for them, which is ridiculous. We can only assume that this is a scheme to mask the increase in USD price, to attempt to collect a large return in the sale of Siafunds, or because Siafund holders don't want to accept that their Siafunds are essentially worthless beyond what they can resell them for.

Conclusion

Overall, we believe that Sia will continue to have limited adoption due to the Siacoin aspect, and that while the price shouldn't matter within the Sia infrastructure, it matters to others outside of the project who will instantly dismiss Sia as being a useless project because the price is so low. We also believe that Siafunds are a failed investment and funding vehicle. The storage marketplace aspect of Sia has shown that prices cannot be expected to hold to any threshold, either arbitrary or realistic, and that hosts will charge what they want and what they can get for their storage. We think that today, based on pricing, renters are better suited to use other services which will give them more value, reliability and features for the price they are paying.


Viability Issues with the Sia/Skynet Team Updated

Nebulous Labs appears to be a house of chaos with no real plans beyond surviving each day.

This is a difficult area to criticize, as we do not intend to speak poorly of any particular member of the Sia team or even the team in general. However, there are definitely criticisms which can be made of how Nebulous handles many things, from development operations to community engagement. Some of these are discussed below.

Lack of Structure in Development and Documentation Processes

The pace and strategy of development on Sia and Skynet are both slow and chaotic. There is no project management, no master plan, and not even an actively maintained roadmap at this point. There are no regular release cycles, or even any prediction most of the time of when the next release will be or what it will contain. It seems that after six years, the team is still scraping by on the most basic startup mentality of working on whatever the project head feels is a priority for the day or the week, getting it to bare minimum functionality, and then pushing it out and working on something else while leaving everything that came before half finished. At this point in the game, a greater form of organization is necessary both in order to make more rapid progress and to be taken seriously by other professionals and organizations interested in the Sia and Skynet projects. It seems that even while the Sia/Nebulous team grows over the years, nobody ever manages to introduce any more control or sanity into the development processes of the project.

Along the above lines, documentation for Sia, Skynet, APIs and SDKs is often left in an incomplete state. Changes will be made to a product and not properly documented, creating frustration among the few developers working to build skapps and other items on Sia and Skynet. The cause is pretty obvious - lack of any formal procedures for development or releases. How can you update documentation when nobody is keeping track of what's changed and what's made it to the master branch? If everyone is so busy and engulfed in chaos that documentation can't be updated, imagine what else is being overlooked each release that may have the potential to wipe out your data, expose it to others, or who knows what else. We believe that the lack of standard planning, procedures, and cycles at this point has no excuse aside from anti-authoritarianism on the behalf of project leadership.

Breaking Changes to Production Servers

Changes to production servers continue to be made on short notice, and in between releases. The predominant Skynet portal which is run by Nebulous, SiaSky.net, is treated as a development server and does not operate with the same software that is currently in a stable release. Why this is done instead of using their development portal, SiaSky.dev, is unknown. Many times, changes will be breaking and users or developers are only notified when they wander into the Sia Discord to ask why something isn't working. There is no commitment from the team to stabilize the main portal to only run release versions - they seem to think that the few hundred portal users absolutely must benefit from the latest and greatest all the time, without any heads up of what is changing or when, and we believe that this is a careless development practice as would most professional developers and serious institutions.

Half-Finished Items and Bugs Never Addressed

As previously mentioned, the team is unresponsive and uncommittal to fixing core Sia issues which have existed for years. These calls come from the most dedicated and regular community members and contributors, but the team continues to build out Skynet, likely because it is the "latest and greatest" and the new potential revenue builder. Many of the things community members request to be fixed are dismissed as unimportant, including things like accurate accounting and bugs with the potential to corrupt a host's data and destroy all pending income like the infamous "Bolt bug". If things are going to be released in a minimum viable state, there needs to be a plan to come back and tidy them up shortly afterwards. To do otherwise is to always operate in a half-broken and half-finished state, which is careless when dealing with a product handling people's important and sensitive data.

With the shift of the core Sia product to the Sia Foundation, we also fear that many of these issues will continue to fall to the wayside, and that the Foundation developers will prioritize things they personally believe to be important while Nebulous/Skynet Labs and the bulk of the developers will focus on Skynet only and will not be concerned about Sia functionality beyond what is required for Skynet to function properly.

Leadership's Instability, Stubbornness and Resistance to Structure Updated

Again, we don't mean to tear down anyone in particular, but unfortunately this point is relevant to the direction and success of the project. The lead Sia developer and co-founder is very opinionated on his vision for the project, and seems to maintain that vision against many facets of reality. He spends a large portion of many days on the Sia Discord defending development or design decisions, even if the community insists that they are not the best way to go. He also appears to speak for what the Sia Foundation (explained later) will and will not be doing, even though he will not be on the Foundation board or officially involved. In general, we've gathered that he wants to maintain control of as many aspects of Sia and Skynet as possible, and is generally unwilling to delegate tasks or consider input requiring significant technical decisions, or submit to others such as in the context of project management. We believe that between the need to control the project and the lack of leadership around development and product structure, this is the biggest reason Sia and Skynet have many of the challenges that they do. It has been an issue for years, and it appears that it will continue to be an issue so long as the current organizational structure exists.

The instability of project management has been displayed to us many times personally over the years through borderline unhinged rants about frustrations with the project, other community members, other projects, or even directed towards us when we criticize the project. The ultimate and most telling demonstration of this culminated with our being banned from the Sia and Skynet social channels like Discord and Reddit in January 2021. More details on how the ban came about are provided in a section below, but the bottom line is that while we worked to make changes and come to a compromise over any outstanding issues after posting this piece, instead Sia leadership continued to rant about our prior slights and how we should have been banned long ago if there was any chance we would have criticized the project someday. Our fate is similar to others before us, such as a user who publicly criticized the project on his blog and drew leadership's wrath many times over the fact that he was bringing negative attention to Sia, even if what he was reporting was accurate per his experience with and testing of the product. While he eventually left the project on his own, ironically before our ban Sia leadership once again brought him up and stated that he should have been banned as well. It is clear to us that leadership is unable to tolerate or address criticism, and that due to their instability any dissent is ultimately likely to be chased out of the project, or failing that, banned instead.

Sporadic Community Engagement

Similar to the development cycle, product updates are sporadic and may be made once a week, every few months, or simply in real-time by a comment made in Discord, depending upon the type of update being provided. Things which were supposed to be regular updates, such as a monthly community update on the Sia blog, have varied from being released every month to updates being six months apart. Often, information given in Discord by the team in response to a question will be counter to what has been announced previously or what the currently understood policy is. Sometimes important announcements are made solely in the Sia Discord under the announcements or contributors channel, and nowhere else. Development updates, sometimes posted weekly and sometimes posted less frequently, are posted by volunteer moderators and not the core team. Along the general lines of inconsistency we see elsewhere from the team, information dissemination is yet another. We suspect that this may be why as of today, about 30% of Sia hosts are still running an old version of Sia, and about 15% aren't running the current v1.5.x series of releases - how would they know that they need to do otherwise if they don't hang out in Discord regularly?

Inventing Problems to Create Solutions New

We stepped away from the project in November 2020 partially due to real-world commitments, but also due to our burnout with the project over all of the reasons listed in this document. For us, this mostly means staying off of the Sia Discord where most of the community activity and discussion occurs, but we still occasionally browse other platforms like Reddit. While we were away, something that really drove home the futility of the project is the showcasing of completely insignificant items as if they are a revolutionary deal. For example, a Reddit post showcasing "decentralized identity profiles" argued that a simple profile page with a picture of you and links to other sites like LinkedIn is such a big deal that you need to make it decentralized so that nobody can take it down - but this proposed solution is not even truly decentralized because it relies on Handshake (HNS and hns.to domain) as it was presented, and the profiles were served insecurely on top of that. Besides, in what universe would any major web service find this sort of thing problematic in any sense?

The reason these things continue to be showcased as major achievements (along with the majority of skapps which are in reality simple applications) is that Nebulous has created an artificial problem in saying that everything you own on the Internet needs to be decentralized to be under your control. While in the very strictest sense this may be true, in practice 99.999% of Internet users have no reason to be concerned that their identity online or any other content they post is at risk of being arbitrarily taken down and lost. Examples of people being "deplatformed" from Twitter, Google, Facebook, or other sites are thrown around as a warning, but these actions are limited to one platform and the offender has typically broken some policy repeatedly in order to earn such treatment, such as making false claims or calling for violence. Even if you create your own decentralized identity page through Skynet and HNS, it's useless beyond being a page with a few links to other platforms - you still need accounts on those other platforms which are subject to being removed if you can't play by the rules. Your "decentralized identity profile" doesn't help you keep your presence anywhere else on the Internet, and showcasing it as a big deal is embarassing to the project because it really demonstrates how a problem has been invented in order to try to justify Sia and Skynet as a solution, which they are not.

Selecting Inexperienced New Hires with No Connection To the Sia/Skynet Projects New

Nebulous most recently hired for two positions in October 2020: a marketing manager, and a "developer evangelist" who would assist in onboarding new developers to the Skynet infrastructure and help get them started with writing code. Both of these hires were made externally to the project, and had little or no experience with Sia or Skynet upon being selected. We've waited a few months to give them the benefit of the doubt and get settled in, as while lack of experience with the project may be somewhat passable for a marketing manager, it is more concerning for a developer. So far, both hires have been disappointing. No doubt we would be told that they need time to spool up and get into the project, but both have demonstrated little understanding of how Sia and Skynet actually work months later, and have made abysmal social appearances such as in a botched AMA or while posting awkward and cringeworthy "12 Days of Skynet" graphics showcasing how many gifts team members were returning after Christmas. Marketing appears to have been focused on trying to create cute content for TikTok, and we're not yet sure what the developer evangelist is doing but the majority of statements made so far have been platitudes about the mission of Skynet and how everyone needs to give skapp developers a chance without hitting on the technicals of how Skynet is going to be better for projects or where it might fill some use cases but not others. Maybe Nebulous knows something we don't, but we doubt Skynet is going to take over the world via cutesy TikTok videos and a developer evangelist who can only sing praises but not articulate technicals, and both of which are still figuring out how Sia and Skynet even work in the first place.

Conclusion

A development team can be one of a project's strengths, or one of its weaknesses. While there are certainly talented and creative developers on the Nebulous team, we believe that the lack of processes and organization squanders their potential. If nothing else, money talks, and the fact that Nebulous has come six years with such a slow rate of progress and no substantial revenue says a lot about the ability of the team to convert the vision into the income under the current organizational structure. While we hope that recent new hires in marketing and developer support and outreach are helpful, we don't see their roles effecting any new structure in the aforementioned areas. We're concerned that there are not many more VC funding rounds left before the funding and/or the equity are dried up, and that the projects will not see profitability before that time without some better organization and focus around how the team moves forward with their work. We also aren't hopeful that the neglect of longstanding issues with the Sia product or inconsistent community engagement are likely to change.


Viability Issues with the Sia Foundation

The Foundation is a new entity with what appears to be an inconsistent plan.

The Sia Foundation was brought forth as an idea to move the core Sia product development to a new nonprofit entity which would be funded by a block subsidy from blocks mined on the Sia blockchain. This would free up Nebulous Labs to work on Skynet, provide dedicated development towards the core Sia protocol and product, and provide funding for a number of community and other initiatives. Of course, this plan would also require community buy-in, as they would need to agree to a Sia fork which implemented the block reward change and effectively diluted the value of Siacoins even moreso with additional Siacoins being minted every block. While the Foundation saw initial community support, the plan has become fuzzier as time has passed after the initial announcement, and many details on which the Foundation was pitched seem to have changed.

Manufactured Community Support

Prior to the public Sia Foundation proposal release, the Sia team contacted several "prominent" community members in private and in advance to garner support for the proposal and gather feedback. This feedback could have been gathered all at once and more publicly, such as in the contributors channel, as none of the information in the proposal was particularly sensitive. However, the way it was approached privately helped the team to ensure that the majority of the main community members would be onboard with the proposal before release. When the proposal was posted publicly on Reddit, community members were encouraged to go offer support immediately, to the point where the proposal looked like it was being brigaded and any dissenting comments were drowned out. Additionally, the next day, the team also announced their switch to focus on Skynet and their intention to rebrand as Skynet Labs, which was not previously discussed with any community members. While it may be expected that the shifting of the core Sia product to the Sia Foundation would logically result in Nebulous/Skynet Labs focusing on Skynet, the way that part of the plan was disclosed which was likely to gain community support but another part was withheld was not very forthcoming. Having the "prominent" community members also brigade support for the Foundation also pushed down the voice of the rest of the community - there was no point in objecting if all of the major players were in support. We assume that the team was attempting to avoid the community division which came about during the Obelisk fiasco and the fork discussions that occurred at that time, but after seeing how the Foundation announcement played out, we were disappointed that it appeared community support was selectively manufactured in advance.

Funding in Perpetuity

One concern that some community members have raised is the fact that the block reward to fund the Foundation will fund the Foundation regardless of performance, and in perpetuity. There is no guarantee that the Foundation will perform any significant work or do anything that they say they will do, and if Siacoin price rises significantly the Foundation could be collecting massive amounts of money each month with no way to interrupt it aside from forking again. The Foundation proposal stipulates that excess funds which cannot be spent will be burned, but in reality there is no way to guarantee that this will occur. Regarding funding, there are also concerns that doubling the Sia block reward and then selling half of it to fund the Foundation will result in immense downward price pressure on what is already an abysmal Siacoin price, further turning people away from the project. It will also cause further coin supply inflation. Regardless of the economic pressure, a block subsidy that continues forever and prints $4 million USD per year in today's prices could be concerning to those who expect that either Siacoin prices will rise, the Foundation will not do what they say they will do, or both.

The Plan Keeps Changing

We realize that any given situation may be fluid and that changes are required as circumstances come up, but many things have changed from the initial Foundation proposal up to today. Some of these are fairly minute and routine details, like how exactly the Foundation subsidy will need to be classified so as to satisfy nonprofit requirements by deeming it a "donation" of sorts and are not a big deal. However, even though not a big deal, we feel that these sorts of details should have been fleshed out in advance, as Foundation implementation has now been delayed from the originaly proposed timeline, and of course in classic Nebulous style there is no firm word on the new plan. Other points are more significant, such as what exactly the community subsidies are likely to cover - for example, initlally subsidies were said to cover Skynet portals, but now that plan has apparently been reconsidered so that community portal operators can no longer rely on a portal subsidy to help cover initial portal costs, and must assume all the costs and risks of starting up their own portal. Other proposals have suggested that any service, such as SiaSetup, must be completely open-sourced before receiving any Foundation funding, a move which other community service members have objected to overall as many projects have sensitive or proprietary components designed to prevent malicious actors from gaming third-party ranking systems or related to other background site functionality.

Funneling Funds to Nebulous

There has also been discussion since the Foundation announcement on selling/transferring the Sia IP (such as the Sia name and logo) to the Foundation in exchange for something like $1 million USD, which we feel is a bit of a money grab as we don't find it realistic that the IP on a product currently generating $0.02/month for each Siafund is worth anywhere near that price. Additionally, it has been said that it is expected the Foundation would hire Nebulous/Skynet Labs for at least the first year to "consult" on Sia and help spin up new Foundation developers, but we are concerned that this arrangement could continue indefinitely and simply serve as a money siphon from the Foundation to Nebulous/Skynet Labs instead of supporting a new and separate entity. This ties into the topic below about potentially insufficient separation between the Sia Foundation and Nebulous.

Aside from what may be more obvious transfers of funds from the Foundation to Nebulous, we also suspect that the timing of the Foundation coincides with the recent venture capital fundraising round held by Nebulous for a reason. It seems likely that investors did not see Sia as a viable product to generate profit under the Siafund model as described in the relevant section above, and wanted a strategy to start generating income while cutting some of the dead weight - that dead weight being the development of Sia, which is a free and open source product which anyone can fork and create a copy of at no cost. Creating a Foundation which was authorized to print $4 million USD (or more) a year out of thin air to cover Sia development and allow the profit-generating arm of Nebulous/Skynet Labs to focus on the Skynet product with a monetization strategy (albeit, a poor one) all seems to be one single plan in hindsight in order to cut Sia development costs and direct VC-funded development to only the profit-seeking venture. Of course, this is only our suspicion, and we doubt that it would ever be admitted if it were the least bit accurate.

Insufficient Separation Between the Foundation and Nebulous

One part of the Foundation proposal described that funds would be held in a multisig address, meaning both Luke and Eddie (the starting Foundation board members) would both need to approve a transaction for any Foundation spending. However, the proposal also stipulated that David Vorick, the lead developer for Skynet and co-founder of Sia which will remain with Nebulous to head Skynet Labs, would hold a dead man's switch for the Foundation funds in case there was a deadlock or another issue with spending the funds. We feel that this is a conflict of interest, and we saw a similar situation with the Obelisk project when David sat on the board of both companies and directed Nebulous employees to work on Obelisk after delays despite insisting that both companies were separate. Additionally, David has been the predominant voice in saying what the Foundation is or isn't likely to do ever since the initial announcement, while Luke and Eddie have remained largely silent. We have concerns that the Foundation will not be sufficiently separated from Nebulous/Skynet Labs as a result, and that David is likely to be pulling Foundation strings in conflict with/in support of Skynet interests even though the Foundation should be a completely separate entity by design. We would prefer to see that the Foundation funds would remain locked in the event that anything happened to Luke or Eddie, and that a new fork with new community support could sort out any logistics around deadlocked funds or a Foundation board member replacement.

Conclusion

When we were originally approached about the Foundation, we were in support of it based upon the information presented - however, the plan has changed enough after the fact and enough has been seen in hindsight regarding how the Foundation proposal was formed that we are now skeptical the goal of the Foundation is primarily to benefit the Sia community. We find it more likely that it was an existential move in order to reduce Sia development costs and shift those costs to miners, with the carrot thrown in that some community initiatives would also be funded in order to gather support. We also have concerns about the separation between the Foundation and Nebulous/Skynet Labs, and we expect that it will ultimately be a conflict of interest where the Foundation is directed to support Skynet in profit-generating endeavors even though the Foundation is a separate nonprofit entity. It is liable to be another mess similar to Obelisk, a mess which the community will come to regret that the project managed to get into.


Viability Issues with the Sia Community Updated

The community, like many crypto communities, stubbornly sees the project as something revolutionary regardless of the obvious issues.

If you've stuck with us through all of the above, congratulations! We know it's a lot of words, a lot of negativity, and a lot to process. However, the last piece in the puzzle that we ultimately believe make Sia and Skynet unlikely to succeed is the Sia community. We hate to pick on our own, but collectively we could probably gather enough blinders from the community to supply the Kentucky Derby. Yeah, we got jokes! But really, though - they're a stubborn bunch which in many cases have dug their heels in to support the project no matter what direction it takes. That being said, on to the issues:

Unrealistic Expectations About Siacoin Being an Investment

If you were to say on the Sia Discord on any given day that Siacoin is not intended to be an investment, but instead that it is a utility token, you'd probably get a number of negative reactions. But to deny this goes against reality - a utility token is a token which is exchanged for goods or services, like a car wash token or a carousel token. If you have a bunch of tokens (i.e. Siacoins) and you want to buy or sell storage, it is typically beneficial to you for the price to remain fairly stable throughout the duration of your storage contract. While there is a chance that the price could swing in your favor, there is also a chance that it could swing the other way, and you would end up paying more for your storage or losing money on the storage you're renting out (this is explained more in the section above about Siacoin and storage pricing).

Buying Siacoins as an investment is kind of like buying car wash tokens hoping that they're going to go up in price and that you can sell them to another guy for $0.25 more than you bought them for - it's a poor investment strategy, and there are no forces compelling it to happen, especially when there is such a large supply of car wash tokens or Siacoins out there. Yet, some community members have purchased as many Siacoins as they can manage in hopes that they will spike in value due to network utilization or some other motivating factor. While there is always the chance that a random price spike like those seen in 2017 could happen again someday, this would be an unlikely and random event precipitated by nothing in particular, and so holding Siacoins in hopes of such a thing happening is random speculation just as much as buying any other altcoin could be - maybe even moreso, considering the purpose of a utility token and the benefit of the coin's price remaining stable. The fact that so many community members deny this fact showcases either their ignorance or their...

Rose-Colored Glasses and Idolation

As we've said, people in many crypto projects seem to truly believe that the particular project is going to take off, be revolutionary, and change the world. They're essentially in love with the project, and see it through rose-colored glasses. Unfortunately, this incites them to defend every decision of the development team, even if the decisions make no sense. There is very little pushback in general from the community on points of the project, plans, features, or anything else which just doesn't make any sense or which is impractical (and we've provided plenty of examples above). Those that do bring up valid concerns and point out that the plan may not be ideal are met with what we can best refer to as "Sia white knighting" from a number of community members and moderators. As you talk to them, you realize that many of these people actually believe what they're saying - they really believe that Sia and Skynet are somehow going to rise up and take over the Internet, despite the realities of the products, their obvious limitations, and the fact that the general population could care less about most of what Sia and Skynet are trying to sell.

In many ways, it's the same kind of delusion that you see in other group followings in that the organization and its leaders can do no wrong, and everything will work out if everyone just continues to believe in it all. Similarly, they praise the project leaders for being visionaries, and the project leaders joke (at least, we hope) about being a god. Where facts and logic cannot prevail, we are left with a cult of sorts, and they will be happy to ride out the project until the very last day when it is undeniable that it is over, assuming that it ultimately happens to steer towards failure. More likely, they will ride it into mediocrity until the money runs out, or until they realize it's been 10, 15 or 20 years and the company still hasn't seen profitability and adoption. Ideally, if nothing else changes, they'll get bored and find better things to do with their time. If chance so happens that the project becomes a huge success, we will be happy to admit that we were wrong about these community member visionary cultists also.

All Aboard The Ban Wagon - Next Stop, Tin Foil Aisle New

Similar to Sia project leadership, certain community members are also unable to tolerate criticism of the project, including some moderators and others who have large interests and investments in the Sia project. Our posting of this concerns page prompted all manner of conspiracy talk as to our motives for writing this piece, posting it originally, and then posting it again later. There were suggestions that we were being paid by other projects or otherwise involved with them, and working to try to take out Sia because it was competition to those projects. It could not be conceived by some that we simply feel strongly about these criticisms and think that others should know about the issues with the project before investing any amount of time or money into it. Explanations were demanded, and when those were unsatisfactory, we were painted as a traitor who should be banned for spreading negativity despite our continued efforts to answer questions for newcomers and keep SiaSetup online in its entirety to continue to be a useful resource on Sia and Skynet.

Ultimately, this talk shifted into calls for a ban of RBZL and discussion of SiaSetup, and a small handful of moderators and community members cheered this course of action on even as the majority of the community and moderation team insisted that a ban for an opinion piece on a separate website would be a very poor decision. However, the ones pushing for a ban were the ones invested heavily in the project, and they likely believed it was of greater benefit to simply silence the criticism in order to try to protect those investments instead of by addressing the real problems within the project. These are the people who will never say a negative word about Sia or Skynet, no matter what the project or its leadership decides to do. They are liabilities to the project because if things continue to deteriorate, by the time they finally decide to speak up it will be too late. They cannot fathom that a criticism of the project comes from good intentions, and that it is written thoroughly and thoughtfully in order to try to contribute to improving the project. Unfortunately for those members, the only thing a ban has done is remove the criticism from their direct view. The criticism will remain online, and the ramifications of banning a long-time community member will also likely result in negative PR and questions as to the tolerance of Skynet and how censorship resistant it actually is.

The Community Isn't Actually That Big - Or That Devoted

Six years into Sia, and any given post or announcement can't generate more than about a hundred upvotes, likes, reactions, or whatever the equivalent is on a platform. A portion of this lack of engagement could be attributed to a lack of marketing, but we feel that plenty of people who are into cryptocurrency have come across Sia, and have then dismissed it for various reasons. Maybe they don't take a project with a low coin price seriously, or maybe they saw the economics and figured they wouldn't make any money with Siacoins (or they actually understood what a utility token is!), or maybe they actually tried the product and thought it was kinda cool but that's about it - or worst case, they had issues they couldn't resolve and gave up on it as being a garbage product. We believe that the people are there, and that plenty of people have seen Sia (and now Skynet), but in general nobody really cares or finds it revolutionary.

There are only a few decentralized blockchain storage projects out there, compared to the thousands of other altcoin projects which do something or other or odd things or nothing, and we have to wonder why there are so few which deal in storage. Our best guess is that people think decentralized storage as a concept is neat, but in reality they don't see a need for it in practical use. We also think that most people in the crypto space are not actually that interested in any crypto project unless it has a very high potential to make them lots of money. The rest that hang around in the Sia community are either in the smaller camp of having idolized the project, or the larger camp of simply hanging out to see what it can do or trying to use it for something in a hobby capacity. And in reality, the campsite isn't all that big, as is evident by the overall lack of community reaction to any given update or news through official channels. It also doesn't help when events like an AMA posted on Reddit mid-December 2020 end up being a circus which attracted a whole five top-level questions, and where the attending Nebulous staff couldn't even coordinate replies or format them properly in response to a list of questions.

Hackathons and Small Projects, But Not Much More

The majority of the projects and integrations which have come about for both Sia and Skynet have been the result of hackathons, which are short-term contests to try to create a product which performs a certain function or uses a certain technology or integration. These hackathons can carry cash prizes (or SC with USD equivalent) of several thousand dollars, which can make it worthwhile to spend a week or two of your free time throwing something together for a few extra bucks. However, these projects are rarely much more than basic proofs of concept, and even more rarely are they built out and maintained after the hackathon. This creates a bunch of small applications for Sia and Skynet which are never really used because they become outdated as changes are made to Sia or Skynet, or because they're never promoted or offered as an actual product with the intention to operate long-term. We believe that hackathons generate articifial community interest in Sia and Skynet only for the prize money, and not as a sustainable way to attract new developers to Skynet or to create products and skapps which will actually be updated and maintained for more than a few months. With the release of Skynet, we've also seen a decrease in community contributions to the Sia codebase, with developers instead focusing on small skapps and hackathons.

In terms of actual sustained Sia community development activity, most of this area is fulfilled by community sites such as SiaSetup, SiaStats, and SiaCentral, which are separate third-party services not running on Sia or Skynet but instead providing services or information related to the Sia network. These projects are mostly maintained as hobbies and community services by the developers out of generosity. A small number of other actual companies have built services upon Sia, such as Storewise for a Sia storage integration with a privacy-focused cell phone and Filebase which provides storage and backup services using Sia for long-term storage. Pixeldrain, a free file upload service, also uses Sia for long-term storage. Even then, these services typically use a third-party Sia client/interface developed by Luke called us and not the standard Sia client/daemon, as the standard Sia client typically doesn't meet their needs due to functionality restrictions. It remains to be seen if these services will manage to be profitable long-term, or if they have other business ventures which are supporting the costs of their Sia endeavors. And with these few short sentences, we can summarize the community development activity surrounding Sia and Skynet over six years, which is very little that is substantial and nothing that is likely to represent significant growth of either Sia or Skynet.

Conclusion

Don't get us wrong - we think that there are several community members which provide valuable services, support, and discussion to the Sia community, and we appreciate their work as much as you hopefully appreciate ours on SiaSetup. However, there are also community members which will go down with the ship if that's what it comes to, even in the face of logical discussions surrounding obvious flaws in the projects. We can only assume that their interests and investments in Sia align with that stance - otherwise, willful ignorance is unflattering and we find it to be one of the biggest frustrations within the community. We also don't see many community members truly sticking around to build out a full-fledged finished and polished product, and we can only assume it is because nobody really believes that it is worth the effort beyond a hackathon where an immediate payoff is the result - or they don't believe that there will be long-term monetization in it for them, or that the project will see enough adoption to merit the efforts. In any event, the Sia community in general either appears to remain hopeful at the project's prospects in spite of all of the challenges we've covered above, or they're just along to lurk or comment or play with it all as it develops but are fairly noncommittal beyond that. The latter camp certainly appears to be the bigger one.


Our Overall Conclusions Updated

As you can see, we feel that there are several challenges to the viability of Sia and Skynet.

We don't really feel that we need to repeat ourselves - everything is laid out for you above. There are obvious viability challenges for both Sia and Skynet, and after our experience with both products over several years we'll be surprised if either sees adoption beyond niche uses or a few thousand actual regular (mostly non-paying) users unless some things change. We expect that most use will end up being for illicit purposes, such as distributing copyrighted and other illegal content. We also expect that any halfway-viable Skynet apps will eventually be filled with garbage - spam, illegal content, and posts from users rejected from other platforms for not being civilized, such as those which sling hate speech or conspiracies. Funnily enough, the "deplatformed" have often done something to deserve it.

The monetization strategy is one of the biggest things which we doubt will be successful, and monetizing the products is obviously key to being able to continue development. We don't want to sit and criticize only - we've considered for some time how to better monetize Sia and Skynet, but with Skynet in particular we just can't come up with anything which we think people would be willing to pay for (developers, users, or otherwise) because Skynet is so limited in what can be done with it. Sia could be monetized with more sane and user-friendly access to the ecosystem in a Dropbox style polished app with fiat payments (similar to SiaStream), but this is something the core team has resisted for years while focusing on the protocol only. However, even with monetization brought into the realm of sanity and practicality, there are so many other challenges around Skynet architecture, liability, and adoption that we think it's a vertical climb to get there.

In any event, with all of the above we feel that we've justified our ceasing of updates on SiaSetup. Following Sia has been like chasing the carrot on the stick - we've realized that we've been walking for years and we really haven't gotten any closer to the carrot, except when it occasionally swings a little closer now and then. We fully expect that the majority of what we've written will be dismissed as not being major issues, not "seeing the potential of Skynet" or not "understanding what it is capable of", and all of the usual defenses which are thrown about whenever these issues are brought up. Those who wax poetic in Sia and Skynet's defense typically have no substance to their arguments, only feel-good platitudes about how it will change the world and anyone who doesn't agree is just too short-sighted to understand. We also expected that as a result of this page, SiaSetup might have been de-facto banned from mention around the Sia communities, but Sia leadership did us one better - they banned us entirely instead. However, this is fine - we have no need to see SiaSetup propagated anywhere, as it is a volunteer project which does not generate revenue or serve any purpose other than to help the community out. Such an action hurts you, the Sia and Skynet user, not us.

This showed up in November 2020, went away, and now it's back in December 2020. Why? New

Obviously because we're bipolar. Ha, no, not really. This page was originally posted on November 27, 2020 in order to express concerns with the project, but it ended up stressing some people out who are involved with the Sia and Skynet projects and who we consider to be friends. We don't think we're that big of a deal here at SiaSetup, and if you make all of your decisions based around a single website or source in anything you do, you're probably going to have a bad time, but people seemed to think we're influential within the project and that this document might end up causing some damage. After a week or so, pretty much everyone had seen these concerns that needed to see them, so we removed the page because we didn't want to be responsible for causing everything to go under. Our real-world obligations also kicked in, and we stepped away for most of December.

While it can be hard to step away from something that you've been involved in for thousands of hours over several years, it can be good in that it gives you a new perspective when you start looking at the project without all of the background knowledge and history fresh in your mind, and you can then see it for what it is today and what a new person coming in would be likely to see it for. And honestly, we see it as a joke. Some may say that's a harsh assessment, but we don't make it lightly. The direction that the project took from focusing on Sia last year to focusing on Skynet now is such an uncharacteristic move compared to the original Sia ideals and vision that we can't help but be disappointed. Skynet takes a lot of what made Sia a legitimately innovative project, and throws a bunch of awkward centralization, monetization, and chest-thumping into it because it appears that Sia was never going to make money at the current trajectory. After stepping away, Skynet now appears to be a hurried last ditch to make the Sia infrastructure more utilized, to generate revenue, and to rebrand as something fresh and new despite all of the obvious issues and limitations, and hope that enough people will find it interesting enough to pay for and start turning a profit. At this point, we see Skynet as more of a death rattle.

More critically, and to the point of our piece here: if one single criticism piece is enough to take down a multi-million dollar project, that project isn't on strong footing in the first place, and there were more serious issues that existed before that piece came along to point all of those issues out. If Sia and Skynet are to succeed, they will do so on their own merit, and not because of what we do or don't say about them here. If Sia and Skynet fail, this piece will be the most insignificant contribution to that ending. After our brief time away from the project, we believe that others should know what they're getting into before sinking any time, effort, or money into a project on such shaky footing, and so this piece has been restored as of December 26, 2020. And apparently the team sees this as a threat to the project, considering that we've now been banned as a result on January 1, 2021. Happy New Year!

What's Next for SiaSetup? Updated

SiaSetup will remain online indefinitely, and automated functions like the consensus download and calculators with real-time pricing should continue to function, but other content will remain as-is. SiaSetup hosting services will no longer be available, and our host will be taken offline once contracts spin down. You can still also contact us via the Contact SiaSetup page via the menu in the bottom right of each page to tell us we suck or whatever. We could previously be contacted via the Sia Discord via @RBZL, but now that we're banned there you're limited to contacting us here or through Reddit via private message at /u/RBZL. If you're still in the Sia Discord, feel free to drop an F in the chat for us.

We suspect that the Sia team's hope behind banning us is that we'd become frustrated or bored with the project, and take this page or the entire SiaSetup website down as a result. Unfortunately for them, SiaSetup will remain online and this page will be displayed for as long as we feel the Sia and Skynet project's direction is one which will not be successful.

Will SiaSetup Resume Updates or Remove This Notice?

Our plan in regards to this matter is similar to Nebulous' general plan in the sense that we don't really have one. We will continue to observe the Sia and Skynet projects, and if we think enough progress is being made to address several of the issues and concerns we've listed above, we may return and remove this page. We expect that it would be a while before that occurred, though, as we'd like to see:

We're not saying that all of these things would need to be implemented or fixed before we'll return, but if things start heading in that direction we're likely to hop back onboard. Rome was not built in a day, after all. We're happy to pitch in and contribute to the Sia ecosystem once we see that it is going to be a project worth contributing to. Until then, please enjoy the content on SiaSetup if it is able to help you out at all, and previously we would have said that we'll see you around now and then in the Sia Discord but having been banned that's probably unlikely.

And of course, if you've decided to throw all your Siacoins away in frustration after reading all of this, feel free to throw them at us.

Much Love,

RBZL


Banned From Social Channels New

Happy New Year 2021! On January 1st, RBZL was banned from the Sia Discord and the Siacoin subreddit - and effectively, the Sia project.

As explained above under "Our Overall Conclusions", this page of concerns was originally posted on November 27, 2020, later removed, and then brought back on December 26, 2020. Accompanying the page was a dismissable red banner at the top of SiaSetup similar to the one there now which explained that SiaSetup was no longer being maintained, and which linked to this page in order to inform users that content on SiaSetup may be out of date and to provide an explanation as to why. For some reason, the red banner in particular really irritated a few community members, including the Sia project's management, as they thought the banner guaranteed that users new to Sia which came across SiaSetup would see what some considered to be nothing more than a bunch of negativity about the Sia and Skynet projects. Strangely, they seemed to fixate on the mere existence of the banner and criticism, not the actual content of the criticism or exactly why any given point contained within was invalid.

Backlash among a small handful of users grew on Discord over the next week, where they theorized about the motives behind posting such a criticism and called for RBZL to be banned for spreading "false information". It was alleged that our sole purpose was now to spread negativity, and even though we've used this site for years to link users to answers for their questions, now doing so meant that it was all a thinly veiled plot to take down the Sia project. Even David Vorick, the Sia lead developer and co-founder known under the username "taek", stated that if he knew SiaSetup would ever post a prominent criticism of the project, he would have banned SiaSetup on day one. Overall, however, most community members seemed to respect our right to criticize the project using our own website, and while some didn't agree with the tone of the criticism or thought that some points were weak or exaggerated, the consensus was that the existence of a criticism on a third-party website was not grounds for a ban from the Sia social ecosystem.

David reached out to us about removing the red banner from our site, as he considered it unnecessary and feared it would steer those new to Sia in the wrong direction. We complied as a matter of good faith early on January 1, 2021, and planned further changes to this page such as the ability for users to submit responses to any or all points we've made in order to create a more fair and balanced view of the Sia and Skynet projects beyond what we've written. We also emphasized both publicly and privately that we were always open to discussing changes to SiaSetup or this page, and suggested that David or others reach out if there was anything else we might be able to work out. However, David continued to publicly lament prior issues and statements, such as the existence of the red banner (even after it had been removed) and other irrelevant points which had nothing to do with our concerns or how we've presented them. Ultimately, we suggested that arguing publicly about the matter would do no good, and that it would be best if David moved on or simply banned discussion of SiaSetup altogether if he was so frustrated with it. We were careful to make the point that we were not suggesting or asking for a ban of our user account, only discussion of SiaSetup - but nevertheless, we found ourselves banned from the Sia Discord on the afternoon of January 1, 2021, about 10 seconds after posting the message suggesting that we all move on. We were informed by other users that recent messages we had posted had been deleted, with David quipping "ok, I moved on, bye" after the ban. A few minutes later, we also received a notification of a permanent ban from the Siacoin subreddit.

The Sia project has previously touted that it is one of the few crypto communities which is tolerant of criticism of the project, allowing more than what other projects would in terms of any dissent from being positive and supporting the project publicly. We would have agreed up until this point. In the face of our offer to discuss and resolve ongoing issues and disagreements, suggesting that we all find a way to move on, and making concessions which were requested, instead we ended up banned - for the existence of this page and these criticisms on our own website. We would have understood if we were hanging out talking negatively about Sia and Skynet all day and posting links directly to this page endlessly in an attempt to deter people from utilizing Sia, but we did no such thing. Is a ban from the social communities really the proper measured response from project leadership in this circumstance? We certainly don't think so, and we frankly believe that David may have some serious issues in order to not be able to tolerate the existence of a thorough and honest criticism against the project - never mind to ban someone who has contributed thousands of hours to this site, to helping new Sia and Skynet users, and who was making an active effort to reconcile any differences and disagreements.

In all honesty, we're also not entirely shocked about the outcome of posting our concerns and receiving a ban in the end; we knew the risks of dissent. We've had concerns about the project's leadership for some time, and those were a significant driving factor in the construction of this writeup. We've discounted the Sia and Skynet projects for several technical and logistical reasons, and being banned for explaining those reasons cements our justification in discounting the projects for their leadership as well. Are you really going to trust someone who preaches about protecting the deplatformed as his reason for building Skynet when he's banned a long-time community member for what are valid (if not sometimes sarcastic or blunt) criticisms of the project posted on an independent website? Are you only protected against deplatforming on Skynet if your views happen to line up with David's, or a skapp developer's? If you see no problems here, we wish you the best of luck in your endeavors with Sia and Skynet. However, hopefully you see this behavior as the major red flag that it is, and a major concern that Skynet is potentially no different from any other centralized service which will suppress what you say or do on their platform if they don't like it. Moreover, now they've shown that they're actually willing to suppress something if it crosses whatever arbitrary line they decide to draw on any given topic. To do so is their right, but the implications of this action when it is performed by a project and organization focused on privacy, decentralization, and censorship resistance are much more significant.

You can no longer trust that you will own your data on Skynet if it transits a portal run by Nebulous, or that your data won't be removed if you say or do something outside of Skynet which they do not agree with. These are the real consequences of our banning, and they should concern everyone. For this reason, we implore you to reconsider your involvement with the Sia and Skynet projects, and to instead use one of the many alternatives available which will provide you with secure storage for far lower prices, less complexity, and higher reliability. If you absolutely must have decentralization or a project tied into cryptocurrency, there are other options available which will fit those criteria as well. The sum of all drawbacks of Sia and Skynet at this point are far too great to ignore.