A Brilliant Response to all the SIA coin Hype - ~A sleeping Giant will soon be woken.
I am a tech entrepreneur, founder of 2 B2B software businesses, one which exited for $130M and another which is growing & profitable, and my area of expertise is product strategy.I have been playing with Sia software for 2 weeks, trying to use it both for file backup as well as for file hosting. I have also read through tons of documentation / forums and blog posts.Right now it seems to me that this project is quite doomed. A fair amount of commentary in this subreddit seems ill-informed, widely enthusiastic, or downright illusory, and this does the project no good.Let me outline the main reasons for my skepticism:
- Software Usability - the software is simply not useable. It doesn't matter what you're trying to accomplish, it's all deeply broken. Not only is the software poorly designed and ugly, it has deeply flawed premises that stem from the overall system design choices. It’s hard to create a wallet. Hard to transfer Sia into the wallet. Hard to announce a host. Hard to understand the parameters. Slow. Most amazingly, the #1 operation which the Sia client should make easy, which is to start backing up files, is the hardest to accomplish. The process of creating an allowance is downright incomprehensible. Uploading any non-trivial amount of data is mega-frustrating with very little feedback about what is going on. It’s a minefield with precious caveats like: you will need to use 3x the bandwidth of what you’re uploading or a 5kb file will take 40Mb. I’m not going to even mention what happens when your allowance expires.
- Developers are all over the map - From what I understand, there are 2.5 people writing code for this project. They are supposed to maintain all the layers of a cryptocurrency, design protocols, a programmatic and a UI client, design an ASIC miner, work with the custom versions of the client that the exchanges use, as well as maintain one or more websites, write documentation, solve bugs… for a project of this scale, this is a laughable amount of resources.
- Poor decision making - If the vision is to create the premier implementation of distributed cloud storage, what sense does it make to maintain a cryptocurrency to begin with? If I want to create app, the last thing I need is to create an operating system for that app first. Given the almost too obvious alternative of using Ethereum ERC20 tokens, I frankly don’t understand the motivation. This is further compounded by the decision to then go ahead and focus on making a miner. While all this goes on, the critical part of the solution, which is a solid reference implementation of a great scaling protocol, suffers.
- Entrenched competition - other areas of the distributed app market offer greenfield opportunities, but distributed storage does not. There are better funded, larger, more focused teams delivering better software right now. They have it hard, disrupting the established non-distributed players will not be easy, so the chances for lagging projects like Sia are even slimmer.
- No solution for the 2-sided marketplace problem - What Sia is trying to fundamentally accomplish is to create a marketplace for distributed storage. In order for to so, it needs to solve one of the toughest business problems out there, which is to create demand without supply, and to convince the supply that there will be demand. See the Ladies Night Strategy If you look at these charts you will notice that the network is at 1% utilization, and that the prices for storage have plummeted. This essentially means that there is almost no demand for storage, and the supply that exists is going to get peanuts for a long long time. The lack of demand means that almost no-one is using Sia for its intended use, to put files up for storage in its distributed network. And given how hard it is to do, it is no wonder. Also, if you dug deeper, you would see that there are many use cases / use patterns for file storage (eg. backup, cloud access, building custom apps that need a storage layer, personal vs smb vs enterprise, file vs versioned storage, escrow…) but that Sia is currently no good fit for any of them.
What can the Sia team do?If the SIA project is to survive, it needs to make radical changes.
- Pick a layer of the stack and relently focus on it - It could be the cryptocurrency itself. Or the miner. Or the storage protocol. Or a consumer-focused client. Or an enterprise API. Or something that I haven’t thought about. But pick one area that you feel you have a great shot at creating great software. Then deliver that software
- Pick a user, its most painful use case and relently focus on solving for it - In the current approach, the user is a tech focused home enthusiast that has tried backing up his home photos and videos to S3 or Dropbox and is unhappy with paying $20/mo for that, and wants to tinker with other tech to see if he can figure out a way to save a bunch. That kind of user needs a Dropbox-like user experience. You could solve for that, but I suspect that there are much more interesting users with much deeper pain points that you actually have a shot of solving. An example: how IPFS is focusing on CDNs and Archival. You say “Enteprise Grade Collaborative Cloud Storage”, but in what significant way have you targeted enterprises?
- Help generate demand for hosters/farmers - Right now your main audience seems to be miners. Not end users, not hosters/farmers, but miners. That is insane, especially considering that you could have avoided this problem altogether with an Ethereum token + ICO. This project in its current form is not going to go anywhere until you have convinced enough people to upload and then fetch their files from your network. So the #1 adoption metric is TB uploaded (right now it’s 34TB, or about 7 hard drives). If you generate demand for hosting, every other problem will be solveable and if you don’t, it doesn’t matter what you solved that the project will collapse. How can you generate demand for storage? You have to convince people who have things to store that your network is better. How is it better? Is it cheaper? Is it faster? Is today’s solution quantifiably better across the dimensions that the users you’re targeting identify as important?
RESPONSE FROM AN ACTUAL SIA DEVELOPER
We appreciate the feedback, but I think you are coming from a position of not really understanding the Sia vision and strategy. I also don't think you understand the history of the project.
Right now it seems to me that this project is quite doomed. A fair amount of commentary in this subreddit seems ill-informed, widely enthusiastic, or downright illusory, and this does the project no good.
We've grown by more than a factor of 10 in the past 3 months. That is going to result in a lot of ill-informed and new users. It's inevitable when you get that kind of growth. We went from a 3 person team running a $4 million cryptocurrency to being a 3 person team running a $400 million cryptocurrency in less than 6 months. A part of our strong commitment to decentralization means that we don't own many coins, and as a result can't even leverage that market cap to hire more developers.I'm really not a fan of all the ICO mania happening right now. When you get investors who are mortgaging their homes to money into a project which has little more than a whitepaper that they haven't even read... it doesn't sit well with me.
They are supposed to maintain all the layers of a cryptocurrency, design protocols, a programmatic and a UI client, design an ASIC miner, work with the custom versions of the client that the exchanges use, as well as maintain one or more websites, write documentation, solve bugs… for a project of this scale, this is a laughable amount of resources.
The cryptocurrency layer is pretty much complete. We finished it in June 2015 and it has changed very little since. The protocols also do not change very often, although many of them do need a few upgrades. We aren't the guys designing the ASIC chip, we're working with a team from California to do that. The exchanges don't use custom versions of the client, they use the same client that everyone else uses. The websites are all largely static (and the one dynamic one - our explorer - hasn't been updated in over a year) and don't require much effort. Documentation and bug fixes should always come from the core team.You are greatly over-stating the amount of spread of our resources. We have one guy who works primarily on the UI, and two who work on the core Sia protocols. The command line tool is something we maintain for ourselves, it's very simple, rather cobbled together, and also not much developer overhead at all.The majority of our dev attention is turned towards improving the storage layer. It's a hard problem, and it takes time. We've also been consistently releasing massive improvements to the Sia core every few months.
Given the almost too obvious alternative of using Ethereum ERC20 tokens, I frankly don’t understand the motivation. This is further compounded by the decision to then go ahead and focus on making a miner. While all this goes on, the critical part of the solution, which is a solid reference implementation of a great scaling protocol, suffers.
Ethereum is a poor choice. Sia also predates Ethereum, but that's another matter. Ethereum has a long history of breaking compatibility and dropping sudden hardforks on the network that obviously favor certain projects. If you aren't one of the chosen ones, you are a second class citizen. Ethereum is going to be hitting massive scalability issues (they already are - blockchain >250 GB at this point), and the decentralization is rotten - the Ethereum foundation controls everything.The core protocol improves by leaps and bounds every 3 months. If you don't believe me, go try using v1.1.1, and then go try using v1.0.3. Each one is night-and-day from the previous.
The lack of demand means that almost no-one is using Sia for its intended use, to put files up for storage in its distributed network. And given how hard it is to do, it is no wonder. Also, if you dug deeper, you would see that there are many use cases / use patterns for file storage (eg. backup, cloud access, building custom apps that need a storage layer, personal vs smb vs enterprise, file vs versioned storage, escrow…) but that Sia is currently no good fit for any of them.
This is where the vast majority of our attention is going. It took time to reach the scales required. When we finally did, we saw a huge liftoff in interest in the project, and had to take a break to put out a bunch of scalability fires. We've finished that work, now we're back to being focused on the core features of the project - the storage layer.
Or an enterprise API.
This is the core focus of the project, and what we've basically angled everything else around. Sia is another 12 months or so from the types of scalability that would let an enterprise drop 10 PB of data onto the network, so in the meantime we're bootstrapping from the enthusiast market.Sia has a clean API with a strong compatibility promise and good documentation: https://github.com/NebulousLabs/Sia/blob/master/doc/API.md
An example: how IPFS is focusing on CDNs and Archival.
This is our focus as well. Disaster recovery first, CDNs second. We can't do disaster recovery at the enterprise scale until the client can handle more data, and that's going to take use time to get running.
Right now your main audience seems to be miners.
This statement bothers me a lot. It's been literally two weeks since we announced our ASICs. In those two weeks, we've spent less than half of our time working on Obelisk related tasks. Look at our github page. We've pushed a ton of code to Sia this week, because our main focus is on improving the Sia technology. We have a release coming later this week, and it solves several significant problems with v1.2.2.
That is insane, especially considering that you could have avoided this problem altogether with an Ethereum token + ICO
As stated before, I strongly disagree that Ethereum would be a good choice for a project with its sights set on long term success. Ethereum is a disaster in many ways, and all of the technical and social and legal debt they've built up is coming for them.
You have to convince people who have things to store that your network is better. How is it better? Is it cheaper? Is it faster? Is today’s solution quantifiably better across the dimensions that the users you’re targeting identify as important?
Sia is the only cloud storage platform today that you can upload files to while maintaining control of the data yourself. Sia's core competency is actual decentralization, and we've achieved it in ways that our competitors are nowhere close to, and we've achieved it with a fraction of the funding. That is the primary way we distinguish, and everything else is just a bonus.
I hope this helps.
I really do appreciate the commentary. It does help me to understand what is going through the heads of our users, but it also makes me wonder how our users image of us fell so far away from what we're actually doing under the hood. It's obvious to me that people are getting introduced to Sia with absolutely no sense of direction, and no ability to see the greater vision of the project, and I'm not sure how to fix that.It wasn't a problem we had until recently. It's really only as of this week and last week that I've started seeing posts that miss the mark on the vision of Sia by such a wide margin.
Yes this is a direct copy and paste of the original article which can be found HERE, but it's answers some very serious points of contention with the system I truly believe the SIA coin is a sleeping giant that will soon be woken by the roar of the markets demand for something different. I personally am sick and tired of Dropbox and Amazon for that matter with their ridiculous premium rates based purely on you purchasing a Brand rather than a functional system.
Very interesting
nice post. thank for share
Just follow and resteem this post if your steem.power is zero i will upvote your one post
upvote this post then comment your post link which you want should I upvote it
If you not following me you should follow me then do it
https://steemit.com/upvote/@crypto-booster/upvoting-started-followers
Thanks for providing the clarity on the overall progress and vision of Sia. For what it's worth, continuous progress updates and content like just provided would go a long way to boost consumer/investor confidence.
Wish you all the luck, thanks for sharing.