To begin with, here’s some context. We have previously discussed a bit about what to anticipate from Frontier in the Ethereum Launch Process blog entry, and Gav has provided us with a detailed summary of the current status of the development efforts. You can also explore some of the ongoing security initiatives, including the bug bounty program. This bug bounty initiative has uncovered my favorite bug thus far: Jonas Nick‘s finding that it was possible to send a negative transaction, shifting value from someone else’s account to yours!
As far as bug bounty discoveries go, this one is an absolute gem. Well done!
Gav has released a new blog post detailing more about what the challenging aspect of the development process entails: crafting an exact specification for Ethereum while synchronizing three different implementations, rather than relying on a definitive implementation that implicitly establishes the protocol specification, with other versions aligning to it in bug-for-bug compatibility. This presents a sturdy and meticulous methodology for assembling a system, and it’s advantageous to read it to grasp how the engineering efforts are progressing.
At this stage, you should have a solid understanding of the development process and the various threads of work that are being interwoven into the release. Now, let’s discuss what specifically needs to be accomplished between now and Frontier.
There are four particular challenges we need to address for Frontier.
- Security
- Network Stability
- Mining
- Exchange
Two of these, Mining and Exchange, rely on collaborations: individuals must implement our software and operate it, sometimes in combination with their own tools, to offer services in the ecosystem. Remember, we are launching an ecosystem, not merely a product: everything contributes to that ecology. Each of these elements in the release process must function correctly, both independently and in conjunction with others, for the ecosystem to undergo a proper stress test.
Furthermore:
- All software needs to be fairly straightforward to install, mining included
- We must assist partners, especially exchanges, in delivering services
- We need to develop our checkpointing and blockchain integrity testing services
- All within the framework of the Frontier environment where we guarantee network integrity (more on this later) at the potential risk of network rollbacks if/when unexpected behaviors or other issues emerge.
Considerable progress is being achieved on all fronts. Instead of providing you with a release date and then facing the risk that something might slip through the cracks under the pressure to meet that date, we will systematically address items in the release process as we progress. This will give you a sense of how far from the launch we are on any specific day, without us making assumptions about how long it will take, publishing those assumptions, and hoping we can align reality with the timeline: as we mentioned earlier, this is a security-driven rather than a schedule-driven approach. We intend to make this process fairly transparent for you, providing summary information so you don’t need to monitor the repositories – but we prefer to engage you in the process so that you comprehend what is occurring and how the work is advancing rather than simply providing an aggregate date and leaving you guessing about the true state of progress. We believe this approach is more transparent, better suited to a crowdfunded development model, and overall significantly more pragmatic.
If you would like to view the current state of affairs, Jeffrey Wilcke (lead on the Go client and the Amsterdam hub) has a Go client issue tracker for Frontier.
This issue list serves as the best reference for the Frontier release: currently, 10 issues are closed, and 26 are open. New issues will arise, and naturally, the issue count is merely a rough proxy for “ship time” but the completed work is the most reliable indicator, so check there if you wish to speculate.
Please be aware: this does not imply that we are operating on “Valve Time” (although Valve Time does produce some remarkable products!) Our development processes are transparent, you can observe the code being created, download the latest branches, and participate in testing. It’s not all concealed behind a corporate PR barrier. However, we are functioning on “releasing groundbreaking software accompanied by significant security processes” time, and estimating schedules for initiatives that have never been achieved before won’t benefit anyone: part of the advantage of Ethereum’s funding model compared to traditional funding methods is that we can prioritize what is correct in the medium- and long-term, rather than being constantly pressured by market demands for next quarter’s figures.
We will state this: there will be a minimum of two weeks’ notice before anything goes live. You’ll receive ample updates from blog entries, you’ll observe the decline of the issue count, and we’ll begin to convey positive feedback regarding security audits and so forth. We’ll keep you informed!
How about the feature list? We’ve primarily gone over this in the Launch Processes post, but to reiterate Frontier will…
- Be a command-line client exclusively
- Mine at 10% of the standard rate, but this will be genuine Ether
- Include the complete blockchain feature set containing smart contracts and logs, though everything other than account balances will be wiped when Homestead is initiated
- We are collaborating with exchanges to facilitate ETH conversion during Frontier
- For security, the blockchain will be manually checkpointed every 24 hours, and any reported irregularities will be thoroughly examined
- Official exchanges will utilize this checkpointing service to safeguard traded assets from potential blockchain rollbacks
- In instances of particularly severe failures, the Foundation may cease checkpointing Frontier entirely and release a client upgrade
In summary, we’re doing everything feasible to create Frontier a secure environment for testing Ethereum with real value, yet we are also strongly advising users against utilizing Ether on the Frontier network if they are not prepared to lose it.
Frontier is primarily aimed at individuals who are developing and testing tools for mining and exchanges, along with possibly a few of the more resilient dApp developers. It is not a general release we expect average users to engage with at all, though you might download a client and mine a little Ether simply because you can. The main event truly begins at Homestead, and we’ll share more details about Homestead’s features soon
Additional updates as they come, and keep an eye on the skies!
