Dash Core Team
Dash Support Group
- Feb 13, 2016
Each month, the ten most popular questions submitted by the Dash community at Dash Nation Discord are collected and sent to Dash Core Group for answers. The intention is to select important and well-thought out questions and filter out unpopular questions, questions that have already been answered and other noise. For February, some remaining questions from the 2018 Q4 Summary Call were also included. Some questions were not answered in time and will be carried over to next month. Apologies for this and the late posting - my capacity was taken up by the new website launch. The submissions for February 2019 are as follows:
- Q: Are there studies on the possibility of having redundancy in masternodes in case of failure, so there can be automatic fallback to another node without interrupting service? (Jol)
A (codablock): Some groundwork for this has already been done with DIP3, which makes something like cold-/hot-deployments easier to implement. There are however still some issues that we need to look into and fix.
- Q: Can we imagine a way to use the masternodes as repository for the Dash source code and binaries? With an app built on the top of evo and IPFS maybe? (Jol)
A (codablock): What many people don’t know or have forgotten is that Github is just an interface to Git, and that Git is a distributed version control system since day one. If Github stops to work at some point in time, or tries to censor projects, hundreds of people already today have a full copy of the source code and its full history locally. Also, every “commit” in Git is represented by a hash that is built from the whole history of the source code, meaning that the source code is already secure against modification. So, there is no need to put it onto the blockchain.
- Q: Do you have any plans on improving UI & UX of current Dash apps in near future? For example Dash Core Wallet does not support Retina displays and still has quite «geeky» design.
A (Liz): Yes, our design team is currently working on UI redesigns for both Android and iOS DashWallet apps. We do not yet have a release date for these, but it is planned as a phased rollout with the goal of creating a fresh and unified look and feel across the DCG apps, and preparing them for the implementation of Evolution features like blockchain users.
A (strophy): Some discussion of specific issues relating to Dash Core is available here. In general, Dash Core is primarily designed to run as a headless server daemon and not a desktop wallet. Development focuses on innovation at the blockchain level, and it is unlikely resources from frontend developers will be redirected to redesign the GUI. That said, I have looked into what it would take to do this myself. The user interface is described in /contrib/dash-qt.pro in the project, so anyone can open this in Qt Creator and make some changes to the *.ui files, and submit this as a pull request. All that should be needed are some basic vector design skills
- Q: Last year, I believe as recommended by DCG developers and reported by Dash News, it was identified as a best practice to use fewer than a dozen inputs when using PrivateSend. When someone creates a masternode collateral using PrivateSend, due to the large amount, this can involve hundreds of PrivateSend denominated inputs. The recent introduction of the 0.001 PS denomination will further increase the average number of inputs for a PrivateSent MN-collateral creation. In Antti Kaikkonen's testing here, it was suggested that PrivateSent transactions can be analyzed by conducting a series of random traversals backward through the blockchain. The reported findings indicated that there is a positive correlation between number of inputs used in the PS transaction, and the accuracy of the probabilistic source of funds. Has DCG analyzed the privacy of MN-collateral PS transactions using this method or other methods? Has Antti-Kaikkonen's method been tested on actual MN-collateral PS txns, and if so, what were the results of the test? How confident are we that MN-collateral creation txns that use PrivateSend are sufficiently anonymous? (TroyDash)
A (UdjinM6): MN collateral creation would already require 100+ inputs regardless of recent introduction of 0.001 PS denomination and thus can be seen as a not-so-anonymous one due to known technics like clustering analysis. New denomination should slightly affect smaller amounts making them less costly to send by adding another input or two instead of spending leftover as a fee. However, adding new inputs also rises the size of the transaction and thus increases the fee. Please note that wallet input selection algorithm should balance between the two (number of inputs and the fee) trying to minimize the fee. We believe that number of inputs shouldn’t increase too much in general which should preserve good privacy. We’ll keep an eye on PS transaction patterns and will make corresponding adjustments in future versions if needed.
- Q: In the quarterly call Dash Ventures was talked about, with the way that it benefits the network being to buy up Dash and destroy it in order to hypothetically drive up the price of Dash on the open market. Have there been internal discussions about using this Dash to instead stabilize the price to increase its utility as a currency? Could you please comment on why destruction of the Dash is favored over this? (name3)
A (Ryan): DCG has no intention of directing Dash Ventures or what it will utilize available funding to do. This will be up to Dash Ventures to decide, and indirectly the network to decide through the supervisors that oversee Dash Ventures activities. Dash Venture profits could be reinvested, used in trading, set aside as investable cash, or returned to the network in the form of buying and destroying Dash - or any combination thereof. We are not favoring or advocating any one option. Rather, our aim was to provide distributions as one additional option, and this was the most effective means we identified to do so. If the network wishes Dash Ventures to play the role of market stabilization, it should elect supervisors that advocate for that role.
- Q: In the spirit of Dash's international nature, would it be possible to translate the blog at https://blog.dash.org into multiple languages? (Jol)
A (strophy): The DCG blog is currently hosted using Medium, which does not support multilingual posts. Instead, each post in another language would be pushed to all followers of the main blog. They are primarily expecting English content, so this is not ideal, particularly if we are publishing each post multiple times in multiple languages.
We may consider exploring other options to publish multilingual content in the future if a clear business benefit can be demonstrated. This may be on Medium or a more suitable platform, but this is not currently planned or budgeted. For now, community and news sites offer the best source of localised news about Dash.