I voted FOR.
Interested to check the findings and the possible next steps.
1 Like
I voted For this proposal on Snapshot. The initiative seems worthy of the relatively low risk and effort required to enable the Discourse API and Discourse public channel access.
1 Like
Thanks for clarifying! So the bot is just a normal member of Arbitrum Discord without any administrator permissions. Even if there is a security issue, it will not have any impact on Arbitrumās Discord, right?
We do not have any specific objections to TogetherCrewās implementation of this research, considering that the DAO has no budgetary requirements tied to this proposal.
Overall, we believe this is an excellent contribution to the academic research on DAO conducted by researchers at TogetherCrew and the University of California, Santa Barbara, as well as potentially benefitting from findings that can share more profound insights on the Arbitrum DAOs.
We look forward to reviewing the research findings.
1 Like
yes, exactly.
The downside is that deploying our bot takes a bit longer as you need to give permission channel by channel sometimes. This is why many other bots opt for admin permissions but we know how that endsā¦
1 Like
Nice! I personally think it is desirable to sacrifice some efficiency for the sake of security! Thank you very much!
1 Like
Iām voting in favor. This research seems really interesting plus it requires no budget and it only uses data that is already publicly available. Initially I was concerned about potential security issues but I saw that it was already brought up in the previous discussions, leaving me with no other major doubts. Iām curious to see the results of this research!
1 Like
I have no issues with this proposal, and it should move forward. No budget is required, and thereās no security risk since it uses an open-source API. Importantly, if approved, this proposal will help identify potential members of the community.
After reviewing feedback from other delegates, we decided to support this proposal as it provides valuable insights to the community. the research would be helpful in understanding the data, and it would be beneficial for gaining a deeper understanding.
voting Against the current offchain proposal because we shouldnāt treat Discord open channels data as public data, since it isnāt public data. I would maybe agree with this proposal if it only used Discourse forum data, which can be argued that it is public, and also if all the analysis and dashboards would be shared with the broad community and submitted under CC0 license.
voting AGAINST the offchain proposal.
The fact that the data is āpubliclyā accessibleāmeaning it is shared on social media, public forums, or platforms like Discord where most people can view it by creating an account or receiving specific permissionādoes not mean it is free to be collected or processed.
The content that will be scraped by the bot includes both personal data (which identifies or makes individuals identifiable) and potential IP rights over proposals and/or feedback. While there are exceptions to the application of the relevant regulations, I havenāt found details on how the data is handled or why it would fall under such an exception.
The DAO cannot approve engaging in a practice (data collection for processing) that it does not have the authority to carry out. The data belongs to individuals, and the IP belongs to the individuals, and they must authorize its use. In any case, the Discourse and Discord policies should be reviewed beforehand
2 Likes
@danielo isnāt it against Discord Privacy Policy and looks similar to other bots data farm such as Spy Pet? cf https://support.discord.com/hc/en-us/articles/360039598252-Protecting-Your-Data
Discourse has probably set up similar privacy policies to fight against these kind of data farms.
1 Like
We went through a discord review of our policies to verify the bot. Past 100 servers itās mandatory, otherwise you canāt add it to new servers.
Also for clarity please note that anyone can request their data be erased or theyād be given a copy, as per our policy (independent of the applicability of GDPR or not in certain jurisdictions, we consider this a fundamental right).
I will be voting against for similar concerns @pedrob raised. Iām not sure the DAO is really one to have say over this, as ultimately the DAO is (as far as Iām aware) not in control of the discord channel. And probably more importantly the DAO shouldnāt really be making decisions for individual users privacy.
A favourable snapshot vote will give the foundation the permission to give us read access to
the Discourse API and public Discord channels.
I guess maybe itās this line that confuses meā¦ if the foundation is the one with the permissions why does the DAO need to be involved at all? I get it as a general sentiment check, but presumably there will be people in the Discord who donāt have voting power that would not have a say. For public channels I donāt see an issue, as privacy isnāt expected but from what others have said it seems the Discord has different rules on this.
All said, Iām not opposed to the broad concept. It seems like it can provide valuable data at no cost - which is great! I guess the question becomes is this something that can be done with publicly available information? Or is there a way users can Opt-In to the Discord so only those who elect to do this can participate.
Edit: I see there is still a day to vote, Iāll hold on the vote until tomorrow incase additional info comes to light
1 Like
In favor of this.
Based on the answers given by @danielo, we are satisfied, as we had similar concerns to those raised by other delegates.
The open-source nature and focus on analyzing only public data ensures transparency while minimizing any privacy concerns. We also appreciate the research-driven approach, especially the collaboration with academic institutions, which adds credibility to the analysis.
Understanding how Web3 communities operate and retain members is crucial for the long-term health of DAOs like Arbitrum. This proposal seems like a great opportunity to gain those insights.
Weāre looking forward to seeing how this works ;ā)
1 Like
We vote FOR the proposal on Snapshot.
We appreciate the potential value to be provided by the research from TogetherCrew team without cost from the DAO treasury. However, we are not certain that the DAO is the one to decide whether the API access to Discord is allowed, while most of the Discourse data is public and much related to discussions around the DAO. Either way, we support the concept and its modified implementation that addresses the concerns from the delegates and related parties.
1 Like
For context, we engaged with the Foundation requesting said access about 2 months ago, and after reviewing everything, they requested we get confirmation from the DAO.
As I understand it, itās not so much about the legal/permissions side of it, but also about confirming buy-in for the idea.
Thank you for your reply.
I appreciate the transparency and your willingness to work toward a positive outcome. However, I think we should not pursue identifying who is who, doing what, how, when, and where. Privacy on Arbitrum should be strengthened, not weakened. A recent example of privacy concerns was demonstrated by the leaks from Fractal ID, an issue that, to my knowledge, the foundation has not yet addressed. We continue to use their services for KYB/C, even though victims are still suffering from the consequences of this leak.
I mean, this is very different. Fractal was focused on verifying someoneās real identity using VERY sensitive data. We donāt even collect email addresses. In that sense, weāre not identifying who is who.
Thank you @danielo for the additional context. An update to my post above (Research on context and retention - #54 by Bob-Rossi) - after sleeping on it and the additional info I will maintain my āAgainstā vote.
I do not have an issue with the research of public content and can see the benefits. When someone posts something publicly accessible there is understanding that data can be harvested. Danielo looks to have put a lot of thought, as well as appreciated guardrails & limits, to the data collection. It is unfortunate the Discord situation has created a legal grey area.
As for the why of the no vote, I will repeat what I mentioned above. The DAO shouldnāt be in charge of determining if user data should be scraped like this. It goes beyond on the scope of the DAO. For a site like this where data is posted publicly, itās not like the DAO can stop people from scraping data so the discussion is a moot point. And for Discord, the AF is noted as being in charge of the Discord so I am failing to see why the AF wants to put that decision on us as a DAO. As far as Iām aware, the DAO has not had to vote on any other Discord-related decisions.
I would be curious to see the AFās @Arbitrum reasoning why the DAO is being asked to give permission on the Discord portion of this given the DAO has no control over the Discord channel. As truthfully it feels like they acknowledge this is a legal grey area and the burden of that is being pushed to the DAO.
A quick edit to add:
As I understand it, itās not so much about the legal/permissions side of it, but also about confirming buy-in for the idea.
If the goal is confirming buy-in, is this going to Tally? And if itās going to Tally does it really need toā¦? The snapshot vote and this thread has provided a lot of context on buy in, going to Tally to re-confirm that seems like a waste of time.
1 Like