decredcommunity / issues

Decred community issue tracker
5 stars 0 forks source link

Proposal discussion and vote duration #117

Open xaur opened 5 years ago

xaur commented 5 years ago

This issue captures a concern that pre-vote discussion and vote itself are too short, which may negatively affect comment and vote engagement.

Personal "late voter" perspective: studying a typical proposal easily takes me 2+ hours. If several proposal surface concurrently, I need some 8-10 hours to form opinions and come up vote choices for the batch. When proposals show up on Politeia, they sort of demand that I allocate time among other activities, usually within ~1 week. As a result, I barely make it with some proposals, and a few I skipped entirely.

This is a manifestation of the "attention span" problem that was anticipated well before the launch of Politeia. People are busy and can only afford X hours per week on the proposal activity.

Perhaps I'm just bad at organizing my time and this issue is to learn if many people are having this problem. If not, close this issue and use it as a reference.

As a general rule, I would suggest longer discussion and voting windows for proposals with bigger impact (in both spend and policy). For anything larger than $50K, I think a default should be 14 days to discuss on Politeia and 14 days to vote.

One idea to study the problem better is to survey how long people spend to decide on an average proposal.

Discussions:

ghost commented 5 years ago

See TLDR & suggestions below I have this problem too. I joined Politeia this week and there were 7 proposal in pre-voting and voting phase. Even for the simple but well prepared proposal like one for YT videos took time to read properly the proposal and comments, and then understand the choices being made like price, future usefulness, details of content, revisions etc. this easily takes few hours. since this proposal was already in voting stage nothing further was possible but i still provided feedback about why this is a good/bad proposal so those who had not voted yet can make a better decision. and then i voted.

For a proposal that is in pre-voting apart from above i asked questions about things that were not clear to me and got answers on them. This question-answer cycle question takes abot 12-24 hours (mostly due to timezones and when people are free i guess) in my short experience with 3-4 proposals Sometimes it takes 2 or 3 such cycles to get clear understanding and answers. Sometimes i had to post links to my comment in chat to get answers, some of them were given in chat which then i had to link back to Politeia in my own comment.

Based on above i think at the very least a week of pre-voting period should be default and i recommend 2 weeks as optimal for someone who missed a proposal being submitted.

This was one side of the problem from stakeholder point of view. Other side of problem is that submitters have experienced no engagement when they submit the proposal on Politeia. This has led to people first discussing it in chats and then on reddit before submitting it to Politeia. This leads to whole process of refinement from draft to final submission remaining undocumented and unarchived on Politeia. Also this might be the reason for low engagement later on Politeia. People have answered that question i am asking was already discussed elsewhere, probably in chats/reddit and probably implying that i should have been involved then to receive the answer and i am just repeating past discussion. I assume this as no links were provided to these past discussions. This is the disadvantage of not doing this process on Politeia, not everyone is or can be present in all chat/platforms all the time and valuable discussion is lost.

So to summarize

Suggestions to improve

More suggestions are welcome.

ghost commented 5 years ago

Additional comment: I think people are not aware of two things which maybe leads to them not joining Politeia, based on my own experience.

  1. They can comment by simply signing up and paying 0.1 DCR once, even if they do not have any tickets or not enough DCR to vote and still be influential via their comments.

  2. Their Politeia account will not reveal how much DCR/tickets they have explicitly.

xaur commented 5 years ago

Great writeup, thanks.

Sometimes i had to post links to my comment in chat to get answers, some of them were given in chat which then i had to link back to Politeia in my own comment.

This rings some bells. People either don't get notified promptly via their preferred channels (see #24), or they are reluctant to "load" Politeia to reply. The latter is a combination of

submitters have experienced no engagement when they submit the proposal on Politeia. This has led to people first discussing it in chats and then on reddit before submitting it to Politeia. This leads to whole process of refinement from draft to final submission remaining undocumented and unarchived on Politeia

Spot on! I think we should recommend to minimize the chat and reddit phase as much as possible. As soon as the idea is deemed worthwile, the author must be sent to draft it on Pi to start discussion there sooner. Also, discussions in chats and reddit should be gently redirected to Pi when reasonable. The challenge here is we're getting into the territory of small and low quality proposals (#33).

There is also a problem of people not wanting to comment on Politeia, even the people submitting proposals.

I believe this is a result of lacking notification features (#24) and difficulty to track unread comments. Put simply, information is lost and/or not delivered to trigger a reply. Compare your Pi experience to GitHub, where you can check the bell icon once or twice a day and stay on top of everything. Besides the bell, linear chronological discussion helps to continue where you left reading commens last time.

maybe announce once a day or every other day via twitter handle @slices_of_pi when a proposal is in drafting or pre voting phase

This can be a combination of hand-crafted slices and to-be-automated pi_crumbs account (tracked in #102).

xaur commented 5 years ago

This leads to whole process of refinement from draft to final submission remaining undocumented and unarchived on Politeia. Also this might be the reason for low engagement later on Politeia. People have answered that question i am asking was already discussed elsewhere, probably in chats/reddit and probably implying that i should have been involved then to receive the answer and i am just repeating past discussion.

Marketing 2019 is a good example of this. Politeia readers may be unaware of a few useful discussions that happened on Reddit pre-proposal, like detailed podcast listing or the idea to break the budget down into returning/flavour/investment buckets. (by the way @Dustorf, linking to past discussions is something to consider for future proposals).

Marketing 2019 pre-proposal is also a good demonstration of the "unarchived on Politeia" aspect: 7 comments "just disappeared" on Reddit.

ghost commented 5 years ago

User comment scores discussed here and presented in latest Pi Digest are a good example of social status incentive that can be used to drive engagement.

xaur commented 5 years ago

Possible plan for rolling out the discussed recommendations:

  1. adjust docs
    • recommend to minimize discussion outside Politeia
    • recommend to let Politeia discussion for min 10 days
    • recommend longer discussion and vote window for proposals with bigger impact
  2. add cosmetic/awareness/help tips to Politeia UI to guide authors
  3. write up and discuss missing Politeia features mentioned in this issue (e.g. drafts tab)
xaur commented 5 years ago

From chat about decredcommunity.org:

I think the closed source nature of community site is another instance of a proposal that validates the call for longer mandatory pre-voting period for proposals. Sometimes people need more time to come up with important questions, I am sure jy-p would have read the comment/asked the question himself and karamble would have had time to update the proposal if it was not already in voting.