I clicked on the green “Vote” button, expecting to be taken to a list of candidates, but only got a “your vote has been recorded” message. I guess I’ve managed to submit a blank vote.
I assume my browser failed to load some external Javascript, but I didn’t notice an error message – it all seemed that the actual voting would happen on the next page.
So, if you’re on a flaky connection or use an ad blocker, beware!
FWIW, the same thing happened to me when I went to vote using my phone. At least I contributed to quorum. 
-eric
Was able to reproduce by blocking the loading of OpaVote’s javascript.
I’m not sure how we can proceed, but I’ll do two things:
I’ve reached out to OpaVote’s support.
We unfortunately do not have a way to invalidate a ballot once it is cast (that seems good… kinda?), we can issue another ballot using a token but that relies heavily on the honor system that a ballot was actually unintentionally cast. Our ballots are anonymized so there is not a way for us to audit if a ballot was actually cast empty for a specific user.
Overall, it is an unfortunate situation. While I would like to be as helpful as possible to resolve this It feels too ambiguous.
I just tried to vote. I retrieved the OpaVote mail from my spam folder and clicked on the link. I saw the list of candidates, checked my preferences, and clicked the “Vote” button. Then…
I got the following error. What to do? I definitely did not vote before.
Doesn’t look like the error you indicated was uploaded.
Thanks for looking into this! Yeah, I’d expect a voting system to behave that way.
Hmm, maybe it would be possible for OpaVote to disable casting empty ballots?
As far as I’m concerned, if these are isolated incidents I’d rather keep my vote blank than make you allow re-casting on a honor system.
Same here.
-eric
Sorry, the mail gateway ate the error. Here it is:
That was unexpected ... This voter has already cast a vote.
Some mitigation ideas to prevent miscast votes going forward:
<noscript> fallback stating that JS must be enabled to vote, as they should be doing anyway
noscript either?Report to OpaVote was released from moderation just now: https://groups.google.com/g/opavote-support/c/rVjSTmmKHZk/m/nisnh5wAAwAJ
@CAM-Gerlach I think those are good suggestions.
That may be worth submitting to their support forum.
They already have sane <noscript> behavior, this is more subtle as its if a specific js file failed to load for any reason. It is also already on the same origin.
There’s no confirmation screen as I recall. Again, that’s a suggestion that could be made separately on the support forum.
So what about my vote? Who stole my vote? I use gmail and have never experienced stolen email before. So I suspect rather something else happened. But what?
OpaVote states that a ballot was cast for your address at 15:45 UTC on June 21st, which is shortly before your initial message stating an error.
Is it possible you double submitted the ballot?
I don’t recall pressing the Vote button twice. But given how much went wrong for others I wouldn’t rule out that the UI made it easy to do so accidentally. Or that some JS bug caused this to happen.
Is there a way for me to recover from the system how it thought I voted? Or is that information lost forever?
OpaVote is configured with anonymous ballots, so it is not recoverable.
The results are in!
The OpaVote page says there were 9 empty ballots. We know from this thread at least two were due to no candidates being listed due to a bug. It’s unknown if the others were intentionally or unintentionally blank.
But the good news is there were more than 9 votes separating each of the top 4 vote winners, and from the number 5 spot.
And the 5th place will also serve a one-year term to fill an off-cycle resignation, and there were more than 9 votes between 5th and 6th.
Which all means: even if all 9 blank votes were caused by the bug, if they had instead been successfully placed, they could not have materially changed the result of the election.
I was just thinking of writing a blog post on the elections: There has been some great strides on the different people being elected on to the board this time round. Do we have data on the amount of members who were eligable to vote? I do know from Deb’s post that affirmed voting members was 877 (which will be a subset of those eligable).
Well folks, for what it’s worth I wrote some thoughts down on the recent 2023 elections. I’ll need to bring these when I next meet with my fellow D&I WG members. Thank you @EWDurbin for managing the elections, as always.
I was a bit confused about what the “Voter Count” section of the linked post was trying to say. It brought up the (entirely valid) point that a like-for-like comparison of turnout percentage between 2022 and 2023 is not representative due to the change in the number of eligible voters, but then appeared to focus on the equally unrepresentative percentage change of the number of eligible voters as a proxy for election quality, which isn’t really a meaningful metric on its own due to the major procedural changes to how eligible voters are counted—something which I don’t see mentioned anywhere in the post.
A much more meaningful number to focus on, I would think, would be (as the section title suggests) the voter count, i.e. the total number of people that actually voted (vs. just those who were counted as eligible to vote). However, I don’t see that mentioned in the post, other than the statement:
Regardless of the efforts we’re putting in, at least from within my community and also from within the D&I Workgroup which I am part of, we can’t seem to get a steadily increasing voter count year-over-year.
In fact, the numbers in the presented in the post show the voter count is actually generally increasing year on year:
| Eligible | Turnout | Voters | Error | YoY Change | Cumulative |
|---|---|---|---|---|---|
| 1151 | 0.40 | 460 | +/- 6 | - | - |
| 1538 | 0.38 | 584 | +/- 8 | +27% | +27% |
| 1459 | 0.39 | 569 | +/- 8 | -2.5% | +24% |
| 877 | 0.70 | 614 | +/- 5 | +8% | +33% |
While of course these data do not establish any casual relationship (nor correlation), it would seem to anecdotally suggest that perhaps the efforts of yourself and others to broaden participation in PSF elections might potentially be more fruitful than you’ve realized 
Thank you for reading my post and also for your comments.
TLDR; I understand what your pointing too, and I agree.
Yes, I was aware of the difference of procedural changes that now requires us to affirm our voting intentions and the reasons behind it, but I naively took that as an apples-to-apples comparison to how many eligible voters we have, when in fact it shouldn’t be.
In my mind during writing my comments concerning this and previous elections, the most important number for me is the eligible voters, or the “voter base” (which in the article is written as “voter count”). This represents not only the amount of people who can vote, but on a broader scale the amount of people who are participating and have participated in efforts for the community and the PSF, i.e top-of-funnel (as opposed to indifferent community members). Increasing this number is an important index to see that our efforts for increased community (not only for the election) participation are bearing fruit.
So that is why I was looking for the same “eligible voter” as I did as last year. But as you pointed out, as least within the context of our elections, this number no longer plays a role. Affirming your intention to vote is just that: Affirming your intention to vote in the elections.
Which brings me to the next data which you pointed out to: The cumulative increase of people voting. This is something which I totally did not look at. And as you have rightly pointed out, for one reason or the other, it has been increasing. Hopefully it’s increasing because the top-of-funnel is also increasing in numbers. But nevertheless, it is still progress, and I am sure all of us are happy to be part of that.
Thank you again for pointing that out.