Engineer on @googlechrome. Involved in CSS and W3C standards. Previously @mozilla, @w3ctag. Mastodon: @dbaron@w3c.social
Rockville, Maryland, USA
Joined March 2008
- Tweets 7,521
- Following 1,244
- Followers 4,296
- Likes 18,519
I'd prefer voting systems that really supported multiparty democracy, including things like mixed-member-proportional legislatures.
But I probably lean against accepting the unpredictability of cross-party IRV (or, worse, California's way) over the current 2-party system. 6/6
1
1
In this particular case, if a relatively small number of voters whose real preference was Palin-Begich-Peltola had instead strategically voted a Begich-Palin-Peltola ballot instead of a Palin-Begich-Peltola ballot, Begich probably would have won instead of Peltola. 5/6
1
Instant Runoff Voting tends to produce results that are very dependent on how the candidates are distributed across the political spectrum, and also most strongly emphasizes the top end of voters' rank lists. (So does FPTP.) It's also pretty susceptible to strategic voting. 4/6
1
2
As far as I could tell from looking at the Alaska elections website a few days ago, I don't think Alaska has made the full ranking data of the ballots available, so I can't verify this. 3/6
1
While folks on the left may be happy about this particular result, I think it's worth being clear that Begich was very likely the Condorcet winner in this election. That is, Begich probably would have won both Begich-Palin and Begich-Peltola runoffs. 2/6
1
Ranked Choice voting may have exciting possibilities... but I'm not so excited about the nearly-standard use of Instant Runoff Voting to count Ranked Choice ballots. Ranked Choice ballots can be counted in many ways. 1/6
1/ I continue to be really excited about the possibilities of Ranked Choice Voting, especially after the Alaska election.
1
2
Not only am I with you on preferring direction... but I wish every city built their subway exits like Sapporo and had a compass on the ground at the top of the stairs.
(I think I've seen this in some other cities in East Asia, but I think it was most reliable in Sapporo.)
1
1
And you can adjust them to be quarterly by adding Q to the end of the URL.
(Also, mentioning that these are employment-population ratio for age 25-54, so I have a chance of finding this again.)
1
Oh, and I found the one I was really looking for (you can adjust them to be 1948-2022):
beta.bls.gov/dataViewer/view… (men, seasonally adjusted)
beta.bls.gov/dataViewer/view… (women, seasonally adjusted)
beta.bls.gov/dataViewer/view… (men, unadjusted)
beta.bls.gov/dataViewer/view… (women, unadjusted)
1
To analyze that final round you need data that's not in the existing transfer data, because you need to know how Palin's first choice votes would have transferred.
2
1
The difference in 1st choice for Palin vs Begich was 3.1%, so shifting 1.6% of the voters would cause the needed 3.2% swing and (making some assumptions about write ins, since they didn't see a chart with write in transfers separate) make the final round Begich vs. Peltola.
1
For example, I think it's likely that if 1.6% of the Alaska at-large voters had voted Begich-Palin rather than Palin-Begich, that Begich would be Representative-elect rather than Peltola.
(I don't know if there's public data to confirm this, though there would in San Francisco.)
1
Depends on the voting system used to count the ranked choice votes, I think.
With STV it's fine to cast a protest vote for a third party that has no chance.
But once it's not clear who the final two are, strategic voting is important again.
1
1
L. David Baron @dbaron@w3c.social retweeted
I knew America had disproportionately high traffic fatalities, but I don't think I realized just how truly bad we are.
319
1,460
341
7,704
This reminds me of engineering-wide emails at Mozilla (many iterations of management ago):
Don't break builds/tests so often
or a few months later:
Don't use so much machine time on try runs
Both messages needed to be targeted. (And they weren't even hard conversations!)
This seems like a common management failing: blasting out feedback intended for specific employees to everyone because you don’t have the, um, guts to have hard one-on-one conversations. joshbarro.com/p/digital-reve…
2
10
The engineers working on shipping it in Chrome showed that it was a lot harder than I thought it would be... but it was, in fact, possible.
(See github.com/w3c/csswg-drafts/… for just one example. It turned out that the way CSS Transitions are started wasn't compatible with CQ.)
1
1
Getting a lot of mentions today about Container Queries shipping in Chrome. It feels a little odd given that it's something I haven't been very involved in for the last 2 years... though I did help start it down its path by claiming an approach was possible to implement.
1
7
L. David Baron @dbaron@w3c.social retweeted
There is no part of America that is more "real" than any other. There are no Americans who are more "real" than any others. I've really come to detest this phrase and what it implies.
65
345
12
2,217
I was able to figure out what changed MXCSR once I knew that was the problem. I don't remember how I did it, but maybe it helped that in my case it was a push/pop change that was on the assertion failure's stack, so I could just go up the stack in gdb+rr. bugs.chromium.org/p/chromium…
1
1