Hacker News Reader

Comment by jawns | Hacker News Reader

The author_is_elon flag doesn't surprise me, but the two political designators are somewhat shocking. I'd sure like to know what changes based on what Twitter knows about your political affiliation.

View on HN

I thought it was interesting how it explicitly doesn't boost independents. So much of the two-party system is self-reinforcing.

Or anti-statists! Not everyone who engages with politics is a bootlicker for authoritarianism but it sure feels like there’s no space made for this perspective (obviously)

The vast majority of self-proclaimed independents vote with one party just as reliably as registered members.

American's lax attitude towards cultivating more than two parties is literally killing the republic from the inside.

You'd think a country that played a central role in a global, decades-long unstable regime of bipolar power that routinely pushed mankind to the brink of nuclear oblivion would know better than to have a bipolar electoral system

> routinely pushed mankind to the brink of nuclear oblivion

I'm struggling to think of a reason why this is anything but bad faith nonsense.

have you never heard of the many times we almost came to setting off full scale nuclear warfare because of the bipolar power war of the ussr v usa et.al.?

the only thing that saved us was cooler heads that prevailed on both sides.

Why would throwing more actors with similar capabilities into the mix make the situation any more stable though? That seems like basically the old European Balance of Power, which broke out into open conflict more frequently.

I take the view that the European balance of power probably broke out into open conflict often because of hidden alliances that made it hard for states to correctly gauge the costs of engaging in such conflict.

In the field International Relations, there's lots of discussion around the stability (or instability) of a bipolar distribution of power. Your stance is closer to neorealists, mine is maybe closer to classical realists. Have fun reading https://en.wikipedia.org/wiki/Polarity_(international_relati... and related links

But from a game theory perspective, having only two powers turns everything into a zero-sum game. I argue that leads to less cooperation and increased divisiveness. Not agreeing with Republicans means you must be a Democrat in their eyes. This "us versus them" mentality is somewhat tribal and leaves little room for nuance.

If I had a magic want, the US would have a 4 party system. Like a cartesian plane with civil liberalism <-> conservatism on one axis and economic liberalism <-> conservatism on the other axis

> Classical realist theorists, such as Hans Morgenthau and E. H. Carr, hold that multipolar systems are more stable than bipolar systems, as great powers can gain power through alliances and petty wars that do not directly challenge other powers; in bipolar systems, classical realists argue, this is not possible.

I suppose they must account for this somehow but isn't that exactly what a series of proxy wars in far-flung places between the United States and the Soviet Union were?

It's been almost 20 years since I studied any of that, but those proxy wars in far-flung places did directly challenge other powers because of the Domino theory

I don't necessarily agree with Morgenthau and Carr as I think most IR Theory is bullshit made up by academia... particularly the stuff around how players "gain power"

So I make my own argument which mostly hinges on the idea that two powers really means "my power" vs. "everyone else" which is not a recipe for peace

Isn't World War 1 a counterexample? Lots of parties, but they aggregated into two competing coalitions.

This balance is the central theme of the seminal work in the field (https://www.jstor.org/stable/j.ctt5vm52s). It may or may not be correct but is one of the most influential texts on the subject.

I've not heard how many times, no. How many times was it?

Luckily there's a whole article on it https://en.wikipedia.org/wiki/List_of_nuclear_close_calls

That's all nuclear incidents ever, not just US/USSR. US/USSR count from that page is more like 10 incidents across 40 years. Some of them occuring within days of others, they're that fine-grained.

I wouldn't call that a "routine" matter.

It isn't an attitudinal problem, it is the logical outcome of our political systems. In political science it is known as Duverger's law: single ballot, winner take all systems inevitably tend toward a two party equilibrium.

Changing this requires states to adopt alternative systems, which can sometimes mean amending state constitutions. It isn't easy or straightforward, and the general sense is that there are better things to spend that effort on.

> single ballot, winner take all systems inevitably tend toward a two party equilibrium.

I'm not convinced it's quite that simple.

For example, Canada also has a first-past-the-post electoral system - yet political parties here have come and gone. And continue to do so.

Parties come and go, but there will be two major parties. In fact under such as system if you're in one of those two big parties and you see a third party rising you need to figure out whether it's you or the other guys getting replaced, 'cos it won't stay a three party system for long.

For example in the UK, the Tories ("Conservative and Unionist Party") and Labour are currently the biggest parties, but a hundred years ago this was a novel situation, Labour were seen as a third party, while the Liberal party (which was eventually absorbed into what is today "Liberal Democrats") had seen success over decades and were often in government prior to that point.

I mean we're on the sixth "party system" in the US too, if that's your standard (it's a new "party system" when there's a significant realignment of which interests find themselves in which parties, either by a reshuffle among existing parties, or by new parties rising and old ones falling). Have you heard much about the Federalists or Whigs lately?

The voting demographics change frequently as well. California has voted to elect more republican presidents than it has democrat ones. It’s voted democrat in the previous 8 elections. In the 10 elections prior to that, 9 times it voted for republican candidates. Texas has also voted for more democrat presidential candidates than republican ones. People who think the US democracy is a rigid and highly predictable system simply have a recency bias.

A lot of it depends on whether your governing coalitions are formed before or after the election.

The US parties are just coalitions of disparate interests joining together until they (maybe) represent enough people to have a majority and be able to enact their collective interests.

In some other countries the different interest groups sort themselves into two factions after being elected but I don't know that it is really that different in practice.

It's not different, or rather it doesn't produce meaningfully different outcomes. I'm not aware of any parliamentary system with a wonderful diversity of thought and a long record of positive accomplishments. You end up with ruling coalitions which are typically pretty awful.

Ironically, America has one of the most open political systems. You register as one party or the other and vote in primaries. This has lead to a huge variety of people replacing hated mainstream politicians. That's way more than you can say for many other countries.

The two parties are essentially coalitions.

Anecdotally, a recurring theme in conversations I've had while living abroad is the desire to prune or consolidate some parties.

While both sides in the US have big tents, they are effective in whipping votes when things need to get done.

It also helps that detracting coalition partners can't torpedo their leadership. Historically, factions within a party, like Blue dog dems or Tea partiers, had to wait for an election to litigate their grievances.

Well, yes, this was where I was going. The coalition is just formed before the election takes place. I live in a city with "nonpartisan" elections and I feel like it mostly just makes it difficult to understand what candidates in local elections even stand for.

Agreed, when I lived in Austin the non partisan elections for city council made me feel ill informed come election day.

It just fueled apathy for me: there's only so much research one can do with limited press coverage and personal time constraints. Shrug.

>it doesn't produce meaningfully different outcomes

It absolutely produces different outcomes than a two party system. Smaller parties make demands as a condition for joining any coalition (or similar arrangement). For example, Canada's NDP only agreed to back the Liberal Government on condition of state funded dental care for children being implemented. Now it is. Millions are affected. Whether you agree with it or not - that is unquestionably a "meaningfully different outcome". Other examples abound if you care to look.

>I'm not aware of any parliamentary system with a wonderful diversity of thought and a long record of positive accomplishments.

That's an impossibly high bar. The standard we're talking about is whether it's better than a two party system.

> It absolutely produces different outcomes than a two party system. Smaller parties make demands as a condition for joining any coalition (or similar arrangement). For example, Canada's NDP only agreed to back the Liberal Government on condition of state funded dental care for children being implemented. Now it is. Millions are affected. Whether you agree with it or not - that is unquestionably a "meaningfully different outcome". Other examples abound if you care to look.

You think logrolling doesn't happen in the American system? Parties are made up of factions who will entertain each other's preferred priorities, which is the exact same thing you're describing except that the parties of the various factions are nominally the same. There's no real natural or obvious philosophical reason why your position on gun ownership should imply a position on environmental regulation, religion, infrastructure buildings, racial politics, abortion, tax policy, and more. We're so used to the groupings that they seem natural but if we go back and looking at older American party systems you'll see parties that don't 100% map onto the contemporary ones, with a blend of some elements we would think of as fitting and others we wouldn't.

I never said it didn't happen. Don't do that.

It's unquestionably a different dynamic. If the party is in power, then the party is in power. Period. They have less incentive to listen to smaller factions. Do they? I sure hope so. But it's not the same.

Furthermore, these internal deals are more likely to be kept private since theres no benefit to airing everything to the press, the public and the opposition.

In a multi-party system almost all those deals are public. Thus, voters can decide if everyone lived up to their promise.

That's just not true. If you want a recent example go look at the protracted battle to elect a Speaker of the House.

In some ways it might be worse. Here we had a party that promised to vote with one faction and partly thanks to it survived the election. A few months later they did a 180 and voted the other faction into government. Opinions aside, imagine voting for Sanders only for him to elect Trump or the other way around. With that said, two party version also comes with major flaws.

IIRC Some states you cannot vote outside your declared party (Washington)

That's only for primaries, which traditionally were only open to members of a party.

It's just different order of operations on coalition building. Other systems divide into a majority and opposition at some point. In the US it just happens earlier, but there is the same diversity of opinions within those groups.

Majority? Probably. Vast majority? No way. If that were true we'd not keep switching parties between presidents. And Florida, a mostly independent voter state, wouldn't have had DeSantis win in a landslide when he just barely tied last time around.

That's attributable to turnout.

You tell me if 90% is a "vast" majority or not. https://www.pewresearch.org/politics/2019/03/14/political-in...

> I thought it was interesting how it explicitly doesn't boost independents. So much of the two-party system is self-reinforcing.

Is it boosting? Others are claiming this code is just for metrics collection: https://news.ycombinator.com/item?id=35391896.

But on the topic of Democrats vs. Republican vs. independent; a big factor may be that "Democrat" and "Republican" are much more cohesive groups and therefore much easier to define. No one can honestly define "independent" except in a kind of "none of the above" sense, since they can range anywhere from extreme right, to the center, to the extreme left.

Why measure something if you don’t intend to change it?

So you can make sure not to accidentally change it.

And if one party trends more towards violent extremism, they get boosted the same as before?

What's that got to do with accidental effects of unrelated software changes?

if your hate speech filter hits one party harder than the other, that doesn't necessarily mean it is broken

The point is to have these statistics remain the same for the individual groups in between changes, not to measure them against each other.

So then why make any changes?

Why not? If it doesn't matter but it's controversial there's no reason not to change it.

What if you're changing the spam filter?

There are many reasons to measure things if you don't intend to change them.

You might want to ask that of, say, astronomers, paleontologists, or even simply journalists.

Observing and measuring leads to understanding. As others here have noted, sometimes you want to measure to ensure that you're not inadvertently affecting an outcome or phenomenon.

> You might want to ask that of, say, astronomers, paleontologists, or even simply journalists.

You compare scientists with businesses. One of them's job and passion is to collect knowledge for the sake of knowledge. For the other one it would be cost without gain and eliminated if they didn't do anything with it!

No idea what journalists are doing in that list, what do they measure? If they want something measured they'd ask or look at other's services. Unless you mean the business that the journalists work for, but hen it's just that, a business.

I gave examples based on well-understood instances for which the reader is presumed to have the capacity to draw inferences to business-related concepts.

Businesses also rely on astronomy and geology, in instances. The former is used for navigation (though far less than in the past, I'll grant), and there are certain extractive sectors with interests in geology. Risk-management as well (insurance and catastrophic risk, whether from landslides, earthquakes, volcanos, tsunami, or other phenomena).

You'll also find businesses keeping tabs on weather, climate, competition (competitive intelligence), demographic data, politics, legal cases, laws, social and cultural movements, etc., etc., etc., which in many cases they have comparatively little capacity to change directly.

You're also jumping late into a thread which has already given numerous other rationales for why such activities might be undertaken.

Sometimes approaching a discussion from the PoV of seeing what you can learn from it rather than automatically adopting a presumptive stance of opposition or disagreement affords benefits. I recommend it strongly.

Depends what the metrics are used for. It doesn't make sense to apply artificial boosts to metrics that are only used for internal accounting. Well, maybe if you have an egotist CEO, but that wouldn't explain the rest of the boosts. We have to assume this code has some sort of effect somewhere.

Metrics collection so they can avoid bias, allegedly

There are probably many other non-political alternatives to select 2 groups to test for bias

Also how about.... everyone else in the world who is not an American voter?

Only anecdotally.

But ever since Musk took over the amount of US political content has significantly increased in particular from the right despite me not living in the US.

It's hard to tell whether previously political content was weighted less and Musk has removed those controls or whether they are now weighted higher.

Some of the power users Musk reportedly had boosted are specifically right wing political posters, like catturd2. But then he's also boosted some high profile left leaning politicians, so it's not exclusive. It does mean you're more likely to see right wing American political content either way, which has to be annoying for people outside of the US.

https://gizmodo.com/twitter-algorithm-aoc-ben-shapiro-cattur...

Would it not be just as annoying to see a slant the other direction, or in any extreme direction?

The parent poster doesn't realize the inherent bias in what they wrote and that you have pointed out.

Boosting both right and left wing content increases your chance to see either type of content, because you boosted them over everything else.

I'm not sure why you wrote that to me. It was already in your prior comment.

I believe their point is that, because the US is generally regarded as 'to the right of' many countries in Europe, for example, a slight rightwing bias looks to us (I'm in the UK) like an extreme rightwing bias. Heck, even a completely neutral stance, as far as the US is concerned, will look like it veers to the right for us. Note that I'm not saying anything about how I believe the twitter algorithm should work, this is just my interpretation of what that commenter meant.

Europe is only a tiny fraction of the world, though.

Compared to India, Japan, Singapore and (depending on the metric) China, the US looks clearly left of center. Compared to most of the Middle East, the US looks extremely left-wing.

Why do you believe it would be annoying to us?

I live in the US in "Trump country" and don't see any political content. Maybe its just related to the people you follow and the content you look at?

This is what it means to let another country own your social media. Their ideas and memes unconsciously get preferential treatment. This is maybe not a good thing, but it is what it is.

> Their ideas and memes unconsciously get preferential treatment.

I think as this repo shows it's conciously, rather than unconsciously, getting preferential treatment.

It’s omission. They built for Americans because they were Americans. No one building this said “let’s ignore Canadian politics” or any other country, they just didn’t think about them at all, because like most Americans they don’t really care about the insidious Québécois plots to annex Prince Edward Island or whatever actual issues are happening in Canada.

The freedom convoy and it's repression with suspension of constitutional rights (because that's a thing over there?) and bank accounts arbitrarily frozen did get a little bit of attention in the American medias.

> the insidious Québécois plots to annex Prince Edward Island

I know it's a joke, but the people of Prince Edward Island would most likely welcome it. Economically that would be a huge boost. Not sure Québec would enjoy it however.

Perhaps they should look at annexing Labrador?

Bank accounts were not "arbitrarily frozen". This is total nonsense.

https://nationalpost.com/news/canada/banks-were-only-asked-t...

Not arbitrarily: the freezing was targeted against political protestors.

They were going pretty damn far beyond "protesting" they were trespassing/loitering/squatting.

Loitering, dear god!

When done in giant trucks blocking traffic and making tons of noise and exhaust fumes for weeks it is a crime.

That sounds straight out of Russia or China.

Your article seems to support the idea that bank accounts were arbitrarily frozen.

> The freedom convoy and it's repression with suspension of constitutional rights

This was needed because the convoy of morons was seriously damaging people's quality of life and the local police did not do their job. The police in other towns DID do their job and blocked the trucks. You should be mad at the cops who just sat by and watched the convoy of morons roll in with their thumbs up their butt.

What exactly is allowed in a protest is quite a thorny issue.

Here in the UK there are environmental protestors gluing themselves to the streets and I have quite mixed feelings about them.

A protest that does not inconvenience anyone is very easy to ignore.

You can see here how wild and inconvenient these protests were. The bouncy castle and hot tub were a big public nuisance!

[0] https://www.youtube.com/watch?v=P74pbbhCZng

[1] https://www.youtube.com/shorts/hzfwaSVIvoA

The constantly idling giant trucks very much were a public nuisance. YOU wouldn't want to live near them.

Indeed their acts of terrorism were not reported by international media.

For example, during the occupation of residential areas of the capital they created deafening noise around the clock that was measured indoors (windows closed) at levels causing permanent hearing damage after minutes of exposure, with police doing nothing to stop them.

https://news.ycombinator.com/item?id=30434168

People can always use their countries' social media (and I assume many do!).

But it's indeed an interesting fact that people seem to specifically seek American tech and social media. And honestly, there's no shortages of foreign nationals commenting on American politics (and it's a good thing, it's their right thanks to the first amendment!).

This is a joke, right?

The vast vast vast vast majority of people use “American” social network because they are the social networks that exist. The US is undeniably the main exporter of SaaS products. See: the incumbents freaking out that China is getting a turn in the front seat. I’ve never once heard of anyone seeking ‘American tech’, except for some punchline in an anti-Soviet joke or movie.

Foreign nationals can comment about American politics because the US doesn’t have jurisdiction over their speech, notwithstanding the back and forth over whether social media companies are liable for disseminating such content in the first place.

A material part of why the USA is seen as The Country in Western culture, and a noted big player in other cultures, is because of the power it projects via the media. That includes both Hollywood exports, and social media.

American social media companies have the benefit of a large homogenous deeply interconnected culture with a vast ad buying community. Simply put, a U.S. social media platform can get more revenue with less effort than a European counterpart (because they have to go country by country, language community by language community), and will generally outcompete those EU counterparts, with social network reinforcement effects doing the rest. This holds up for most ad-supported online services, which together with a much stronger private investment sector is the reason American platforms absolutely dominate Europe. The EU has a stronger government subsidy system for software but that system does not reward or expect market success, which is why it delivers very little value.

The importance of US cultural power cannot be overstated.

> I’ve never once heard of anyone seeking ‘American tech’

One reason people have a distorted idea of what physical products the US produces, so I've read, is that most of them are products for industrial use, in factories and so on.

You look at consumer products that are made in..., because that's where the finishing touches are put on. That's not representative of the global economy though.

> I’ve never once heard of anyone seeking ‘American tech’, except for some punchline in an anti-Soviet joke or movie.

Really? Why is everyone using MacBooks and iPhones? Actually most Soviet computers were copies of western designs. [0]

> is because of the power it projects via the media. That includes both Hollywood exports, and social media.

The thing is, is takes to to project, someone has to export a product and someone else has to import it. Since people seek it and want to consume it it's easy to export.

[0] https://en.wikipedia.org/wiki/List_of_Soviet_computer_system...

Before I deliberately locked myself out of it (well before Musk), I asked for my data.

They classify me as:

  * speaks Indonesian
  
  Interested in:
  * Beer
  * Cricket
  * DJs
  * Dance
  * Enterprise software
  * Horror
  * NFL football
  * South America

  And aged either between 13-54 or (and?) over 65

Other than the age (I'm neither under 13 nor between 55-64), everything I've listed is incorrect.

On that basis, they'd probably call me a Republican.

Carries a different meaning when you're British, that name does.

I'm a guy in my 30s but they had me classified as a 65+ year old woman. They also thought I was a radiologist.

What a strange age classification — no idea about what your age actually is, just that it definitely isn't 55-64. I assume it would never conclude you are under 13 for legal reasons.

Aren't "enterprise software" and "horror" the same thing?

Fair.

I try to avoid both, when I can. :)

This is an interesting point... they're making sure their A/B experiments don't adversely affect Republicans or Democrats. But lots of European countries are skewed more to the left and would broadly look more similar to the "Democrats" group.

Isn't it intrinsically self-reinforcing, if you have a winner takes all system? It's almost always better to join an existing team than start a new one.

I thought it was interesting that twitter thinks the U.S is the world

For a big tech company, the regulatory regime is the world. For Twitter that's the US.

Most big companies I've worked at are international and are regulated in many regimes, although sure the one with the main corporate headquarters is more important.

Twitter thinks Twitter is the world.

I'd be interested in which bin people like Joe Manchin, Kyrsten Sinema, Susan Collins, et. al. are placed.

So many questions. How are users tagged D or R? Is that a manual process or automated somehow? What is the effect of these tags? Can I find out if my Twitter account is in one of those buckets?

Especially for people that aren't... you know.. Americans.

Unless they mean actual public figure party members which are known and probably verified.

Then it probably means which side you'd be on if you were American.

you could probably algorithmically determine it in most cases based on any number of indicators from phrases used, to communities interacted with, which hashtages are included, which cohort retweets and likes most etc... thats not even getting into simply tagging political figures with the party they officially affiliate themselves with

And how are they choosing to balance them, per capita, or just both sides should get 50%? It seems pretty clear they are making editorial decisions here. Does that break their section 230 protections?

"Section 230 protections" have nothing to do with "editorial decisions"

Please read something. Preferably section 230, which is short. Alternatively something entirely different, as long as that keeps you occupied.

>It seems pretty clear they are making editorial decisions here. Does that break their section 230 protections?

No.

https://www.techdirt.com/2020/06/23/hello-youve-been-referre...

Likely they are tracking performance verified politician accounts based on registered party affiliation. Why republican should count equal in the evaluation metric to democrat when nunerically there are less republican voters, let alone proportions on Twitter, is another question.

I suspect that these are used for metrics tracking rather than being fed back into the recommendation engine. But there's no real way to know for sure given the limited release. These predicates aren't actually used anywhere in the code that's been made available.

It's not that shocking...

Half the people that got promoted on my timeline were perpetually candidates for elections I couldn't vote in, and they _self-identified_ as Republican or Democrat in their own bios, or via the registration of their candidacy...

This is why I exclusively used to use Twitter in the "people I follow only" mode, and simply shut my account down when they pushed harder on the algorithm.

Facebook guesses your political affiliation as well, you can even look über your settings to see what they guessed.

The repo suggests it's about tracking engagement metrics[0], so Team Red people see more Team Red content and vice versa. Nothing nefarious.

[0]https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92...

Having a separate flag to track engagement specifically for Elon tweets isn't nefarious, but weird.

You say it’s not nefarious but isn’t that how echo chambers are created?

I don't believe echo chambers are nefarious - there's no hidden agenda involved with them. That's just how recommendation algorithms work, and it's what most people want.

But if someone finds some code that suppresses recommendations from a specific political ideology across the board, that would be nefarious, IMO.

Echo chambers may not be nefarious, but they are insidious.

They can be if they're involuntary and inescapable, but neither is the case for Twitter. It's designed around letting you curate your own feed, but it also constantly throws random stuff in through retweets and quote tweets - which is what people hate the most about the platform.

The echo chamber in your first point does the suppression in the second.

Why specifically track political parties? Where is author_is_american? Or author_is_mayonnaise_enjoyer?

Maybe it was a choice made many years ago that they thought was appropriate, but we can't yet know it's not used for other purposes. We can at least be reasonably sure they've added the author_is_elon within the past year. I would have thought there would be many more descriptors, or non-controversial descriptors.

Or maybe Elon specifically added those before releasing this code to get people riled up.

Twitter became a very politically charged platform after 2016, mainly among Americans. They'd be idiots not to take advantage of that.

The point is probably to check that changes they make aren’t accidentally politically biased.

Exactly. I don't see NPA (no party affiliation) anywhere.

Did you not get the memo? "If you're not with us, you're against us".

It's probably more of a conservative/liberal identifier based on US political party ideals... And they likely would filter any metrics from this by the users country

Reply on news.ycombinator.com