Hacker News Reader

Comment by toqy | Hacker News Reader

This is begging for a console.table

     console.table([...document.querySelectorAll('.fatitem table .athing')].map(el => [el.textContent.trim(), el.nextSibling.textContent.trim()]).sort(([,a], [,b]) => parseInt(b) - parseInt(a)))

View on HN

Current top 10:

  'US: California' '79 points'
  'Canada' '48 points'
  'US: Washington' '39 points'
  'US: New York' '36 points'
  'Germany' '35 points'
  'United Kingdom' '34 points'
  'US: Massachusetts' '19 points'
  'Netherlands' '18 points'
  'US: Colorado' '17 points'
  'Sweden' '16 points'

Colorado #9? Really?

Denver and Boulder have a good amount of tech. A bit surprised it comes out ahead of Texas though.

Also the UK and Germany being about the same is interesting. Germany does have a slightly larger population, but of course the UK is natively English speaking.

Texas is without power at the moment.

Thought you were joking at first, how can an entire state be without power? But seems you're right, how the hell does that happen in a first-world country?

Tried to access some dashboards that are supposed to show some data, but not sure if I can't access it because I'm not in the US/Texas or if it's down as well (because they don't have power?)

- https://www.ercot.com/gridmktinfo/dashboards

- https://www.ercot.com/gridmktinfo/dashboards/gridconditions

- https://www.ercot.com/gridinfo

Says "This request was blocked by the security rules" so I'm assuming it's the former.

https://www.ercot.com/gridmktinfo/dashboards/gridconditions

Is showing green for me, so some people might be without power but most of the state is fine.

ERCOT doesn't do power delivery, you wouldn't see how many people are without power by checking out ERCOT. You would want to look at one of the several power delivery companies, such as Oncor.

https://stormcenter.oncor.com/

They show 7,073 customers affected by outages out of 3,835,901 customers served. That's 0.18% of customers affected.

Texas-New Mexico Power is another power delivery company in Texas.

https://www.tnmp.com/power-outage-map

They are showing 1,195 customers affected. They serve ~400k customers. So ~0.3% of customers are without power for that company.

Austin Energy is another power company in Texas, they serve pretty much the entire city of Austin so ~1M customers. They are reporting 24 affected customers. So 0.0024% of customers affected.

https://outagemap.austinenergy.com/external/default.html

The above poster is not right, and neither are you for suggesting some massive amount of the state is without power.

Thank you, that oncor website seems to be available to me. I was browsing around some news article that seemed to echo the same sentiment, tried to access the source data myself and couldn't, so made an assumption. That was an error on my part and I'm sorry for that. Thank you for providing a better source that seems accessible worldwide.

> so made an assumption

You shared lies and slandered a state suggesting its a third world country based on false data.

To be fair, our power grid does have a tendency to fail regularly across the US. Texas last year, California's rolling blackouts, NY blackouts, etc. There's a reason that upper-middle-class homeowners have been buying Generac generators en masse.

Power grids tend to fail just about everywhere that experiences extreme weather. Its not necessarily a US-only thing. Its just that the US experiences a pretty large share of extreme weather compared to most of the developed world. Far more tornadoes, hurricanes, wildfires, etc. affect the US than say, Europe.

California's rolling blackouts were not the result of any particularly extreme (for California!) weather. Similarly, hurricanes are not new to most of the hurricane-afflicted parts of the country. Wildfires have been a force in the US for a long time, and the West has experienced seasonal wildfires for decades.

The difference is the US doesn't generally build or maintain infrastructure in a fashion commensurate with the extant risks here (which are different than those of e.g. Germany). Obviously, it is possible to keep a grid running when it is extremely cold outside, see the Nordics and Russia, for example. The US just makes systematic choices not to build & maintain to that standard.

What kind of similar weather risks do the Nordic countries or Germany experience that would be an equivalent to the tornadoes, hurricanes, and wildfires the US experiences on a very regular basis? I truly don't know what kind of similar risks they face and would like to rectify my ignorance.

But seems you're right, how the hell does that happen in a first-world country?

An interesting note is that Texas has its own power grid, more or less, and it isn't federally regulated.

more info: https://en.wikipedia.org/wiki/Texas_Interconnection https://slate.com/news-and-politics/2003/08/why-texas-has-it...

Less than 1% of people in Texas lost power from this storm, and most who did lose power lost power for only a few hours tops. More people lost power in Ohio and Tennessee from this storm than people in Texas.

Only two people in my network of people I know lost power in Texas, one of which was a coworker's neighbor because said coworker's tree dropped a branch on the neighbor's overhead power delivery line.

Actually our grid has held up pretty well this time around.

It also didn't get nearly as cold as last year, and not for as long. Plus, the providers ramped up production the first evening to the point where wholesale rates were negative to incentivize more usage.

Another week+ of 0 degrees would likely have not been as smooth...

My parents house in California had no electricity for over a week (Christmas through New Year's, and it was sporadic until about Jan 8). I think the state is still doing fine as a tech hub.

I think the point is that if a large portion of Texas currently didn't have power (which seems to not be the case, but whatever), then they might be under-represented in this poll because I imagine fewer people in place without power would be browsing HN than usual.

It's funny how the UK and Germany are growing with the same rate. I checked 10 minutes later and they were 56/57, then I did it again few minutes later and they were 67/68.

As I type this Texas is at 35 votes; looks like 5th behind California, Canada, Germany, and the UK

All the relevant people in Germany as well the rest of the continent speak English pretty well.

I think that’s overselling it slightly. Yeah most techies in Germany are proficient or better in English, but that’s still gonna lead to less English website usage than a natively English speaking population.

Well, EastEnders was on...

There is an admittedly surprising amount of tech work here (CO). I moved here from Chicago nearly a decade ago for a startup job (not telecommunications, ad tech or government related), and have found that the marketplace for tech jobs has grown heavily since then.

Also, given the new(ish) move to remote work that many firms nationally have begun to embrace there is also a strong showing here of people that work in Colorado, but for firms located elsewhere. This is my personal current situation, and I work with a guy who is moving here from the midwest in a couple months to work 100% remote. It's more and more common.

I know a programmer who moved to Colorado, mainly because he likes to be able to go out on hikes and that sort of thing.

I lived in Boulder for 2 years, and it's not just that there are gorgeous places to explore in CO - there's a culture of adventure and exploration here; everyone is much more likely to finish work early and go hiking, or take a snowboarding break in the middle of the day. It's encouraging and awesome.

As percent of total responses, as of right now:

    'US: California', '10.87%'
    'Canada', '5.79%'
    'United Kingdom', '4.51%'
    'US: Washington', '4.45%'
    'Germany', '4.45%'
    'US: New York', '4.32%'
    'US: Colorado', '2.10%'
    'US: Texas', '1.97%'
    'US: Massachusetts', '1.84%'
    'France', '1.78%'

The US in total makes up 49% of responses so far (even though it is nearly 10pm in most of Europe)

Edit: quick and dirty one-liner to get the full table with relative values:

    data = [...document.querySelectorAll('.fatitem table .athing')].map(el => [el.textContent.trim(), el.nextSibling.textContent.trim()]).sort(([,a], [,b]) => parseInt(b) - parseInt(a)); total = data.map(a => parseInt(a[1])-1).reduce((a,b) => a+b, 0); console.table(data.map(a => [a[0], ((parseInt(a[1])-1)/total*100).toFixed(2) + "%"])); us = data.filter(a => a[0].startsWith("US:")).map(a => parseInt(a[1])-1).reduce((a,b) => a+b, 0); `US makes up ${(us/total*100).toFixed(2)}% of ${total} responses`

Edit2: made one-liner even dirtier to account for counts starting at 1

It would be cool to see it relative to the population of the country.

Be mindful that the dataset is 1-indexed (i.e. zero votes equal a score of 1), so your script is probably a bit skewed towards non-US.

Good catch. Accounting for that brings the US up from currently 50.2% to 52.8%. It also brings California up from 11.75% to 12.86%. I updated my code above to account for that (having to fight the urge to refactor it to be more sensible)

Is there are reasonable way to do this on a mobile browser?

Maybe rewriting it as a bookmarklet that displays the results on the page?

Anyways, here's the current result in a pastebin: https://pastebin.com/fR89qMKk

Good tip, thank you. Bookmarklets work as normal(?) on Firefox on Android, and can be invoked via the address bar on Chrome:

https://paul.kinlan.me/use-bookmarklets-on-chrome-on-android...

A random intro to bookmarklets: https://betterprogramming.pub/are-bookmarklets-still-useful-...

I posted these:

https://news.ycombinator.com/item?id=30213991

https://news.ycombinator.com/item?id=30215244

Edit (this version subtracts 1 to see the real score and pops open in a new window):

  javascript:open().document.write('<meta name=viewport content="width=device-width, initial-scale=1.0"><table>' + [...document.querySelectorAll('.fatitem table .athing')].map(el => ({place: el.textContent.trim(), score: parseInt(el.nextSibling.textContent) - 1})).sort((a, b) => b.score - a.score || a.place.localeCompare(b.place)).map(e => `<tr><td>${e.place}<td align=right>${e.score}`).join('') + '</table>')

TIL about console.table. Thanks for sharing!

Can't we just aggregate all of the US or worth splitting out cities/states/provinces for other countries also.. Think it makes the point of even this forum being US centric

This forum is run by a US company and was originally made for their US portfolio.

Large US states are more populous than most countries, and the US tech industry is orders of magnitude larger than most countries'.

If US weren't broken out, it would be a huge outlier at the top and wouldn't be nearly as interesting.

I don't know of any other country that would make the chart that much more interesting by being broken up.

Google and Amazon have a pretty big showing in Denver/Boulder area and Apple's expanding an engineering office here.

Denver's also a big telecommunications and defense hub. Both industries employ armies of developers

More efficient, uses integers, and sortable in the table:

   console.table([...document.querySelectorAll('.fatitem table .athing')].map(el => ({place: el.textContent.trim(), score: parseInt(el.nextSibling.textContent)})).sort((a, b) => b.score - a.score))

with tie breakers

     console.table([...document.querySelectorAll('.fatitem table .athing')].map(el => ({place: el.textContent.trim(), score: parseInt(el.nextSibling.textContent)})).sort((a, b) => b.score - a.score || a.place.localeCompare(b.place)))

If you're on a mobile browser, paste this into the address bar:

    javascript:document.write('<table>' + [...document.querySelectorAll('.fatitem table .athing')].map(el => ({place: el.textContent.trim(), score: parseInt(el.nextSibling.textContent)})).sort((a, b) => b.score - a.score || a.place.localeCompare(b.place)).map(e => `<tr><td>${e.place}<td align=right>${e.score}`).join('') + '</table>')

That doesn't seem to work on Firefox for Android, sadly

Sometimes the browser will remove the leading 'javascript'. In that case, type it back in.

It doesn't seem to be supported. Ironic that the bookmarklet was invented by them.

It works on Chrome for Android.

I wanted a total for the US:

    console.table([...document.querySelectorAll('.fatitem table .athing')].map(el => [el.textContent.replace(/:.*/,'').trim(), parseInt(el.nextSibling.textContent.trim())]).reduce((d, [k,v]) => {d[k]=v + (d[k] || 0); return d}, {}))

I can't figure out how to sort that, though...

  Object.entries

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Refe...

Oh, I need to type Object.entries(x) instead of x.entries()

Makes sense, otherwise entries would contain itself.

Thank you!

Entries only iterates enumerable items (unless you make something evil). None of the native types with entries in their prototype iterate over that method

I created a page with the current results that includes a total for the US:

https://nate.org/hacker-news-location-poll

Percentage next to totals would make this perfect

Good idea—I just added a percentage column, as well as a row at the bottom for the total count

Haha, great minds think alike.

Are you accepting PRs? If so, align=right. Thanks!

To see the sorted results: https://gabrielsroka.github.io/hnpoll.html?id=30210378

Source: https://github.com/gabrielsroka/gabrielsroka.github.io/blob/...

Just click on 'Value' in console to sort by value

To see the sorted results: https://gabrielsroka.github.io/hnpoll.html?id=30210378

Source: https://github.com/gabrielsroka/gabrielsroka.github.io/blob/...

Neat!

Noob question: is there a similar way to dump it to JSON, instead of a table?

Just replace console.table with JSON.stringify for the quick-and-easy dump of exactly the same data as is in the table...

JSON.stringify([...document.querySelectorAll('.fatitem table .athing')].map(el => [el.textContent.trim(), el.nextSibling.textContent.trim()]).sort(([,a], [,b]) => parseInt(b) - parseInt(a)))

Replace `console.table` with `JSON.stringify`:

JSON.stringify([...document.querySelectorAll('.fatitem table .athing')].map(el => [el.textContent.trim(), el.nextSibling.textContent.trim()]).sort(([,a], [,b]) => parseInt(b) - parseInt(a)))

this is fricking awesome!! thanks!!

TIL. Cool!

Good call!

Reply on news.ycombinator.com