https://news.ycombinator.com/item?id=21942143
IMO, OP’s predictions may be safe, but that says more about reality than OP’s lack of imagination.
Otherwise you will get what we see in TV everyday. a bunch of people making outrageous predictions just to sound exciting. Then when the prediction comes true, they will claim all the credit. when it doesn’t, they will say “haha I was just kidding of course”
Any serious prediction should include a probability estimate that differs from conventional probability estimates. Almost not one actually does this, because everyone is doing it wrong.
Sure they are, it's a nice counterpoint to the people who think in 10 years we'll have flying cars and a fully immersive VR matrix and nobody will have a real job since AI and robots will do everything for us.
The AI not leading to job displacement prediction is just silly, given that it's already leading to job loss.
The prediction was about job loss, not displacement, and you are incorrect about the current state. Job displacement and job loss are two very different things. AI is leading to job displacement (less jobs in certain industries, but not necessarily overall reduction in jobs), it does not seem to be leading, so far, to job loss (less jobs overall). It's harder to tell if it is leading to a greater share of jobs with lower wages, which has similar economic and policy implications to it causing net job loss, because there are lots of confounding factors when looking at the dynamics of wage distribution.
A lot of AI fear is based around net job loss (which has not generally historically been associated with technological advancements), not mere job displacement (which is a common effect of technological advancements).
Per-capita CO emissions [1]:
United States 17.5
Australia 16.75
Canada 14.67
Russian Fed. 12.18
South Korea 11.78
Norway 11.71
Finland 11.53
Greenland 11.07
Netherlands 10.96
...
China 6.18
Mexico 3.91
Brazil 2.15
India 1.64
[1] https://cotap.org/per-capita-carbon-co2-emissions-by-country...
Qatar 40.1
Trinidad & Tobago 37.78
Kuwait 34.24
Netherlands Antilles 23.55
Brunei Darussalam 22.96
United Arab Emirates 22.31
Aruba 21.59
Luxembourg 21.34
...
Here is the list of top CO2 emitting countries:
China 10,877.2
United States 5,107.4
India 2,454.8
Russia 1,764.9
Japan 1,320.8
Germany 796.5
South Korea 673.3
Iran 671.5
...
If you think in terms of net change relative to a given level of emissions at some year... eu bad, US badder, China & India Terrible.
If you think in terms of "every person is entitled to X emissions per year," where X is current or desired average emissions (inc 0-emissions) then... China & India are best, EU & US are bad.
The more nationalistically we think of these things, the more nationalistic they seem and become.
That's quite outdated; there's a huge internal market in China, India, etc, and plenty of trade between the non-"1st world" countries. In fact, the US did have an initiative to reduce consumption from there (in the form of tariffs) and it didn't impact their economy much.
Just as an example, China now has about as many cars as the US. And the market there keeps growing, whereas the US has mostly plateaued.
If "1st world" were truly worried, they would stop buying or stop off-shoring. Saying "3rd world lacks initiative" as if it's a national issue is a joke.
Yeah, I definitely said that. Continue being mad at me and America.
The Node ecosystem is a mess and it starts with core being a hodgepodge of deprecated mush. I think it's more likely that wasm picks up steam.
Rails, PHP, Django, etc aren't going anywhere.
The former will keep going and keep solving problems, but it'll be relatively quiet in the grand scheme of things just like it is now. The JS ecosystem will continue to thrive in the latter world where there's more VC money, hype & RAM than common sense, so we'll keep hearing about it. As long as there's enough VC money to fuel the fire this situation will keep going (along with NPM's storage expenses growing alarmingly) but the second the VC money dries up and hype-driven startups die off, companies will mostly come back to the first approach and the majority of "React developers" will be out of a job.
The secret, or perhaps not-so-secret, is that the vast majority of web applications are still CRUD, and the average person doesn't understand the difference between a SPA and a server-rendered app. Most of the engineering effort that goes beyond that is probably a waste of time unless you really plan on doing things beyond CRUD. So the best thing is to actually understand the fundamentals, learn to do a good job at those, and use the tools that make sense for said job. That means that maybe, just maybe, your WebGL game doesn't need to use React.
But comparing Rails (a framework for complete web dev functionality) to entirely different paradigms and use cases seems kind of silly. These frameworks don’t solve everything, but they make it far easier and cheaper to develop a website with under, say, 500 million users. Once you get that large you can start exploring paying more devs for better efficiency.
Ya let's blame it all on other people.
The Americans loudly refuse to do anything, loudly deny that there's a problem, and loudly proclaim their hatred of ecology.
The Chinese say yes to everything but they have absolutely no intention of actually doing anything, they just lie.
Check this out : https://www.telegraph.co.uk/news/2019/05/23/china-factories-...
https://www.theguardian.com/world/2015/nov/04/china-underrep...
> This means that according to our assessment, China will meet its 2020 pledge and its NDC targets, but still be above current emissions levels. China’s chief climate negotiator, Xie Zhenhua, has expressed the opinion that China could meet its 2030 peaking targets early, reflecting the conclusions of other studies (Green and Stern, 2017; Xu, Stanway and Daly, 2018; Gallagher et al., 2019b; Wang et al., 2019). CAT analysis confirms that, based on current GDP projections, China is indeed likely to meet the carbon intensity target of its 2030 pledge early.
So the TL;DR is China is likely to meet its pledges but the pledges themselves are pretty weak — rated “highly insufficient” and only consistent with a 4 degrees warming limit.
That’s inconsistent with your “say yes to everything but ... just lie” assessment. It appears they don’t say yes to everything (hence the weak pledge) but will hit the promised target.
Your links don’t support your point either. The first link doesn’t pinpoint the cause (could be a lapse in enforcement) and the second link is about a revision in official figures; Guardian’s reporting is based on a NYT article with a more neutral tone, but even the Guardian article with its loaded title (surprise surprise) didn’t call it lying.
But with current developments this will be far too little too late and it will be a rough ride forward.
Unfortunately we can't blame it on ourselves as our carbon emissions, assuming it's even relevant, have decreased massively compared to predictions.
United States per-capita CO2 emmisions are 17.5 tons per year vs China's 6.18, and that's not even taking into account the fact that China emits a lot of CO2 producing cheap products to sell to the US.
https://cotap.org/per-capita-carbon-co2-emissions-by-country...
I’m not trying to absolve the US here just noting that rate of increase/decrease matters in addition to per capita and absolute values
https://www.livescience.com/49196-nasa-satellite-oco2-carbon...
China shows up very red, but they have 5x the people living in a much smaller area.
I'm not sure why you'd expect it to be otherwise. This is a per-capita figure, and carbon emissions are strongly correlated with income and purchasing power. CO2 emissions aren't driven by consumer goods production alone.
There's an adjusted metric that puts things into an even better perspective: https://ourworldindata.org/co2-and-other-greenhouse-gas-emis... .
The reality is all these data sources will he estimates. Estimates can be severely wrong — they can be produced by local governments with incentives to lie, or estimated by outside observers with poor actual knowledge. China has had a long history of empirical lying in their economic data locally due to internal party targets, for example. I see no reason for them to give accurate numbers to the UN/external world, where as with the US there’s enough decentralization and public data that there’s a better chance of being more accurate (but still probably quite off).
There is less data available for China and therefore the estimates are likely wrong by a higher margin than the estimates for the US. But the figures do very much fit the demographics, economy and geography of China so I doubt that the margin is that high.
Wait... https://co2.jpl.nasa.gov/
When azinman2 claimed the reported COTAP numbers for China are probably wrong I was saying the same is true for the US for exactly the same reasons - measuring emissions on a national scale is hard, there are good reasons not to believe any self-reported data that business relies on, and frankly America has worse climate emissions policies than China at the moment.
I'm not suggesting that other agencies that independently measure emissions are wrong. It is possible to measure emissions remotely (eg the satellite that SmellyGeekBoy linked to), and that provides proper unbiased data.
Does this mean we need to multiply China's values by 4.2 (1386 mln Chinese vs 327 mln Americans) to account for all those capitas? ;-)
[Edit] Also, I could see being offline becoming a kind of cultural prestige or aspiration, like something only rich people could do by delegating their online presence to hired guns, etc.
This reminds me of the 1970s, when being 'ex-directory' - paying to not have your phone number included in the telephone directory - was seen as cultural prestige.
All it takes is a series of stories about offliners from a couple of major news outlets.
I suspect WASM will lead to a lot less stuff being done in JS.
I agree with your intuition, though -- I don't think it'll become too widespread. Projects will start in all-js in the front-end, and no language will provide sufficient benefit to motivate rewriting half of the client.
I'm less sure about back-end development trends but I think "less standardised than the front-end" is a gimme. Maybe Node/Express or similar will pick up, but I feel like they've been waning the last few years. Go servers seem trendier.
> 6. But a small yet growing culture of "offliners" becomes
> mainstream. Being offline is the new "Yoga" and allows
> bragging rights.
> 8. Web development matures and a "standard" stack is accepted,
> all in JS.
I'd walk into traffic.
HN for example gets by just fine with minimal JS. JS is not even required for functional websites.
Amazon, unironically. I did an entire transaction the other day with noscript on before I realized it. I'd never want to work for them but their tech is impressive.
Only thing that didn't work was the logout button.
JavaScript today is much more usable than what came out in 1995. You still can't use the Set type to usefully contain anything but primitives, but there's a Set type!
Based on something like the Gartner Hype Cycle I will guess that we might be nearing something in 10 year. Or it might be that these things work in a 30-40 year cycle (based on observations of the last time AI was hot).
8:
For better (we can use languages that doesn't have more built in footguns than PHP) and for worse (openness of frontend will disappear) WASM will make Javascript unnecessary in the long run.
I'm kind of seasoned with JS - and a number of other languages - and the only real strength that JS has compared to TypeScript, C# and Java, language wise, is that it is ubiquitous.
Besides: if you get developers tp agree on a single stack in 10 years then you are dangerously good at negotiations and should spend your time in embassies not on HN;-)
This! But my take is it will be out of sheer necessity.
Killing all my social media presence for the last 3 years or so has been more cathartic than anything else I tried in order to acquire perfect peace. That, plus completely cutting out alcohol. I have near zero anxiety since embarking on this and I highly recommend it.
... Then again, you could just go knock on your neighbor's door and say hello, but that's too easy, too natural. ;)
Tinder for friendly neighbors /s
Jokes aside, it does sound good though. I wouldn't mind getting my hands dirty for a few hours on the weekend. As far as mobile apps go, my only competence is in react native. If interested let me know how I can get in touch.
Adding renewables or nuclear while you massively increase your world-leading emissions output, won't actually improve anything.
"China Is Still Building an Insane Number of New Coal Plants"
https://www.wired.com/story/china-is-still-building-an-insan...
https://www.npr.org/2019/04/29/716347646/why-is-china-placin...
https://www.washingtonpost.com/world/asia-pacific/years-afte...
> However, few plants have commenced construction since 2015, and it is now unlikely that this target will be met.
Also, take a look at the bar-charts on the Wikipedia page.
https://twitter.com/globaltimesnews/status/11535899795340738...
Meanwhile also claiming "China says it has already hit 2020 carbon reduction goal".
https://thehill.com/policy/energy-environment/472224-china-s...
I don't think they can be trusted with anything they say.
Per capita, it's not the "3rd" world countries that are polluting, but the "1st" worlds (https://cotap.org/per-capita-carbon-co2-emissions-by-country...). I'm sad to see someone still referring to countries as "1st" and "3rd" world, as that's an incredibly outdated term. Also to refer to China as a "3rd world country" is astonishing. I recommend you read the book Factfulness (https://www.gapminder.org/factfulness-book/) to get a better sense in how the world works.
> by China and 3rd world countries
By Mary and the guys (indicating that Mary is not a guy).
> Per capita, it's not the "3rd" world countries that are polluting, but the "1st" worlds
From your own source, the top polluters per capita are Qatar, Trinidad & Tobago and Kuwait. Qatar is wealthy, but probably not "1st world".
> I'm sad to see someone still referring to countries as "1st" and "3rd" world, as that's an incredibly outdated term
The AP standard was updated in 2015. The terms "1st", "2nd" and "3rd" world are still in common parlance even though the AP prefers "Developed" and "Developing".
I reiterate what I said before, please read Factfulness, and then get back to me.
Also, funnily enough, China was never considered a third-world country, but a second-world one, you know, in Cold war terms.
Again you're really struggling with the grammar here.
"Mary and the boys".
Is Mary a boy? No, Mary is not a boy. We are indicating Mary is in fact, not part of "the boys". Likewise, "China and 3rd world" indicates China is not 3rd world.
> Show me the AP standard
https://twitter.com/apstylebook/status/631085111220547584?la...
https://www.npr.org/sections/goatsandsoda/2015/01/04/3726844...
> I reiterate what I said before, please read Factfulness, and then get back to me.
Wonderful book, not related to our current disagreement. If we're throwing out recommendations though, then I'd recommend Strunk and White.
The US and other more developed economies are plateauing in terms of emissions per capita. While there's ample opportunity to begin to claw back those numbers, it is not where the problem lies moving forward.
United States: 16.5 metric tons of CO2 per capita
--
Other developed countries:
Germany: 8.9 metric tons of CO2 per capita
France: 4.6 metric tons of CO2 per capita
UK: 6.5 metric tons of CO2 per capita
Spain: 5 metric tons of CO2 per capita
--
This is 2014 data (for comparability), but in 2018 China still had half (7.95 metric tons per capita) the CO2 per capita than the US
https://news.ycombinator.com/item?id=21942100
I've invested in rideshare companies with the assumption that humans will still largely be behind the wheel in 10 years.
(Because nobody's going back to taxis; rideshare is more backward compatible than public transit buildout; it's harder than you think to start a new rideshare company, etc. Also once something becomes popular, people underestimate how long it sticks. Similar to somebody predicting on the 2010 thread that Facebook would be gone by now.)
Side note: I've made bearish predictions about self-driving on HN for at least 2-3 years if you check my comment history. I'm don't want to be negative, just realistic. I'm bullish on the software industry in general and video in particular.
Have you ever seen longbets.org?
My favorites:
Would you still do your bet with your “average” restricted to the US? What about major US cities? (By the end of your comment, it seemed you were at “average person in the whole world”, including developing, such that nobody should take your bet within a 10-year timeframe).
I think the biggest risks are regulatory and weather conditions. For the first: will NYC really allow AVs in time for 2030 to have any serious adoption? That looks questionable currently. As for weather conditions, you didn’t explicitly state it, but an autonomous system that only works 50% of the year means people need their current mobility and the AV option, leading to high cost.
The Bay Area and Los Angeles only have rain though, so it’s not too hard to imagine a world where AVs can operate in big California cities successfully, most of the year. I’d make that bet: Californians will have driverless cars with material impact (but perhaps not car replacement for the masses) for the median person in the major metro areas by Jan 1, 2030.
Edit: updated with my own bet and phrasing.
I live in SF and worked at Google so I've been hearing all about self-driving since 2009 (and I remember when co-workers went to the DARPA competitions in 2006).
Look at the top comment on this article from February 2018:
https://news.ycombinator.com/item?id=16352467
It’s great to explore this topic. You can have a lot of fun in picking a random industry and imagining the effect that self driving will have
A lot of that is true IF we get to level 5 soon -- but it seems like most people don't believe that anymore (for good reason).
https://a16z.com/2018/02/03/autonomy-ecosystem-frank-chen-su...
One thing’s for sure: We’re at a critical inflection point with this technology, so the shift is going to happen a lot faster than we think
Here's my comment which I think should have been totally uncontroversial since I simply quoted three people who are very close to the problem: https://news.ycombinator.com/item?id=16353541
However I got pushback on that comment, and also when I said similar things in person to people even further away from the problem than I am. I remember a summer 2018 conversation where an engineer was poking fun at the tech-unsophisticated for doubting that self-driving is possible.
----
Take a look at this 2013 nytimes article, which might have been the height of it:
That city of the future could have narrower streets because parking spots would no longer be necessary. And the air would be cleaner because people would drive less.
I would definitely take a bet against changes in city layout or the the air being cleaner in 2030 because of self-driving. (Both of them will improve in many places for other reasons though.)
I would also take a bet against self-driving impacting your average senior citizen or disabled person, which has been in the marketing of at least a few self-driving companies.
“I could sleep in my driverless car, or have an exercise bike in the back of the car to work out on the way to work,” he said. “My time spent in my car will essentially be very different.”
note: the article does present both sides: I'm picking a few people's opinions. But I heard a lot of that in person over the last 10 years.
-----
So really I'm not predicting anything controversial... it looks like the hype has already died down. I'm more interested in my other predictions on this thread.
But I also won't rule out the possibility of a "black swan": if there's a big innovation in self-driving, it probably won't come from Waymo, Tesla, or any incumbent.
That was the story that Uber/Lyft sold to their investors prior to their IPOs. Here's a fantastic (and long) article by Tim O'Reilly that:
https://qz.com/1540608/the-problem-with-silicon-valleys-obse...
Despite that I still think rideshare is valuable and there's no going back. I think Uber and Lyft will raise their prices significantly and people will take them anyway.
"The majority of people don't draw the privacy line where I draw it, therefore they don't value privacy at all."
Huh, I never thought of this before but you are right, seems like rejecting technology is becoming more acceptable in the mainstream with the 5G conspiracy.
People who can afford to reject the services provided, or can afford to buy them non-sticky elsewhere, will. Same as it ever was - money buys freedom.
Man! You really know how to drop a bummer for your last one :(
Spoiler: it will float down.
Wealthy nations have produced massive emissions during their industrialization, and continue to do so on a per-capita basis. The global north has the responsibility to cut back to a greater extent so poorer countries can develop.
Isn't that already the case and not a prediction per se?
> 6. But a small yet growing culture of "offliners" becomes mainstream. Being offline is the new "Yoga" and allows bragging rights.
Meh, not going to happen. On the contrary being connected will be more than ever the "new utility" to do just about anything in life.
> 7. Increase in adoption of non-scientific beliefs such as astrology/anti-vaxx/religion/flat-earth as a counterbalance to the increased complexity of everyday life.
That's already the case with most people have a very poor understanding of scientific principles or even basic stats.
> 9. Global carbon emissions are not reduced, mostly because of lack of initiative by China and 3rd world countries.
China is building more nuclear power plants than any other country on Earth.
Similarly, it looks like wall street already has job losses from automating trading. Is that an ai automation? The problem is you could label any job losses due to automating things as not ai, even though actual jobs will continue to be replaced more and more by computers. Have you seen the mcdonald's where you order it yourself instead of using a panel, there must have been fewer people hired to take orders there.
Which year do you live in?
This happened already, so long ago in fact that even Instagram is already on the way out as the mid-30s social network, and TikTok is taking over with it’s real-life softporn focus.
> No major job losses due to AI automation
First off, #2 is really 2 predictions. 1) No AGI 2) No major job loss due to AI automation. There can be job loss due to automation without AGI (which I'm EXTREMELY bearish on AGI, bearish for the century actually).
Second, here's a scenario I think we could see job loss due to autonomous vehicles. Trucking. If we get level 3. We're defining level 3 as
> In the right conditions, the car can manage most aspects of driving, including monitoring the environment. The system prompts the driver to intervene when it encounters a scenario it can’t navigate. Driver involvement: The driver must be available to take over at any time.
I see this as near tech and could cause disruption. We already have level 2. If regulations change, then I could see job loss.
Or another scenario. Let's say that trucks are level 4 on highways in clear conditions, level 2 in cities, and level 1 in bad conditions. Regulations could change so that a driver could "drive" for longer hours and even sleep given that trucks are driving in clear conditions, automatically pulling over to the side of the road/sounding an alarm/slowing down when conditions are deemed unsuitable (with a low threshold). Such a situation seems near possible with current tech and have the potential to disrupt the market. Knowing if that's going to create more drivers or less drivers is harder to say.
I'm also bearish on level 4/5 vehicles in the next decade, but I'm not bearish on driving staying the same within the next decade. Especially as we see more cars becoming level 2. I am bullish on disruption here, including regulation. Though I'm not going to try to define what the disruption is, other than increased safety (which is pretty huge).
I'm not bullish on AI causing major job loss, but I am bullish on automation (not necessarily %s/AI/ML/g) creating disruption. In fact, I think we're already seeing this. And I want to be clear, I don't think we need ML to automate things. We've done a lot of automation already without it. I think COTs electronics becoming cheaper is the bigger driving force to that along with consumer preferences of using digital systems (Amazon vs store, self checkout vs cashier, movie ticket touch screens vs box office, etc).
But they exist right now! Just not for the average consumer. https://m.youtube.com/results?search_query=flying+hoverboard...
Same with jetpacks (Gravity Industries will sell you a jet engine based jetpack for ~400k)
How do you see a culture of offliners developing when they can't organise by social media?
I think this is about 50/50, but will take infrastructure changes to really bring about level 4 -- think smart roads and moving pedestrian walkways above/below street level, or possibly every car as a mag-lev/car hybrid. Leaving mag-lev turns into manual.
> 2. Same with AI. The panacea hype dies down. No AGI at all. No major job losses due to AI automation.
I disagree AI will not die or have a nuclear winter, it'll have tremendous leaps/bounds, but nobody claimed AGI would happen in next 10 years, most estimates put it at 2060 or greater
> 3. Facebook (the SN) still exists but ages along with it's current user base. i.e it's the "old people's" SN. Facebook (the company) is still going strong, with either Instagram or one of it's acquisitions being the current "hip" SN.
This is definitely how I see it playing out, FB will monopolize any social media sphere that looks like it's going to 'explode' and be the next thing through acquisitions until they get a monopoly ruling by the gov't and get split up (if/when I'd say 40% likelihood).
> 4. Google still dominates search and email but losses value and "glory" compared to today.
I think google won't lose too much value, assuming they go all in on cloud. I think GCP is where they'll make the most money. Possibly also Google Business accounts for email/etc.
> 5. Majority of people still don't care about privacy.
Not until the country enacts a social credit system like china, which I'd say is a 60% likelihood but it'll be 'privatized' not state ran, so that makes it better right?
> 6. But a small yet growing culture of "offliners" becomes mainstream. Being offline is the new "Yoga" and allows bragging rights.
I don't think this will happen, except either among extreme hippies (and I'm a progressive) but I mean the live in a van off the grid types or the other side live off the grid but right-wing survivalist types. But there's already folks like this.
> 7. Increase in adoption of non-scientific beliefs such as astrology/anti-vaxx/religion/flat-earth as a counterbalance to the increased complexity of everyday life.
God save us from this future. I hope we can educate the dumb out of people, with some free college and better public school systems. Maybe end home-schooling unless we can make sure this doesn't propagate as a result.
> 8. Web development matures and a "standard" stack is accepted, all in JS. HAHAHAHA HAHAAHA HAHAHA. Wait, are you serious? If anything it'd be probably rust/web assembly for it's speed/benchmarks but I don't see anything in web ever becoming one solid framework or stack. If anything it just keeps splintering. The problem is devs are WAY too opinionated about the 'right' way or 'better' way or 'familiar' way of doing things.
> 9. Global carbon emissions are not reduced, mostly because of lack of initiative by China and 3rd world countries.
This depends on 2020's election. If we see Bernie or Warren win, I think we could lead the world in curbing emissions. We won't win
> 10. Still no hoverboards. Never will be. Nor hover cars (except maybe mag-lev).
I laughed!
> 9. Global carbon emissions are not reduced, mostly because of lack of initiative by China and 3rd world countries.
I cried.
> 10. Still no hoverboards.
:( I wanted to believe.
1. Still no level 4/5 autonomous cars anywhere in sight. The promise of being "just around the corner" fizzles down and people just forget the hype.
2. Same with AI. The panacea hype dies down. No AGI at all. No major job losses due to AI automation.
3. Facebook (the SN) still exists but ages along with it's current user base. i.e it's the "old people's" SN. Facebook (the company) is still going strong, with either Instagram or one of it's acquisitions being the current "hip" SN.
4. Google still dominates search and email but losses value and "glory" compared to today.
5. Majority of people still don't care about privacy.
6. But a small yet growing culture of "offliners" becomes mainstream. Being offline is the new "Yoga" and allows bragging rights.
7. Increase in adoption of non-scientific beliefs such as astrology/anti-vaxx/religion/flat-earth as a counterbalance to the increased complexity of everyday life.
8. Web development matures and a "standard" stack is accepted, all in JS.
9. Global carbon emissions are not reduced, mostly because of lack of initiative by China and 3rd world countries.
10. Still no hoverboards.