Hacker News new | ask | show | jobs
by pc86 900 days ago
Protectionism doesn't work beyond selfish short-term interests. American cars are not exactly paragons of technical or mechanical prowess compared to their Japanese or German counterparts.

I've been a software developer for the better part of two decades, I'm not worried about the C-tier code coming out of rural India. You shouldn't be either unless you're a really bad dev.

4 comments

I’m okay with my fellow citizens being selfish and protecting their livelihood. No one else will. “Whose interests?” You’re advocating for shareholder returns (labor arb savings->profits). Ain’t nobody coming to save us.

If we can mandate EV batteries be built in the US to get subsidies (Inflation Reduction Act), other protectionism mechanisms should be on the table. Otherwise, businesses will do their best to maximize profits in the market they’re offering in without any labor contribution back, extractionist style.

And at the end of the day, the ones doing the outsourcing will have outsourced themselves out of a job too. China won’t need American MBAs when they can do everything in house.
Jack Welch died before getting outsourced after driving GE into the ground, and there are lots of folks like him still alive, empowered, and with that belief system. These are the folks controls are needed against.
I appreciate you telling me what I'm advocating for. Any basic macroeconomics class covers the effects of protectionism. Yes, in the short term wages may be artificially propped up, industries may be (temporarily) saved, but you do long-term damage to the economy, and the people are better off with free, open markets long term.
> but you do long-term damage to the economy, and the people are better off with free, open markets long term.

The evidence does not show the American worker being better off after these policies you support were enacted and have had decades to run. Free trade is great for shareholders and some consumer cohorts who get excess utility, but terrible for workers. “The definition of insanity is doing the same thing over and over again and expecting different results.”

https://www.epi.org/publication/briefingpapers_bp147/

https://www.epi.org/publication/botched-policy-responses-to-...

https://www.epi.org/press/globalization-lowered-wages-americ...

https://www.npr.org/sections/thetwo-way/2015/12/09/459087477...

> Any basic macroeconomics class covers the effects of protectionism. Yes, in the short term wages may be artificially propped up, industries may be (temporarily) saved, but you do long-term damage to the economy, and the people are better off with free, open markets long term.

You're almost certainly correct in the sense that the people of the entire system will be better off, but your own domestic market could suffer at the gain of the other market where the business is now being outsourced to.

A good example of this might be tech in the EU. The EU basically has no major tech companies because we "import" all our tech services the US (Facebook, Google, Amazon, etc). It's great for us in the sense that we didn't have to pay anyone to build amazing online services like Facebook, Google, etc – it's just free stuff we get here from the US. But who benefits the most from this arrangement, is the US or the EU? I'd argue that the EU allowing the US to provide all of our major tech services has been great for US growth, but it's stagnated the EU economy in recent years as we've had no real reason to build 21st century companies here. The free stuff we get from the US actually comes at a cost for us even if overall the economy as a whole (EU + US) is better off for it.

Similarly, imagine an extreme scenario where US companies outsource all work to low-cost labour countries (I know this is impossible, but assume the US is 100% service sector jobs which could be outsourced). Would this hypothetical scenario be good for the US economy? It might be good for companies registered in the US because now they can provide their services to markets they serve for a fraction of the cost, and it would be great for those low-cost labour countries getting all this foreign work, but it would be awful for the actual US economy that's allowing this to happen in the pursuit of efficient markets.

So yeah, you might be growing the whole pie at a faster rate, but it's possible mass outsourcing doesn't help grow your share of the pie. And like with manufacturing, you also need to consider how you'll lose technical competency within your domestic market over time if you outsource too much, and this will likely lead to the country you out sourced to eventually out competing you in your own industries. We see this today in China.

If you want to cripple tech innovation in the US, outsource all your software engineers so there's no one in the US with the skills or resources to start the next Google or Facebook.

> and the people are better off with free, open markets long term.

The theory only says that all the people globally will be better off. It does not say anything about citizens of a specific country that is applying protectionism. They may be better off for it, they may be worse - it depends on the particulars and, as most economic interventions, can only really be judged post-factum.

Do people forget that a 40 hour work week exists because of unions? Did we forget that having weekends off and holidays off was not a thing until unions? Did we also forget that children were put to work at a young age until unions?
I think many well-off people just don't care. They might support them in a poll but question their utility in a specific sector like software. Or how another commentor said they weren't concerned about competing with people writting c grade code in India since they have been writing software for 20 years.

I don't think unionizing will help with off-shoring for development. I do think unionizing could bring about more equal treatment though. It seems most companies ignore their own policies when it comes to ratings, work assignment, hours, etc. Devs have very little recourse. Of course the assholes that do well and aren't afraid of c grade coders are the ones who don't want unions since they might loose their edge over others comp-wise.

No, a huge amount of Americans are actively hostile to unions as a concept. The very idea that employees should be able to negotiate as a block instead of individually has been scapegoated as basically all of society's ills.

Consider the common refrain: I'd rather negotiate for myself

It's like, a fundamental misunderstanding of how power dynamics work, as if you as a solo person in a 1000 person company could somehow EVER be more valuable to the company than the entire labor pool.

Newsflash: If your company doesn't throw a fit any time you try to take time off, like CEO comes and talks to you personally fit, they think they could replace you just fine. 40 years of project management has attempted to build things just for that.

> Do people forget that a 40 hour work week exists because of unions?

Is it still forgetting if it never happened? Both the idea of the 8-hour work day and weekends predate the first recognized union. The 40-hour workweek became the norm during the Great Depression by way of government initiative in an attempt to spread the work out across more people.

Unions have long supported the 40-hour work week, but are not meaningfully responsible for it. If showing support for something is necessary for something to exist, then you could probably say that just about everything exists because of unions...

Even still, we're talking nearly 100 (when it became common) to well over 200 years ago (when it was conceived). Even if unions actually were responsible, people are going to naturally ask "What have you done for me lately?". Where is the 10-hour workweek?

The weekend was due to organized religions. The Atlantic has a great article on the history of it. Henry Ford pioneered the 40 day workweek without union prodding, only through his own experimentation.

Besides, let's say youre right and unions have given us all those things. Those standards are over 100 years old. If it were true, it would mean nationwide unions did some things 5 generations ago and have collected literal trillions in inflation adjusted dues* since and have provided nothing in return.

* 11% of the US is unionized, representing about 1.1 trillion in annual payroll. Average union due is 1.5%, that's $16.5B per year in union dues collected, over 100 years = $1.6 trillion. But union membership used to be much much higher than 11% so its actually a much bigger number than $1.6T. But you get the math.

Only Sunday was off thanks to Christianity. Free Saturday only came in early XX century. Before that it was work 6 days a week.
In the long term we're all dead.

Short and mid term matters a lot to people.

"and the people are better off with free, open markets long term."

Eh, maybe not. It depends on the demand and availability of skill/labor. If you have a high percentage of low skill labor and you can outsource low skill labor to cheaper markets, then what are the current low skill citizens going to do? Surely the rust belt is not better off now than when coal and steel (and other manufacturing) were still a domestic thing. Maybe other areas of the county faired better, but with median wages dropping over the past 50 years, it doesn't seem like a strong case.

> You shouldn't be either unless you're a really bad dev.

I dislike the tone as there are plenty of good devs who've been cut and replaced (sort of) by offshore. Don't equate laid off/replaced with "really bad dev".

The downside is for people disabled. Maybe they aren't a terrible dev but they aren't a great dev either. If they're on par with outsourced labor, they aren't so safe. But what else can they do?

If I lost my job right now, I'd be totally fucked. I'd end up working at Walmart. Masters degrees might as well be toilet paper.

I remember driving through West Virginia with my parents to visit family as a kid, and my dad was lamenting the fact that its full of the haves and the have nots, with the distinct implication that the haves did something wrong to end up there, and the have nots would be just fine if it wasn't for those pesky rich people. I was just left thinking that if life was haves and have nots, shouldn't you spend your time trying to be one of the haves rather than lamenting the way reality was? But in reality, both those views are overly simplistic.

It's a pretty big leap to go from a software engineer to Walmart. The median software developer (~$110k/yr last I checked but could be outdated) is somewhere in the upper teens as far as income percentile (20% being around $100k and 10% being around $150k[0]). Pretty much any non-management role at Walmart is going to land you in the bottom half.

I'd be curious (but it's none of my business) what about your situation makes that the most likely outcome. I'd bet there are ways to head that off.

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

Just in general, what do you think an alternate job would be for an ex-dev (without spending 10s of thousands on reschooling)?

Seems like retail, warehouses, and other unskilled labor are the main options. Even something like teaching would require a certification.

Basically anything tech-adjacent - product, "business analysts," management. Maybe something like tech writing but there are going to a lot of decent devs who make terrible tech writers so that's much more on an individual basis. And then there's the devs who still code but that's not their focus - SDET, devrel, that sort of thing.

Devrel would actually probably be the easiest thing to get into with a few years' foresight. Building a following on YouTube, Twitch, Twitter/X, etc. will make it infinitely easier to land that first devrel role.

You bringing up teaching is a good point - I'm not sure about where I currently live but where I used to live you could get a substitute teaching cert basically by just passing the background check and having a college degree. It was pretty easy to get an add-on certification as well to teach your subject or closely related ones. I can't say what it's like everywhere though, and to be honest most of the people I know who have teaching degrees have left or wished they could. But if you can get a job and can deal with the bullshit you're basically set from a put-food-on-the-table standpoint.

Most of those roles you list require strong interpersonal skills. That's usually not something an autistic person is strong at. Not a big market for many of those either (dev rel, tech writer).
I'm looking to switch into IT from teaching. The job security in teaching is the best but half of my colleagues are in psychotherapy. Last year we had 30% of our teachers on leave. It's not a happy life unless you're a social butterfly and unless you enjoy yelling at kids.

Can't wait till I'm out.

Options:

Tech Writers - nope. Nothing dealing with boilerplate text is safe, in any field.

Twitch/YouTube - nope. That is just celebrity economics again. For every half dozen people that make it (and make it is just back to median dev salary) there are thousands that only get a few viewers.

Teaching - this is option. But as noted by others. Can have own problems and a lot of people leave.

I'm an old Dev looking for second career, and it is tough. The option is to just re-skill and be dev again in another industry. Dev's be Dev's. It doesn't seem like there are many upward paths, and limited sideways paths.

What, product manager, analyst, marketing? Tried them, they all have downsides.

Even with all the crap, I only find comfort in creating things. Coding.

If talking percentiles, you probably mean 80th percentile instead of 20%, and similarly for the other number - 90, not 10.
the new CEO of Google is from India, no?

I see a lot of very qualified people coming from India, around a big University here.. They have had twenty years to build to this.

There are extremely talented Indian engineers, but I don’t think this discussion is about the elite “cream of the crop” but more so the rank and file.

I have been assigned to lead offshore teams with engineers that need direct guidance on very basic coding tasks, produce low quality code and become combative when receiving feedback.

So many times I’ve reviewed and requested the same changes to code, classic example: a try/catch then completely ignore a caught exception, just to get some code “working” for the example inputs. When I call this out as a problem, it’s met with “well you did not say that in the specification, I have completed what was asked”. Another commenter had a similar anecdote where all input validation was removed to get test cases passing… are we expected to write things like that into a work spec? Seems like this would take longer than for me to write the damn code myself

There's gotta be something between Satya and "C-tier code out of rural India".

There are of course lots of bad coders in India, but there are also many really good ones. And whereas in the past they had to emigrate to US or Europe to fully make use of their talents, nowadays some(many?) choose to remain in India and work remotely. It's silly to dismiss and underestimate their skills.

As far your experience with developers that follow the specs literally, in an almost maliciously compliant way, that might be learned behavior from working on projects where the tasks are spec-ed and estimated and any attempts at going above and beyond ultimately result in late delivery and punishment, so developers quickly learn to only do the bare minimum of what is described. Granted your examples are extreme and pathological, so maybe you just had the misfortune of working with really bad people.

Additionally, unless you pick the developers yourself, you're at the mercy of the agencies who assemble those offshore teams, and often the economics are such that it doesn't incentivize them to hire the best people available. From my experience, many good developers find work on their own, outside of an agency, contracting directly with the remote company.

Those talented indian engineers are paid on the global market rate though. The logic also goes the other way, those don't want to be underpaid.

That's why these outsourcing threats from companies make me laugh, they don't understand that software is a global market and they also are competing in it.

The world class FAANG-level Indian engineers which are underpaid just do not exist.