Hacker News Reader

Comment by echelon | Hacker News Reader

Our industry feels like it's in the decline of the Roman Empire right now.

Our craft is becoming arcane: Microsoft is building VSCode for the cloud. Few people own desktop computers anymore. Apple is locking down MacOS to signed binaries. Soon we'll all have thin clients and will write software in the cloud for the cloud. Few will be able to learn how to engineer at home anymore - you'll have to go to a university or trade school. Building a simple website will require JS package management, kubernetes, and AWS. The "free tier" will fall away.

Employees aren't excited by what they do: People job hop for promos, raises, and resume fodder, not because there's an actual interest in the product. The majority of software projects are mundane glue, pipes, optimization tweaks, or migrations.

Companies don't care about employees: Wages are adjusted for cost of living. Companies are trying to hire more outside of California in order to lower wages across the board. Employees at "passion project" companies (gaming, SpaceX, Tesla) are overworked and underpaid.

Incentives are wrong: Google bestows promos on launches, not KTLO. Google kills products that make customers and engineers happy and doesn't spend enough effort to make them better.

The industry is eating the commons: The web is turning into shit. Devices are being locked down. Ads and tracking are everywhere. Open source is being co-opted, but the benefits are not returned. The cool things to hack on and integrate with (remember "mashups"?) are all siloed up, full of ads, and dumbed down to the most basic products.

The Internet now treats everyone using it like we're all brainless idiots to be injected with dopamine. Like we're not smart enough to choose how to consume it. And working for companies in this field now feels like working for the insurance industry. Optimize for the average.

View on HN

Unfortunately I feel it is hard for me to disagree with the overall sentiment of what you are describing.

But I'd like to offer a little glimmer of hope from someone with too many gray hairs to count.

Computers are still machines, and we still tell them what to do by programming them. It is still possible to stay out of what you describe above, it is still possible to work on low-level code, on real products, on things that matter to people and companies.

The reality though is that the mundane things are where the money is though. While we all enjoy "cool" things, millions of companies need their accounting and payroll running, their truck drivers planned, their work schedules planned, ...

So you might find that more interesting things happen in e.g. industrial engineering companies or electronics companies, more so than in web companies. It's an entirely different world than what you describe above.

Hackaday might be a more interesting hangout than Hacker News for you ? So let's not lose our excitement and sense of wonder, after all it's all bits and bytes and we tell them how to behave.

> Computers are still machines, and we still tell them what to do by programming them. It is still possible to stay out of what you describe above, it is still possible to work on low-level code, on real products, on things that matter to people and companies.

Yes it is possible, but in practice one will (sometimes have to) spend most of the time with devices/technologies that make one die inside a little.

Part of the problem is that "doing the right thing" is often not competitively viable.

This may not be specific to IT only, either.

Some of the most inspirational stories in tech are along the lines of "I had this terrible job where we had to use tech X so I wrote adapter Y in more cool tech Z and automated the whole thing and took long lunches."

> Hackaday might be a more interesting hangout than Hacker News for you?

The Hackaday focus on the dumbed-down format of video presentation is infuriating. It would be very nice to fund something community-driven, focused on actual content, with the same subject.

Until then, I think the few articles on the Hackaday subject that appear here on HN are worth more than the entire Hackaday site.

(I still visit both, of course, but tend to only look at the headlines there.)

> So let's not lose our excitement and sense of wonder

Exactly. It’s wild the potential that many of us hold.

And then age...

I'm an eighties whizz-kid, educated hardware and software background with borderline epiphanic understanding of computer technology. From firmware to systems where peoples lives literally depend on. In 2000 jumped on a plane to Taiwan to develop the first Arduino/RaspberryPi class devices. Still up to date and currently busy polishing a ShowHN which will blow you off your socks.

Everything nowadays is shrink-wrapped out-of-the-box, hidden behind layers of frameworks. Programming languages becoming a fashion statement. Everyone seems to copy-paste stack-overflow and even then the answers are like "Why?" because it seems nobody reads specifications.

But...

Finding a job position is impossible. The excuses for rejection range from unbelief to insulting. It is so bad that I have now seriously considered leaving the field, taking 40+ years experience with me.

> It is so bad that I have now seriously considered leaving the field, taking 40+ years experience with me.

Please don't. We need more people like you to infuse much needed energy in a field/profession that is becoming mundane and steadily going downhill day by day.

Talk to people you know. Should help find a job without the inane interviews. Blessings to you friend.

I left two years ago and have been largely pursuing my own extended interests. One result is that 90% of the HN postings seem to be absolute wastes of time (to me at least.)

Stay curious.

> Finding a job position is impossible. The excuses for rejection range from unbelief to insulting. It is so bad that I have now seriously considered leaving the field, taking 40+ years experience with me.

Before you do, are you up for a Zoom call or something? Talk about anything you're passionate about, and perhaps we'll post it as a show HN.

I'm 31 myself and at the beginning of my career (I've been longer in the education system than most).

My email is in my profile.

> Finding a job position is impossible.

What's your email? I'd love to reach out to you and discuss some openings that we have on our team.

Thank you. It's mailto:xyzzy%40rockingship.org

I love the %40, that's a great way of obfuscating an email address that I've never seen before.

> The excuses for rejection range from unbelief to insulting.

I was once rejected because I did not know inane crap like what "this" is bound to at the top-level in JavaScript. Their answer: "the window object". a) Yeah, okay, if your code only runs in the browser. But not nodejs. And b) if anyone uses a top-level "this" it's an instant reject in code review from me.

Ultimately, a lot of these interviewers are waving huge warning flags saying "stay away!"

Even FAANG. Every single algorithm there is a "trick" to it. Once you memorize that trick, every leetcode becomes trivial. But it took a guy like Knuth some 30 fucking years to come up with that "trick". It's so goddamn stupid memorizing bubble sort when every language has a built in sort you're going to use. And if you ever see a developer writing their own algorithm for bubble sort at work, you're going to think something is wrong with this person. What the fuck are you doing guy? You're going to get it wrong and there is a library for all of this, used by thousands of people with hundreds of bug fixes.

It's because the business people got interested and are increasingly taking leadership roles. As an engineer who has shifted over to the product management side I can safely say business people suck fun out of stuff for a living. Margins, cost savings, metrics to manage, each and every slow step of creating a new, less fun, normal erodes a culture built on making "Something cool" or "Just for fun".

Worse yet is when the business minded masquerade "Fun" activities as them still being cool such as hackathons. The sad truth of the matter is that this is what happens in a maturing industry. A lot of the green field money that sat around for Facebook and Google to gobble up has been eaten. The remaining scraps are enough to prop up small startups, but really only until they get absorbed into the giants with fat checks for the founders. The regular workers get conformed into the system, with assurances "Everything will totally stay the same" as they slowly change everything.

Businesses exist to make people money. Engineers are great at solving problems and innovating. Business people are great at squeezing every last drop from those solutions till only a husk remains. To be fair to the business people, myself included, if you do anything less than squeeze for money as hard as you can, you get beat out by the competition who copy you, and nothing is as not fun as being jobless. The more and more people who get coding knowledge the more and more being a coder will be just like every other engineering position. Someone who executes. Which is why I bailed and got into PM so I could actually have a strong say to the business in what we execute on.

To end on a happy note I have never seen more people getting into robotics clubs, trying to make small static sites, learning the basics, and I am certain that every day more and more people are running their first hello world programs. So I think there some ups and downs to everyone learning coding.

Agile/Scrum was the biggest bait-and-switch con out there - developers thinking the methodologies would empower them while in fact being a vector for the worst excesses of micromanagement. I feel less empowered now than I did in my earliest days as a junior dev.

Yep, I call it collectivised micromanagement in an attempt to commoditize development.

Whats driving business types to think leetcode is a good test? This should be an easy conversation with HR. I think software engineers want this arbitrary barrier.

Leetcoding involves preparation for known problems and churning out solutions in least possible time. Most of the business have problems along the same lines (think of CRUD, authentication, authorization, UI animations etc ...) and they think people who excel in the former might do well in latter.

Edit: better sentence formation.

Honestly this is probably because other businesses started doing it. When businesses have no clue they look to the market for the "standard" way of doing things. We don't make mechanical engineers design mechanical solutions for fun.It is just a consequence of the early culture of "hackers" and people who do that stuff for fun and for work. Hopefully dies over time, but it is unlikely given that now seems like the standard thing to do. Businesses are not rational thinking creatures. They just do what feels like lower risk if they can.

> Our industry feels like it's in the decline of the Roman Empire right now. ....

And here I am, writing a desktop native app that targets a small niche of users, with no marketing, no advertising, and we've had our strongest year yet, and overall usage just grows slowly year over year.

Maybe it's how you think things ought to be that is misaligned with reality, rather than "the decline of the Roman Empire".

Rome took quite some time to decline, and to fall. And there were always pockets of stability. Maybe your app is Byzantium.

And hopefully it's not Byzantine.

> And here I am, writing a desktop native app that targets a small niche of users

What is it?

Looks like it’s Ardour DAW https://ardour.org/

Yikes, tried to use Ardour multiple times back in the day (I used to work as an independent radio producer and sound engineer). Classic open source desktop software in that it promised the world, but in practice was extremely difficult to even get to install, and reliant on external packages that were themselves glitchy and baroque (I believe Jack was the common issue in this case).

The prebuilt binaries of Ardour that we supply have no external dependencies except for libc, the native window system (X11, Cocoa/Quartz, etc), and libstc++. Since Ardour 5.0, JACK has been optional.

Every week, more or less, someone tells me that Ardour is a piece of shit, and someone else tells me it's the best DAW they've ever used. I gave up believing that any given DAW could ever make everyone happy about a decade ago.

Looks like Ardour is open source. How do you make money?

Source is always available without cost. Pre-built binaries are available via a pay-what-you-want pay tunnel. Subscriptions are encouraged to allow us to continue developing the software. We made on the order of US$200k this year.

Thanks for replying. This is a great approach and seems really well suited for the target audience.

Just because one household in the empire is having a good year doesn't mean the empire isn't on the verge of collapse.

The thing is, it's not one household. Every other company I'm in touch with in the audio software world (quite a few) had a banner year. Yes, the pandemic helped us by forcing lots of people to be stuck at home and turning to that song they'd always meant to record/write. But it was the icing on the cake for a market that generally continues to improve, despite stupid levels of competition.

Now of course, if by "household" you meant "the entire niche of audio software", then sure. Or maybe even more broadly, "the entire niche of creative software", then even more sure.

But these are not tiny niches, even if they are small.

>Companies are trying to hire more outside of California in order to lower wages across the board.

Lower labor costs mean lower costs passed on to the end users which means more products in the wild which means a bigger industry. It's only a problem if you're worried about being out-competed by "those hicks in the flyover states".

Techies complaining about companies moving out sounds a lot like the guy who worked the UAW system for 30yr so he could land a buddy position where he does nothing for $60/hr complaining about non-union factories and companies moving to Mexico. In both cases it's a tacit acknowledgement that the gravy train is unsustainable.

Yeah wages might be slightly lower in 40yr but that's a reflection of the difference between being an electrician in 1910 vs being an electrician in 1950 after the last of the line shaft industries had long been converted, once all the heavy lifting and low hanging fruit has been picked there's just a lot less cash sloshing around driving up prices and wages. The people who just want to be mediocre and punch a clock will still make a decent wage but the difference is that mediocrity and job hopping aren't enough to make you big bucks in a stable industry like they are in a rapidly growing industry flush with cash. In a stable industry you actually have to be worth the big bucks.

It's hard not to see Bay Area software developer wages, especially at the big "tech" companies, as something of a bubble. They're completely out of line with engineering and science jobs more broadly, including software engineering jobs in much of the US.

Which does suggest that someone going into software just for the money or, as GP or so commented, going through the motions and spending all their time prepping for Google interviews may be chasing a pretty transitory target.

Historically there's very little to suggest that the segment of one profession that's largely in one geographic area can sustainably capture super-premium compensation even though it doesn't really require a higher level of skill than many others have throughout the country (and the world).

And the fact that a lot more flexibility in living location has come in essentially overnight means that a lot of companies are going to start looking more skeptically at some of the premiums they're paying.

This is the result of a war that was fought two decades ago and the hackers/coders/sysadmins, the people to built things, lost and lost big. It was the dot-com crash. Looking back it's absurd. Everyone was convinced there was this big bubble and everyone got caught up in the irrational exuberance of the future of computing. But was it really? The real financial damage was done by the likes of Enron and Anderson consulting. If it was so irrational why did that supposed bubble give rise to some of the largest companies that now exist? Because it wasn't. That's why we spend the next two decades wringing our hands and fretting that there was another bubble right around the corner. This never happened. Why? Because there wasn't a bubble in the first place to worry about repeating.

What did happen is a ton of people lost their jobs and those people, passionate about what they do, kept right on doing it in the form of open source software. Google, flush with cash, hovered up talent for bargain basement prices. The board room got the message that software was eating the world and if they didn't do something quick the people to wrote the software and built the systems might want a say in how those companies were run. So they outsourced what they could and doubled up what they couldn't. The message was clear. I don't need you. I've got a second one of you ready to go. The waged a war on GPL calling it "viral" and "insidious" and convinced everyone it would be much better to use more liberal licenses that would allow them to build and sell products.

The war was lost when they convinced you that management was "icky" and you didn't want that boring old stuff. You wanted to be coding so they hired the MBA to be the scrum master. The result is you were cut out of the decision making process. Your only input was on a technical level and even there you were kept a close eye on. Some people would even proudly exclaim that they liked this system because they thought they were winners based on their slightly higher than average salary all the wile cranking out multi-million or even billion dollar software for someone else. Doug Cutting, the creator of Hadoop and basically everything that was to become Cloudera was only given a job. They made so much money even they were embarrassed but only embarrassed enough to allow him to keep working for them.

You lost and they won and this is the result. You're made to jump through these absurd hoops because it's cheaper to create these pathological systems and waste your time than to do something more efficient.

I feel a common thread in the events and behaviors you describe is the prioritization of decisions by valuation, IP, and boundaries over execution, progress, and flow.

Maybe the only way out of this mess is for business schools to start teaching new metrics for the latter category.

Companies are trying to hire more outside of California in order to lower wages across the board

This could turn out to be a very good thing for the country. It's a social & economic problem if all the money & opportunity are squeezed into just one metro.

Why did you assume those jobs will stay in US?

Have you seen the offers for remote work? Almost every one is "remote (us citizen)", "remote (same timezone)", "remote (monthly onsites)". Unless the companies are prepared to open whole divisions with colocated management in other countries, the jobs are staying local.

Anyone who has worked with remote programmers, unless they happen to have hit the jackpot on them, really has no desire to ever do it again. Timezone differences, language differences, cultural differences, communication differences, etc, etc, all combine to chew up so much more time and money than the nominal savings of hiring someone local. And the good remote people don't cost appreciably less anyway.

You seem to be agreeing with the parent. Large timezone differences, cultural differences, and language differences can be real remote challenges. (Although, in my experience, US--at least ET--to Europe works pretty well. Takes more to coordinate Asia in general and probably want to minimize the amount of synchronous communications.)

After a decade or so working remotely I'd describe that differently. If someone working remotely finds a company with culture where remote work is fully functional rather than something odd that people have to work around, they hit jackpot. Unless everything's documented online, every meeting is online, every water-cooler chat has a link published, etc. then remote work is going to suck for everyone.

Yes. Even then, a lot of the tech jobs remain in the US.

Look at the job listings for US tech company EMEA HQs. It's almost all sales and legal. Occasionally a pre-sales Engineer position comes up.

Meanwhile, they are hiring dozens of Devs at their US HQ.

Hiring foreigners is a pain. Companies did it because the headache offset the savings, without the savings the headache may no longer be worth it.

Many of them definitely won't, but daylight and the location of management's homes are still limiting factors to off-shoring. Synchronously managing people in distant time zones is no fun, and trying to do it asynchronously is really inefficient. There may be a solution to this 'problem', but until they come up with it, letting people work remotely from their homes in the same city isn't a slippery slope to all jobs being off-shored.

Or, probably more accurately, it is a slippery slope, but not a very steep one. I do believe that jobs (and wages) are going to get distributed differently in the future. But, I'm hopeful that it'll be an improvement over having every high-paying job be in the same 5 dystopian cities.

They’re not hiring people in the US.

> This could turn out to be a very good thing for the country.

Once you cross the hump to not seeking physical proximity when seeking to minimize labor costs, there's very little reason to go to the rest of the US once you leave the SFBA.

The country it's good for isn't the US.

Have you ever looked at the map of US HP campuses in their heyday? Dispersed across the country, with regionally clumped physical proximity, a foot in numerous labor markets and all tidily within a couple timezones.

You might think of HP as a California company but they had large engineering hubs all over the country.

> Our craft is becoming arcane

I would say is that a lot of hard problems (of last 20-30 yrs) are getting commoditised (e.g. S3, DynamoDB, Redis) and barrier to entry reduced. This would mean increasing the work is "glue & pipes".

Yes it's glue and pipes. Services mean that you can focus on your core product. And most products are...pretty boring.

exactly, "software engineering" nowadays is just over-glorified plumbing

> Google kills products that make customers and engineers happy and doesn't spend enough effort to make them better.

Heck, as far back as 2010 I got a promo at Google because I was the only one on the Google Maps data team (or the org) who wanted to keep the data up to date. (Also the only one who wanted to live in Seattle -- which might as well have been Siberia to them.) So they shipped me up there and gave me a team. It was like somewhere, deep down, they knew it was a good idea, but each person could forcefully argue why it wasn't -- so that it didn't have to be them.

It was great! Well, other than encountering more of the famously toxic culture, where people get promoted for being smart but nasty. But I'm sure that little problem has been solved.

Could it be that this is a very SV problem? I'm from a European state, I know some people doing programming in Asian regions. What you are describing to me sounds like a bubble. Or I am in a bubble, IDK.

Yes devices can do less than 10 years ago (less arbitrary thing), but they can also do more than 10 years ago. Of course the market in this segment might give you the feeling that everything is dumbed down. These devices are not used solely by very smart people. The most basic of all humans has a phone nowadays. And If apple decides to make MacOS only run singed binaries. That shows that they are targeting people who do not know how to run arbitrary code safely. I am a programmer and when I have to do that I have to think really hard. If my main PC runs stuff that I would rather not like to leave my house. I have to set up a VM and check it beforehand. You still have Linux to do whatever you want.

If people in the job market and Open Source community are getting exploited, it is only a matter of time until something develops to create a force against the exploitment. There are unions (at least there should be) and there are multiple ways to fund Open Source nowadays (Patreon, Gitcoin, Bountys,...).

It doesn't seem to me as everything is moving backwards. It is getting harder to live out the freedom of the individual. But on a macro level, I think things are moving forward.

Yeah, a lot of posts here seem to be trying to say the following, but they don't want, or don't know, how to word it:

Technology was made accessible for the masses. That means 100 IQ people. And then it was refined so the 85 IQ people could work most of it. I have no doubt that if you gave standardized IQ tests to every reader of Hacker News, you would not see much of an overlap with the bell curve of the American, or even world, population.

The people coming here are generally very technical, very smart (I'm defining this as at least one standard deviation above the norm, so 115 and higher IQs).

Most people aren't going to build their own enormous media server from used enterprise gear. Most people aren't going to build their own home surveillance system even though the hardware and open-source software out there exists. I could list another hundred projects for which we have specialized "consumer-friendly" devices.

Watch the YouTube video where Jordan Peterson talks about working with an 85 IQ individual and how it was a struggle for that person to consistently and accurately fold a sheet of paper into thirds and put it into an envelope. And now realize that for many of you, that's someone for which you're developing.

> The Internet now treats everyone using it like we're all brainless idiots to be injected with dopamine. Like we're not smart enough to choose how to consume it.

Have we generally proven otherwise?

The problem is that most people are. Look at the statistical breakdown of intelligence. Half of the world is 100 IQ and below. Half. Hell, 16% of all people are 85 IQ and below. That's over a billion people worldwide.

Any technology that you want to see enormous widespread adoption of, has to be usable by some of those people.

And believe me, no business-minded person wants to throw a billion potential customers to the wayside. They want to simplify the technology to capture those dollars.

Sorry if this offends sensibilities of anyone, but I guarantee you if you could get inside the heads of a lot of these billionaires and multi-millionaires, you'd hear the same shit I'm telling you right here. The reason I know this is because my Uncle is one of those people. Smart. Ruthless. And has absolutely no qualms about exploiting people of lower intelligence.

Don't think for a minute that the SV crowd is any different.

Goodness, you and the OP have really hit the nail on the head.

I spent the last 5 years quite happy at a simple dev shop. Then they hired a bunch of people, paid them more than me, and allowed them complete free reign to write what ever kind of code on some brand new projects. After a year of this, one person quit mid-project, another died mysteriously (covid?), and then another just recently quit. My boss expected me to just pick up the slack on these three projects that were written in the most arcane and buffoon way possible. I basically just said no, and left. I had already been driven half insane over the year just trying to figure out the crap they were committing.

This was in the spring and I'm still unemployed. I'm mostly Java and some older languages, and was really looking forward to jumping onto python and django. But it's been a struggle really learning django good enough to interview for without any good experience behind me. I have a family and on top of covid issues and this year in general, sitting at home trying to learn a giant web framework is not very motivating when I could be spending it with my family (which were already severely affected by all my time I "wasted" going above and beyond at my last job for 5 years).

I just recently did an interview for a small django shop that didn't even test on django. Instead, I was given a half page of python code and asked to "fix" the problems. No requirements or tips. There was a syntax error, a bug, and a few cosmetic things.. I managed to find 8. All this while being stared at over Zoom. I was informed that "junior" devs can find 10-12 "problems", but they only hire those that "find" more than 18 (apparently 25 is the max).

I still got the code and look at it every now and then. Half a page of code that does nothing but take a list of "item" objects and print out an aggregate of a total price by item type. A loop over nothing but:

  total[item.type] += item.cost

Of course there was some random data mangling above and below that, but that is the logic. Find the problems?! Should I raise errors or use a namedtuple or defaultdict? Half a page of code that could have just been a Select/Group by in SQL. No senor dev should be summing data like that. Sadly they said the code was from "production". They have "interviewed" over a hundred people in this year and hired only 1. All this for $45/hr 1099.

In the beginning, I briefly thought I could use my savings to start some kind of business, not sure what. But I've blown most of that just paying the bills while I "interview".

I used to love software and would dream and breath it. But I've never been so unmotivated over these last 6 months. I'm debating on just applying at a local small town municipal IT job for $30 an hour. Giving up basically. Just show up at 8am, answer the phone mindlessly for IT "problems" and tell people to try restarting their windows machine (at least 3 times minimum) and screw of playing solitaire the rest of the day. But it seems so pointless and waste of a life. But even that doesn't motivate me, but just makes me more depressed and even slower going.

Your comment highlights so many problems with our industry so poignantly, and at the same time is an extremely sad personal story...

I don’t actually know what to say as encouragement but I do feel for you :(

I was in a similar position to you on your last job, and after looking at every single company I could think of in my city (London) I ended up finding just one that seemed good in all possible counts...

Maybe there is a company out there that you’d like to work for? My approach was to filter by the product, and only look into working for companies whose end product I thought was good and interesting.

Best of luck and don’t give up!!

The Django shop test had nothing to do with your skills as a developer, and everything to do with language gatekeeping. It's just overcompensating for boring work and poor pay. It's no different to how the gaming industry exploits workers.

Don't give up. And for the love of all that is holy, don't work municipal IT, it would only demotivate you further. Keep at it, improve your skills and keep applying.

Is there anything you're passionate about? If so, why not spin up a website on that subject?

Django is pretty easy to learn. Model view Etc in Python instead of Ruby. Not rocket science.

Building something you care about will force you to dig into the details and become enough of an expert in a few weeks to get in the door somewhere where your skills will be appreciated.

That probably won't be Facebook, Google, or the next Silicon Valley unicorn. But there are a lot of other places and other Industries where interesting things are going on.

Good luck

Yeah, this echoes my experiences interviewing. Spent 6 months looking a few years back

It’s just incredibly demotivating to hear all the self-professed hot shots shit all over your past 10 years of delivering unambiguous value because you don’t know that [] === ‘’ is true in Javascript.

I finally found a position that’s financially great, but it’s more in spite of their dev process than thanks to it.

I haven’t had anything quite that bad, but I’ve had an interview where I was being tested on SQL, but the interviewer didn’t actually know the SQL keywords I used.

It happens. It’s a strong sign you want to work elsewhere anyway. The problem is one of finding the right Elsewhere.

> Giving up basically. Just show up at 8am, answer the phone mindlessly for IT "problems" and tell people to try restarting their windows machine (at least 3 times minimum) and screw of playing solitaire the rest of the day.

You have a very holier-than-thou view of small scale IT work. There are a whole lot of problems to be solved at this level, many of them created by the people working the positions you're currently interviewing for.

It's a numbers game, learn from mistakes and continue to the next interview. Eventually you have seen them all and you will pass the gatekeeping. Steel in forged in fire. After hundreds of bullshit interviews, nothing surprises me anymore and it's just all automatic process now.

This sounds exactly like a job I interviewed for a couple of years ago, a Django shop. I'm skilled at Python but the questions were ridiculous. I feel like you probably dodged a bullet and don't want to work there anyway.

I’m spending some time in Constantinople right now, if anyone wants to spend some time lamenting the fall of the Western Roman Empire together. That is, after the weekend plague curfew. Email transitively in profile.

How is the plague of Justinian treating you so far?

How did you know my name? Also, you’ve come a bit late.

> Google kills products that make customers and engineers happy and doesn't spend enough effort to make them better.

Small correction here: The non-paying users are not their customers. So it’s understandable that they kill all products where they don’t have any ROI. As another user here put it recently: Ads are like a hose where money pours out. And they’re desperately looking for another such hose.

And yet, there are still so much value one can create with software... A lot of it is even outside of the reach of the spy and divisiveness as a service companies, it's the kind of thing that if they touch it, they destroy it.

The late years of Rome is actually a great metaphor. We are suffering from human capital loss, and have signs of institutional collapse. A shock can destroy our institutions and throw us into something like the early Medieval period: much less wealthy at the beginning, and messy too, but fast growing and with a much higher ceiling than the previous age.

This doesn't feel new to me. I entered the industry in the mid-90's, so I have no reference from before that period. But I'll cherrypick a few items:

> Microsoft is building VSCode for the cloud. Few people own desktop computers anymore.

I heard the same thing in my early years about these newfangled HTML email clients. Email should be consumed using Pine or Eudora, just as God intended.

> People job hop for promos, raises, and resume fodder, not because there's an actual interest in the product.

Oh yeah. I did this for the first 15 years of my career. It was faster to get a promo by changing jobs. I didn't care where I worked, as long as the company name had prestige. I kind of still do, but I transfer internally. Same effect, but I get to keep my vacation accrual.

> Companies are trying to hire more outside of California

In my early years I worked for a company that used Russian developers for some non-critical parts because it was cheaper to fly someone out there once a quarter than it was to hire the staff in CA. This was before the massive rise of outsourcing in India. See the Dilbert cartoons about Elbonia for more.

> The web is turning into shit.

Heard the same argument about MySpace. And Geocities before that. And AOL before that.

Geez that was bleak. And I don't think you're wrong, either.

There are so many pieces of software that are latency dependent. Aspects of them can be in the cloud, but I doubt we will ever truly lose track of bare metal. Right now I'm hoping for a real time OS that has enough of the common apps that most people are happy, but that satisfies the speed requirements necessary for digital audio, lighting controls, and other tangible and interactive behaviors. Hopefully the natural world's fundamental interactivity and tactility in real time is something that we all consider going forward enough that the cloud / thin client model becomes only a part of our human system.

Gaming is a pretty latency sensitive application, and yet Amazon is working hard on moving that to the cloud: https://www.amazon.com/luna/landing-page

While there are some things that are even more latency sensitive, I think that could become a very small niche in the near future.

The technology to "move gaming to the cloud" was around at least a decade ago, when Gaikai started doing zero-day "play this game in your browser via Java". It worked well then, it works well today (HN had a article several months back by a sort of follow-up to the Gaikai concept). The problem is convincing gaming companies to actually use it(they have some good reasons to be concerned).

I think the real issue remains latency. Gaming companies are using it but the users don’t want it. Try playing call of duty or doom over the network and it’s going to be a terrible experience.

Stadia disagrees (Doom is actually on there).

You can say what you want about the business model or Google itself, but the UX is impeccable.

I am periodically in disbelief when I think that what I’m playing on the screen is being streamed by a server.

Doom is on there, but I wouldn't say the experience is great. I've played it both locally and on Stadia and while a quality-inelastic buyer might be OK with it there it just doesn't match up for a serious/invested consumer. And that's totally fine; not all things have to be for all people. But it's a worse experience once you know what you're looking for, and will stay so.

And, as a TO for fighting games, there's no substitute there for the barest minimum latency possible (to the point where older games legitimately feel better because you can play them over analog connections to a CRT).

> UX iS iMpEcCaBlE

https://img-9gag-fun.9cache.com/photo/aDgDE8x_460svvp9.webm

That looks like Borderlands 3... I've beaten the game and never had such a problem.

I'll grant: it's impeccable under the right network conditions.

The optimist in me says that cloud gaming will expand the category, not move it from local to cloud. Cloud gaming can't beat the laws of physics, and cloud hardware isn't free. So action/fps games will always have a home on local hardware.

But cloud games have an advantage too and maybe we'll see new kinds of games/experiences that are not possible with local PCs.

I've tried out Google's beta with assasin's creed, and the results weren't promising. I had gigabit fiber and while it worked pretty well for the most part, I still got quite a few gamebreaking lag spikes and fps slowdowns. It stuttered like a youtube video: freezing, then resuming in 144p quality which made the in game text completely unreadable on my screen. And that's on a beefy internet connection, imagine the situation for the median American or someone reliant on a mobile data plan. People can scarcely tolerate playing a game with high ping as it is.

People can scarcely tolerate playing a game with high ping as it is.

People who classify themselves as "hardcore gamers" perhaps, but those people will also drop upwards of $1000 on a graphics card. They're not the majority in gaming. Most people will happily play a game even if there are a few problems if it's cheap and entertaining.

Anyone will throw a controller across the room after putting a whole clip into someone taking no damage, watching them rubber band across the map, and pistol whip you in the head. That's your life on a crappy connection in most FPS games which are some of the best sellers every year.

I've ran my own cloud gaming VM on AWS and with an enterprise connection (1ms latency to the AWS datacenter) it's been good enough to play multiplayer shooting games and maintain a positive kill/death ratio against those who play locally.

Maybe the problem isn't that it's impossible per-se but that the consumer-grade services underprovision their machines/bandwidth, or that consumer-grade gigabit fiber isn't actually that good (or problems in the "last mile" like using Wi-Fi instead of good old Ethernet)?

> Employees aren't excited by what they do: People job hop for promos, raises, and resume fodder, not because there's an actual interest in the product.

Welcome to the other 99% of humanity. Work sucks, but someone has to do it.

It can all be summed up in two words: corporate greed.

>Few people own desktop computers anymore. Apple is locking down MacOS to signed binaries. Soon we'll all have thin clients and will write software in the cloud for the cloud. Few will be able to learn how to engineer at home anymore - you'll have to go to a university or trade school. Building a simple website will require JS package management, kubernetes, and AWS. The "free tier" will fall away.

Building a simple website is easier and cheaper now than it ever was.

When I was a kid, I dreamed of being able to have my own domain name and an account on a shared server where I could run CGI scripts. Now most people who can afford a laptop can afford to host their own website on a virtual server.

Spot on. You nailed it poetically

> Companies are trying to hire more outside of California in order to lower wages across the board.

California isn't the industry...

I need to create several projects or companies to reverse these trends. Your insight is appreciated.

Any successful projects you can show so everybody knows you are as effective as you want to be ?

You can arbitrage AWS with your own servers as long as network costs don't eat you. Run say a TravisCI type service on your own hardware?

The big lockin is network pricing. No egress fees when their data is already in S3. You have to run where the consumer has their data or you eat that network transfer as overhead for your SAAS.

I see GitHub slowly charging for egress downloads as the next lockin. You "save" by paying for builds on Azure.

Welcome to MBA-run tech.

Thank you for the truth glummy glooma.

The moment the MBA types discovered there were plenty of VC money to get if you exploit the socially inept nerds it was game over.

I guess for some people the social part is either painful or difficult so for them it's easier to just become masters at the leetcode game.

"Few people own desktop computers anymore."

I will own a desktop computer until the day I die.

Excellent post too.

I mean - desktop processor sales an insanely strong, AMD can't make enough.

Gamers will buy every RTX3xxx card that hits the shelf and every heavy duty programmer I know has at least a fast desktop and often server racks.

I don't think we need to worry about the future of the desktop just yet.

What is a heavy duty programmer?

Someone with an expensive desktop and possibly even server racks.

Once who can spew thousands of lines of code each week :)

I was hired in at a company a couple years ago. My soon-to-be manager asked, "What should I order you for a computer?"

I responded with, "Whatever you guys buy - Dell, Lenovo. A 14" laptop and an external monitor would be fine."

He torted, "I've never seen anyone be productive on a laptop. Nobody here uses laptops. I'll get you a desktop with 2 monitors."

After I started, I notice he did almost all of his work from a conference room, using a Surface Book remoted into a desktop in his office.

Side note: In an open office, people quickly learn to cheat and claim unused spaces like conference rooms for working undisturbed.

We have people who claim to be engineers but start thorwing a fit when they're required to do unit testing. (Aka the anti-vaxxers of software engineering)

Reply on news.ycombinator.com