This is a solid take, the only thing I'll quibble with is the idea that VC funding is going to dry up. Sure, it will be less for a while, and diligence will be much higher, but there's too much capital sloshing around that will continue looking for big returns, and where else can they get it? Theoretically you might find it in the developing world, but it's hard to make a convincing argument that the structural and corruption problems there are going to be fixed more readily than traditional VC funds who have far fewer stakeholders to wrangle when trying to recalibrate their structure and strategy.
Relatedly, i think about being overpaid. But everytime i end up watching folks work in non software roles, all i can think of is how badly want to automate everything they are doing. Ive built a lot of worthless stuff in my career, but every time i do build something useful, its automating some value producing activity that used to require a person. Now it can be done 10-1000x as fast, for free, indefinitely.
Its easy to both under and oversell tech but i think ultimately when done well, tech produced value and experimentation is relatively cheap. Im not sure if we have passed our heyday but to me it feels there is still so much left to do and a lot of it is still pretty basic crud and workflow automation stuff.
Everyone looks like a sophisticated investor during a bull market. Now it looks like the emperor has no clothes. These investors are going to have a hard time justifying their dumb money investments.
Personally, I'm most concerned that index funds might hide bad behavior - both on the part of the investee (being overvalued), and non-index investors.
> And much like Twitter, Reddit has remained valuable not because of any amount of venture investment, but because of millions of users that choose to post there for free.
That's the money quote from the article. It's also true of stack overflow, and flickr and Tumblr. That's what Elon and spez miss. Why do they miss this? Is it arrogance or a genuine blind spot?
As long as they're getting enough user engagement, which drives ad revenue, they don't care.
Apart from the Reddit protests (which they will just force an end to by replacing mods) has there been any objective evidence of engagement or number of active users going down on Twitter or Reddit so far?
I thought Twitter user numbers had dropped at least some. I mean even Elon acknowledged that ad revenue was down and user numbers have to affect ad revenue.
I'd heard that it was because certain advertising partners had pulled out of Twitter for various reasons.
You're probably right that user numbers could be having an impact there as well, but I haven't seen any sources that are showing direct evidence that they have dropped
Until 2021 or so, even a really crappy startup could raise 8 digits with relative ease. Then all of a sudden those investors realized that the economy was bad enough that those companies may never make it big enough to justify the investment.
I don't think this recession will mean the end of crappy tech at scale, but it will make it really hard to justify certain strategies such as burning mountains of cash to build a (micro)blogging app, to name an example.
Yes, this is 100% the most accurate take on what's going on with Reddit and a number of other companies. It's common sense to many I'm sure, but this is well expressed.
Reddit isn't a tech company. They sold themselves to Conde Nast back in 2006, which is a news company. Would you call NYTimes a tech company? News people hate tech. They don't want to be part of the tech culture that seeded Reddit's community. Their primary interest is to extract rent and manipulate public opinion. So don't tell me that's tech.
Finally someone has mentioned this in the entire Reddit saga. Reddit has basically been PE-owned for more than a decade now (technically a family office but Advance Publications operates like a PE).
I get in conversations about this and I flail for examples of good things tech has built. Dear lazyweb: what good things has big tech built in the last 20 years? Let's stipulate that the smartphone was good and that I already know about it.
This is crazy. The electronics and telecommunications industries have just kept trucking on and improving hardware and connectivity by orders of magnitude. Meanwhile web technologies have provided the software complement to smartphones that let people manage their whole digital lives online, at the same time as hundreds of millions of people come online for the first time.
Are you defining 'tech' as purely social media companies? Or demanding entirely new classes of product? There has been an incredible amount of iterative improvement.
I think op is referring to what I call “capital T tech”, or alternatively, “big internet tech”.
There’s plenty of technological improvements within the past 20 years, we just tend to ignore them the further they fall outside of internet-adjacent technology. As incredible as it is that solar panels keep getting cheaper and more efficient, the real miracle is that no one gives a shit about them. What captures people’s interests is internet-adjacent stuff, and that’s especially true if said thing can be compared to as “a new internet”. Hence why when people say they want to “work in STEM” they implicitly are referring only to programming most of the time, and also implicitly referring to being a web dev or phone dev, and not some guy that writes software for power plants or whatever. That’s “capital T tech”.
I also refer to it as “zeitgeist tech” because I think we’re currently in a phase where said tech has captured the public imagination, but I think we’re also seeing it’s end, the decline of all major social networking players being the most visible side-effect (I don’t consider TikTok to be a social network). Or, you know, a “tech reckoning”.
Kind of hope we return to 2008 with Occupy Wall St and more environmental consciousness personally. Maybe it was because I was 18 at the time, but man, there was so much more optimism.
I build infrastructure (raised beds, drainage, chicken runs, composting areas) for our garden and even with that minor engineering get a good look at the level of manufacturing sophistication required to do basic things well, down to the accuracy of our mitre saw (which doesn't reset to 90 degrees very well) or staples that flatten instead of penetrating wood, or the timber I buy that isn't extremely straight. The lumber yard I go to often discards 40% of the timber in a pallet when going through it with me because it's warped or knotted or cracked and can't be used for my needs.
The manufacturing sophistication required to build a reusable space rocket is astonishing and frankly incomprehensible. It'd never be achieved in my country. We have 'barely good enough' materials for the simple stuff, and sophistication of manufacturing stands on the shoulders of layers and layers and layers of other advanced tools and processes and measurements and minute tolerances.
Even building a shed, based on the instructional videos I see on YouTube, in the US is so simple and requires so few corrections for i.e inaccuratly squared plywood. But people shrug off amazing improvements like watches that detect heart palpitations.
I know this is poor educate on HN, but surely you can’t be serious? Sure, there’s a lot of paper value, but there are a ton of good things.
I’ll throw out a softball - learning and self education. The tech field loves to be open and talk about details of what they build. There’s an insane amount of information that’s readily available to nearly anyone.
Let’s throw out a more controversial one - smart thermostats. They’re far over hyped, but I save so much energy with a smart thermostat. Being able to set an away function remotely is massive for me (I tend to forget until we hit the road).
Tech is insanely important to the efficiency and safety of your vehicle. Modern automotive engineering uses insane amounts of simulation and digital analysis. Not too long ago, a lot of this work had to be done on real, physical objects. Dozens/hundreds of cars scrapped. You do still need to do really world physical testing, but involves less waste. That doesn’t even get into the digital controls that allow tiny engines to output insane amounts of power all while getting much better fuel economy.
Telecommunications. 20 years ago, my dad would have to physically leave the office and drive 30+ miles to meetings. The amount of fuel that uses is insane. Now you punch a button on your laptop and use a dozen or so watts of energy.
Smart thermostats are also important for the grid, not only do they encourage setting a schedule (and adapting to it), they run fans to circulate in case the ac isn't needed, and finally they can use eco settings for load balancing and saving power.
iTunes was progressive. AppleTV is garbage. Apple makes toys for adults; nothing we can't live without.
Solar is accessible to the masses.
Cloudflare seems useful, until you need them. Never trust a company headed by a lawyer.
Netflix makes good stuff, but the bad softcore lesbian porn in every show is getting tiring. (Protip: an obligatory lesbian romance does not make boring female characters interesting!). Wokeness and all, TV is better than the cheesy 90s-00s.
TVs themselves are cheaper, but that's not the result of FAANG. Most computer parts are cheaper than in the 90s too.
Software is sortof better. The nightmare of fucking with IRQ settings to get SoundBlaster to work were behind us, but have since been replaced now replaced with virtual environments, cloud bullshit and virtual environments within cloud bullshit. We had a decent run of software that "just worked" until recently.
Google Search is garbage now. Used to be revolutionary.
AI is cool if you like motel art and corporate literature. And ignore the misery of out-of-work creatives.
Your washing machine is the same as it was in the 70s, only now the basin of water is turned sideways and a bunch of expensive computer chips are stacked around it. Laundry reinvented, now with more mildew and electrical problems!
Microwaves are exactly the same.
Food doesn't cook itself, but you can buy a fridge for $10k that shows you what's inside-- from the outside. Glass, reinvented. You can access it from Bulgaria though!
Flying cars? Still working on the self-driving part.
We don't live longer, and life is no easier. The promises of Big Tech are underwhelming.
I'm a hardware engineer, so my point of view is biased a bit towards atoms and what they can do for me:
Cell data networks and Starlink are pretty awesome. Mostly-good connectivity just about anywhere is a huge advance. wifi is great.
The ubiquity of brushless motors, controllers, and (for some applications) batteries means that my fridge, HVAC, and power tools are loads better than they used to be.
More software-y:
Remote meeting software means I rarely go on business trips any more, which is nice because flying is way worse than it used to be.
Google Maps / Apple Maps / OSM have made it dramatically easier to get around, especially unfamiliar places.
After two or three decades of nonsense, I finally have a decent no-frills printer driver that works.
We have flying cars, Cessna makes them. We’ve had them since them 1950’s. I feel like there is room for innovation in the general aviation space but the FAA has very draconian and discriminatory rules for who can fly even as a private pilot.
My point here is that innovation in tech is dependent on how well they can squeeze it into the current regulatory scene in addition to find market fit, and produce a viable product.
I’m a current student pilot in FAA-land, soon to be a private pilot.
I actually really don’t think the training requirements are very draconian. 40 hours, a few maneuvers, some knowledge about some stuff. Less than driving in some states. The DPE system is mildly corrupt, but that could be fixed by just adding a bunch more of them.
What is driving prices up (and normal people out of flying Cessna-cars) is certification and maintenance requirements. You can’t change the sunshade in your Cessna without an STC. It’s nearly impossible to certify a new engine for a plane, so all those 172’s are using engines designed in the 60s.
There was a story posted here a while back about a company who figured out automotive-marine engines work just fine for GA. Much better, even - more power, plenty reliable, better fuel economy. And the FAA just wouldn’t let them certify it. As in, they had investors who would pay certification costs and the FAA just wouldn’t tell them how, other than that everything they tried was wrong. We’re talking “you can’t put an experimental engine in a certified airplane, but certification of an engine must be done in a certified airplane” levels of bureaucracy.
If you want GA to get more accessible, focus on making engine certification possible and a generally more accessible R&D cert process.
> I actually really don’t think the training requirements are very draconian. 40 hours, a few maneuvers, some knowledge about some stuff. Less than driving in some states.
I am in the process of obtaining my private pilots license as well, the goal of eventually getting my IR and CFI. The logging of 40 hours basic competency is the least troubling part. I was referring to getting a medical certificate to be clear to fly.
High blood pressure? Potential heart problems? Sleep apnea? I guess that’s okay, just makes sure you are taking 2-3 medications we are okay with and use a cpap. Ever filled an Adderall script? You are a danger to the skies and you would only be destined to fly a plane into the ground - no fly, no go. The FAA only just started the process of getting Wellbutrin approved, which has been a side-effect free, well-tolerated medication for a number of years.
The 1500 hours before you can fly for a commercial airline is sort of strange[0], considering the event the caused that regulation wouldn’t have prevented the pilot involved from flying[1], and I’m sure I would have a stronger opinion on that if getting my ATP was a goal.
I was interested in building experimental aircraft, playing with different propulsion systems. I am not sure I am anymore given the landscape in which one must operate in.
One take I’ve heard on the certification mess is that the FAA wants small plane aviation to stay small and niche and maybe even die off for any purpose other than training pilots for commercial.
Basically the argument is that the FAA sees it as a hassle and a cost center and really only cares about commercial and government aviation.
This is pretty much my take too. I think the FAA knows that if GA becomes accessible, people will start using it, and they aren’t ready for an influx. Which is… understandable, kind of.
What’s tech? I recently had surgery and the surgeon used a robot. I was an out patient the same day and had basically no pain without medication. Compared to a similar surgery I had roughly twenty years ago this experience was like science fiction. Does that count?
This.
The appropriation of “tech” to mean web-first business has gone to our collective heads. Real tech is hard, expensive and mostly progressed by the interplay of science and blue-sky inventions. Not really something easily adaptable to the venture-startup model. We’ve seen a couple of attempts at funding fusion/fission energy along the same lines as SaaS, but honestly, it seems more like a tech-bro indulgence than real technological progress.
Meanwhile the conventional tech industries such as medical equipment, have seen decent progress at the incremental pace of the respective industries. Adding a voice interface to your car, or allowing remote surgery over the net is hardly something we should credit the “tech” companies for.
After 20-some years in the “tech” industry, hoping to get-rich-fast, and THEN do the real tech stuff with that capital, I’m feeling like an extremely well-prepared failure. By now I would do nearly anything for access to a physics lab and funding from a ROI tolerant source. Only problem is that academic funding agencies have been mimicking the “successful” VC model for decades as well…
Big tech has built a lot of good things, even if they have subsequently ruined them. Google search was initially a good thing, before it became a platform primarily for ads and thus SEO. Amazon was good for a while, before they became infested with counterfeits, fake reviews, etc. Netflix and its imitators have provided many hours of enjoyment for millions of people, and I would say still do despite their recent missteps.
Things that are still good? I'd say Maps and its competitors would be near the top of my own list. Despite the well known issues with both the platforms and even more so the apps, both iOS and Android also qualify because they provide the basis for connection to other non-big-tech gems. Wikipedia is that for many, though Etymology Online might be my personal favorite. It's not all bad. Just mostly.
Also, as several others have already pointed out, there's plenty of good outside of internet-oriented FAANG stuff.
I mean, what's your definition of "good"? Is it "valuable as a business"? "Helpful to consumers?" "Good for the environment?"
Tech has produced lots of useful things, but many of them are simply not good businesses, and many of the ones that are good businesses are run in a way that is so odious that it makes it hard for anyone to benefit.
Reddit is a great platform for lots of things, but making profit's not one of them. Amazon has revolutionized how people shop and consume all sorts of products but at massive economic and environmental cost. Discord and Slack and Zoom have made communication with friends/family/colleagues much much easier but struggle to make money.
Craigslist and Wikipedia are both pretty great, though. Don't see much downside to either of them.
I'm just going to say it, the ai revolution is amazing, and Bing AI is crazy good when it comes to doing technical searches.
Also for natural language questions "does this sound weird in X language?"
Encryption and that we have zero knowledge day to day, wouldn't be ever possible without tech and precisely Whatsapp who made E2EE mainstream.
(There are huge caveats though, like unencrypted backups and so forth)
But i agree, the way the venture capitalists work in the U.S. is just ridiculous.
If reddit didn't exist, forums would still out there, but by giving infinite money to one company, it can centralize everything, then kill the entire community trying to seek profit at any cost.
Going back to the old times isn't easy, especially when users are used to these endless free centralized platforms, which no body can actually compete with, because they have Infinite money.
Life has improved infinitely for millions of people with disabilities. Just 20 years ago there were a lot of everyday tasks that someone who's totally blind or totally deaf couldn't do without assistance. Today technology has enabled them to do those things totally independently.
20 years ago, you could have access to the world's information...as long as you were in front of a computer. Today, you can have access to the world's information anywhere you want, because mobile networks, wifi, and cheap, fast satellite are now ubiquitous and cover the whole world.
That access to the Internet from everywhere is what makes things like Uber and Lyft possible. And no, those aren't just fancy taxis - in most major cities, Uber and Lyft are doing 5x as many rides as taxis ever did. They're enabling people to not own a car. They're providing new mobility for people who are blind or have epilepsy and can't drive. They're saving lives, as fewer people drive drunk when there's a convenient, safe alternative.
Aside from the many software application examples on this thread, have you also considered how they also pushed innovation in the semiconductor industry that in turn contributed to non-obvious advancements in non-Big Tech sectors, from particle physics to drug discoveries?
It was the zero interest money that effectively removed price as a signal from capitalism. Capitalism relies on negative feedback to correct errant behavior, usually in the form of lower stock prices, or bankruptcy.
Combine the end of zero interest money with the retirement of the baby boomers, AND the people Long Covid took out of the work force, and we're looking at a decade or more of 10% interest rates.
Fiscal Discipline is coming back, the physical economy (an odd, but precise, phrase used by Lyndon LaRouche) always lurked in the background, masked by the free money.
I agree with your larger point, but do you have any #'s of "long covid" workforce removals? My gut says it's in the 1/10'th of a %, but I don't know either.
also, i hope this is a good thing. Hopefully people will spend their resources on actual tangible problems compared to the massive, absolutely insane money burning machine that is the VC based startup environment.
I have trouble taking people seriously who don’t understand basic economic constraints. The company loses about $170m/year. Without VC money it would have either been dead, sold, or a shadow of what it is.
To be clear, this would have been an improvement. Reddit absolutely does not need the headcount it has. It provides no value to either users or to the company. It should have never grown above 50 people.
50 is probably too few. Because it has an international reach, something closer to 200 people is probably necessary. Of course, having 2,000 workers is far too many.
Amongst the UK-oriented subreddits, there is a phenomenon called the Forbidden Meatball Problem. There is a dish associated with the West Country of England that you can't discuss on Reddit because Americans don't know that its name can mean something other than a homophobic slur. The moderators of those subreddits have raised the issue of banning people for discussing food and received no adequate response.
Reddit enforces its own moderation policies on subreddits, regardless of the wishes of the moderators or communities, which can only be done effectively if the company has awareness of more social and cultural issues than the ones in Silicon Valley.
My other similar example is the same word, but used in the sense of cigarette and combined with Irish idioms for "fun" ('craic' pronounced crack) and "borrowing or begging" ('bumming').
why are you reducing the reddit situation to one number instead of the nuance that should be?
reddit bought up dubstep for a number that must be north of the 20million that dubstep raised. reddit increased headcount to 2000 when it certainly doesn't need it. reddit could have charged the api at a reasonable fee, a fee that was passed directly to consumers, serve ads to the api, require reddit gold to use or whatever that would have been more amicable than what happened.
If it loses $170m a year without any real hope of turning that around very soon then this "business" should die, and it should die quickly because it is a money burning machine and not a business.
We are seeing a contraction in the spending of magic dream money and the the process of pruning shitty businesses like reddit or Twitter is economically healthy.
It's only losing $170m/year due to the way it's being run. Remove the bad investments, empower the community, turn it into a non-profit and you've got a sustainable business.
Next you say they should move to a 4-day workweek. Non-profit is a non-starter for the same reason; some admittedly dumb people with a lot of money (for one reason or another) want that money to become even more money come hell or high water.
I'd reckon a vast majority of people have not, and some major subset of those don't want to; why do you say this like it's bad? People opine about things they don't have firsthand experience in all the time.
Would it be alright with you if an economist weighed in on the Fed’s decision to hold interest rates, or are they not qualified to do so because they’ve never sat on the Fed’s board of governors?
This is amazing. Especially this: "In Huffman’s eyes — just like Elon Musk — apps should pay Reddit to give Reddit traffic and make Reddit more money, because said apps make a profit themselves."
They fundamentally do not understand the myriad of reasons Reddit wants to end third party apps.
