blog: http://www.davidralbrecht.com/
gmail: albrecht.dr
live: Oakland, CA
- I tried to automate a big chunk of property management (mostly commercial) over the past few years. The main thing I realized is that it's mostly a people business, and AI (or computers generally) are never going to stand over a vendor's shoulder and keep them honest (e.g. make sure they sweep up or don't scuff the walls), or have a difficult conversation about late rent with a tenant, or show up after hours when a pipe breaks, if only to show face with a tenant.
There are definitely workflow and process elements that can be automated. But if wealth management is any indication, there are a lot of people willing to pay a premium for having a person involved. Not sure why real estate would be different.
The wildcard in all this is the NAR court decision. If buyers have to pay for their own representation, that might make them shop around a little more.
- I think these aren't more popular because actually, there are things that genuinely suck about small business ownership. Offhand: stress, uncertainty/having to make payroll, having a significant % of your net worth tied up in a single, undiversified investment that's near-impossible to sell.
Given the choice, I think a normal cash-paying salary job (maybe with bonus) is vastly preferable to most people.
- A perspective I've found helpful is to think about a fund as a business, which it is.
Let's assume a hypothetical case of running a $100 million tech investment fund. The first 10-11% of return should go straight to investors (not the manager), for the simple reason that 7-8% is available in the stock market, and unlike the public markets, tech funds are illiquid--unlike investing in SPY, you can't wake up one day and trade out of your interest in a fund like this. You're locked in. That shifts the return expectation upward.
So assuming the manager's cost of equity (what investors demand) is around 10-11%, maybe a great manager can get the return number to 15-16%. That's a pretty good return on the fund but it's only $5 million in absolute gross returns (10->15% on $100 million). Considering the manager might get only 10-20% of this (the rest goes to investors), it's just a lot of work to earn $1 million in performance-based comp, over 2-3 years of active work and perhaps a decade of full fund life. These things are also typically run by teams (several partners) so the returns are split.
The point is that $100 million funds just aren't making their managers rich. The two outcomes of this, which you see over and over, are (1) for managers to try to manage much bigger funds ($500+ million) or (2) big management fees of 2%/year or more, which significantly erode returns.
The net result being, VC is a very hard business that almost always delivers substandard returns to its investors, after long lock-ins with very little liquidity. It's very tough and I'm not surprised to see these guys shutting down.
- So I think it's worth discussing this a bit.
Maybe there was a time when this wasn't true. I spent a lot of the 2010s working in San Francisco tech jobs. It was quite enjoyable and I just got back from a guy's wedding who I worked with closely 2012-2015. Three different startups, one mine, other two as an employee. All three crashed and burned.
I think there are three possibilities.
One is that tech is somehow different/exceptional and bound to stay that way forever. I doubt it.
The second is that we were always lying to ourselves, it was always just "a job", but we were all young, stupid, and naive. This feels too cynical.
The third, which I feel is most accurate these days, is that tech was different, but now is a more mature industry, and is, as you put it "just some shit corporate world". Maybe there was a time when it was genuinely true that people got "unlimited vacation", that "titles don't matter" and that the CEO ate with the hoi polloi. But I think those days have passed. Everything's different now. The people coming into this industry today are 4.0 GPA high school kids, not misfits tinkering with computers in their basement. Competitive parents no longer feel they have to justify why their kid is going into tech rather than law, finance, or medicine. Salaries have increased 3-5x (!!). Tech influences elections, mints billionaires, and controls many facets of American life.
The problem is that the attitudes haven't kept up. These days, Google, facebook, etc are just standard American megacorps. They are some of the most valuable companies in the world with huge lobbying budgets, and tremendous pressure to deliver shareholder value.
The simple fact is that margins always get compressed as industries mature. Google has been under greater and greater margin pressure over the last decade as Apple demands higher payments to be their default search engine, OpenAI starts to steal share, and more people head directly to Amazon for search results. Google of course is going to blame the pandemic for this, but the actual issue is long-term erosion of margin. It's hard to see how any of this is going to reverse course over time.
- I think this portends a larger culture shift in Silicon Valley tech that, in my opinion, cannot come quickly enough.
Here you've got a guy, 18 years at Google, probably earning somewhere between 500k-1mil per year, probably $5-10 million in his Schwab account without breaking a sweat. With a little blurb at the top of his blog about "How to Leader", feeling the need to explain whether any of this is "fair" or why it's ok that "Google did this to you".
Honestly, as an industry--we need to grow the fuck up. Using the wrong part of speech or talking about what is or isn't "fair" are things I do with my three-year old when she's throwing a tantrum. Not something I expect from an emotionally mature professional in his 40s or 50s who's likely earning a million/year or more. Google is a trillion-dollar, global multinational with shareholders, and a board, and a stock price. If you don't deliver, you will be cut, period.
It's not that I even blame this author--I think this post shows a lot of maturity and self-awareness. It's the broader culture of unseemly whining by some of the most mature, intelligent, and professionally successful individuals that needs to die in a fire.
- Bit of unsolicited advice. It's very possible you've had a great career as a programmer. But for better or worse, high schools are academic environments where you can't write "elligable" and expect to be taken seriously.
If you aren't willing to correct this, this may not be the path for you.
- Keep in mind this has to include downside protection to cover the very real risk of someone initiating a suit in case they screw up (which I'm sure happens from time to time).
I think overall, what a lot of this entire discussion misses is that 90, maybe 99% of transactions are pretty straightforward, but the 1% where things go haywire, might cost 10, even 100x as much in wasted time, money, etc. Real estate brokerage is a regulated professional service which means by law, there's a minimum standard of care owed to anyone who uses these peoples' services. This costs money to provide which ultimately means a lot of the costs of the problem cases end up socialized/spread out, over everyone.
This feels quite unfair if you've got your act together and everything goes smoothly, but it seems hard to know upfront which ones will be the hard cases, or not.
- Not really. 20-30% effective on income including CA's 9.4% at ~40k, plus 10-20k in property taxes, 10%+ in sales taxes, excise taxes on everything from gasoline to alcohol, and various "fees" that are really taxes including auto registration, building permit fees, DMV charges, ride share fees, motor oil and battery disposal charges. Also, some cities (e.g. NYC) have a personal income tax.
Obviously you're right that marginal != effective, but Americans really do pay a LOT in taxes, when you add it up, for a relatively middling level of services (e.g. no public healthcare, no subsidized daycares, marginal-quality public schools)
- The problem with licensure is that there are many markets where the number of licensed professionals is zero. My wife reminds me of this--she grew up in rural Idaho, where many things I take for granted (e.g. electricians, medical care) growing up in the Chicago suburbs, weren't available.
Let's consider the choices of a homeowner in this market, who needs electrical work done: (a) pay a licensed professional to travel into their market (expensive), (b) have someone unlicensed do the work (or DIY it), or (c) just don't do it.
This is why I think certification (not mandatory licensing) is the way to go. It's not clear that criminalizing economic activity when it's someone's only real choice--and there may be major health/safety hazards if the work isn't done--is beneficial.
- This is it. There's real skill in creating simple solutions to complex problems. Knowing the general landscape of what's out there, and easily available off the shelf, really does help.
Developers grow. It starts with simple code that doesn't work. The next step is, complicated code that solves the problem, in messy unmaintainable ways. The next step is writing super clean, almost boring code, that's highly readable and "dumb" and does exactly what it's supposed to do.
The other thing to realize, a lot of great code doesn't just spring from the developer's hands in its final form -- it's extensively edited and rewritten into its final, good, form.
- I think this is more in line with how hiring works for senior jobs. I've been part of 50+ hiring decisions and do it like this.
Social stuff is absolutely the first thing I'm checking. It's a quick test to spot total lies, e.g. you claim you worked somewhere but are connected with zero people on any social network, or claim attendance at a school with zero connection to anyone. If nothing else, it helps to build rapport for the interview.
I would 100%, absolutely trust a personal recommendation over a dumb whiteboard interview.
- I think we're going to see a lot of this. I worked in self-driving and the stuff always 95% worked. Never 100.
This is useful in some ways. Thinking about situations like pre-release software testing, there are exploratory test cases that are simply too numerous to ever have a human perform economically. A lot of AI is going to do this kind of very low-valued grunt work where it doesn't matter if it's 90% or 99% correct, it's the fact that it can get done at all. A lot of this work is "additive" in the sense that, it's just too expensive to do today (with a human).
The work product of these systems is best seen as a "rough draft" or "suggestion". It's a first cut, not the last word.
On the other hand, a lot (most?) of the meat-and-potatoes coding done today, is situations where things have to WORK. Stuff where correctness absolutely matters--billing/money/settlement (calculating tax, handling returns, moving money between accounts), a lot of OS code for things like memory management / locking / resource management, drug dosing, reservation management, etc.
Granted, this stuff is a lot more complex and nuanced than the code of an average CRUD app, but then, I also don't spend my days implementing bcrypt, quicksort, or self-synchronizing Unicode parsing. We have libraries for that. The question is whether we're better off relying on agents to write a bunch of grunty code, or come up with better top-level organization / code structures, that doing it "by hand" is the better approach.
I'm actually optimistic that we can do better code-wise. But I'd love to see how things develop. Maybe we wouldn't need AI if we just had better programming languages.
- Forget management. This post is about the proper exercise of power. Don't run from it. Parents, landlords, even political leaders deal with this stuff.
First, good leadership is something we should celebrate. It is very hard, requires making a lot of tough decisions, and comes with a real personal cost. Unless you're a total psychopath, which most managers aren't, firing people or delivering bad news isn't easy--for anyone. Doing this all day is difficult.
However, there's a reason most human undertakings of any consequence involve hierarchy (armies, organized religion, etc): it works. It's needed to get things done. Companies aren't charities and in order to ensure the paychecks keep cashing, things have to get done. Goals have to be hit, decisions have to be made, and occasionally, staff let go or fired. All of this falls to management.
You don't have to do it, but someone does.
My favorite essay on this topic: https://medium.com/8vc-news/a-deficit-of-leadership-38bb888a...
- I used to feel this way.
But then I realized this is the natural push/pull of conservatism against progress. Incumbents get old and conservative. New companies crop up and try to do things differently. Sometimes the new ideas are stupid, and incumbents are right--the tried and true really is best.
Other times, the incumbents have become so bureaucratic, slow, and out of touch with what matters (customers), that a new company can come in, do things dramatically better, and burn an old competitor down.
Macro-wise, I think we should be glad for this. It's very hard to tell what will and won't work at the beginning. VCs get paid a lot of money for trying to guess this and even they can't get it right. I myself have been surprised many, many times when an idea I thought was incredibly stupid, turned out to have legs.
One way to change your mind on this, write down your predictions about what companies/approaches/techniques will/won't work 10 years from now. If you do this say, once/year, and you're honest, you'll see how hilariously bad you, and more or less everyone else actually is, at predicting the future and knowing what will and won't work. Sometimes you really do just have to try it, and see what happens.
- Came here to say this.
I'm a small real estate owner (alongside running a tech co). I spent about 90 minutes yesterday reviewing an agency agreement giving a brokerage the exclusive right to market one of my units (commercial) for lease.
Even in such a "simple" contract, I demanded many changes. They wanted the exclusive right to represent us in case the building was sold - I took that out. They wanted a flat 5% commission for the length of the lease - I changed that to only the first seven years. They wanted indemnity, and to be reimbursed for expenses, and many other things I just wasn't going to do.
It was the same thing with a lease earlier this year. We argued for months over the terms - how much rent, who was responsible for HVAC (me as owner, for replacing one unit, the tenant for ordinary repair and maintenance), late charges, whether or not we were doing ACH (my preference) or paper checks, whether or not a renewal option was included (I took that out), and a bunch of junk their lawyer wanted that I mostly managed to keep out.
Now, maybe there are some very basic agreements that can be relatively boilerplate, low-value things, or maybe it's valuable to start from some kind of common standard.
But in my experience, the whole concept of this product is flawed. It actually reminds me of something I built 5-6 years ago, a simple chat bot (called Interval) to handle scheduling for low-value items like tennis courts, or conference rooms. My view of scheduling was a practical problem, almost algorithmic, of making it easy to reserve something. What I didn't understand, was the huge amount of social context inherent in scheduling--who's important, who's friends with who, who's doing somebody a favor, who gets special treatment, and why.
There's a lot of social context there that means scheduling is going to stay in the human realm, for a LONG time, long, long after 50 more people try replacing it with an LLM.
This product feels the same. Every business contract is a complicated dance involving negotiating power, who needs who more, who cares more, what language stays in vs comes out, settling some kind of score (I did this for you last time, you owe me this time), and 100 other things.
There might some kind of low-value contracts that aren't customarily negotiated, or maybe real estate is a particularly negotiation-intensive field (very possible), but I have a very, very hard time seeing all of this getting "standardized" in the name of efficiency improvement. It just isn't going to happen.
- I've been reading that this change (hub and spoke to decentralized) was already a big effect of COVID.
I like trains, but an alternative could be electric buses, with Uber Pool-like dynamic routing. I'm not sure how the speed would compare, but, it gets pretty wild when you consider what's possible with mobile phones, AVs, and electric vehicles.
I for one, am excited by this.
- I think this is right. I'm married with kids and have little desire to go to an office. It was indeed different when I Was younger -- I didn't cook at home as much so enjoyed office lunches with coworkers, I lived in a smaller apartment which was easier to find closer to work, etc.
It occurs that shows including "Call my agent", "The Newsroom", and to some extent "Madam Secretary" depict the better version of office life (particularly for young people) pretty well.
- Narrowing down a bit, I think it's the "downtown business district" that's on the way out -- and the accompanying high office rents, long commutes, etc.
My home is a place of rest and family, not work. Work happens at a dedicated location five minutes from home, on foot.
I think we'll see more of this in the future--mixed neighborhoods with a lot of residential space next to offices. What I think we won't see as much of, are huge office blocks very far from residential areas, where the majority of the workforce commutes an hour+ every day, each way, to work.
- I've noticed this with a lot of professions. Comp is back-loaded. They start much lower than software, but the good ones make it up in the end by being owners and having a crew of 10-20 work for them, and earning margin on their labor. Regulation and professionalization (e.g. you need to be licensed) seem to encourage this.
Software, assuming one doesn't go into management, seem to have a more linear comp structure. You start much higher, but it doesn't ramp as dramatically as most trade/licensed professions as you go up. Probably because there's less regulation/capture so even entry-level people keep more of what they produce, vs the almost-feudal nature of many trades/professions.
Just what I've seen, anyway.
- As someone who's taken a couple semesters of college-level accounting, and filed my corporate taxes correctly many years without audit (but am not an accountant/CPA), this seems broadly correct.
I'm actually surprised people are freaking out about this. Of course software is R&D. And of course you don't just get to expense it all at once. It's long-lived, like you said.
Maybe we could have some tax breaks like our friends over in real estate, but I very much think the base assumption should be that software dev is capitalized.
On recession: cost of living is becoming crisis-level. I read recently that 67% of Americans are paycheck-to-paycheck. 150k/yr is 12k/month. If groceries go from 500 to 1000/month, a 150k wage-earner save less for retirement. For someone making 30-40k (basically minimum wage), it's a huge hit. Then consider it's the same story for cars, housing, medical care...it goes on and on. It doesn't look "recessionary" because GDP keeps going up. But we're getting so much less for it with every passing year.
I also agree that we need to consider what brownfield dev looks like. It's where the vast majority of my time has gone over 15+ years in software and I'm not convinced all the coordination / sequencing / thinking will be assisted with LLMs. Particularly because they aren't trained on large proprietary codebases.
What we might both be missing, is that for most people, writing the actual code is hard. LLMs help with that a lot. That's what a lot of junior/entry-level work, actually is (not as much planning/thinking as seniors do).