Something that’s occurred to me recently is that there seems to be a pretty big disconnect between how hackers see the world, and how many others (reporters, market analysts, investors) see the world. When non-hackers look at a company, they might measure it by the quality of the furnishings, the prestige of the office building, the market cap or valuation, or the category of products that they sell. The first thing a non-hacker asks about a new company is–where is it? Physically. This is a proxy for prestige. A company that commands its own sprawling campus is the best. Next is a downtown skyscraper office. Next is a large rented office building in the affluent suburb. Finally, there’s the industrial blue-collar part of town.
Companies like Oracle and Microsoft are very impressive by this sort of yardstick. I was recently invited to hear a patent attorney give a lecture–with the expectation that, because this person had a high hourly rate, it was obviously an incredible opportunity. Leaving the question of the ethics of patent litigation for another blog post, I cannot imagine a more valueless way to spend an afternoon.
I do not mean to suggest that the non-hackers are stupid or non-rigorous. For example, when the iPad was introduced, it was met with some skepticism. Analysts knew the size of the tablet market at that time, and it was small. Their reasoning was based on facts: how large is the market, who uses tablets. They massively underestimated its success–on the basis of somewhat reasonable statistical models.
Now in 2012, analysts know the market for tablets is huge, so they predict success for arbitrary webOS tablet, again on the basis of real world facts. Wrong again!
It’s not that they’re stupid. It’s that innovation, by definition, does not follow the previous patterns. You can probably predict how well an A5 chip will sell based on the sales data for an A4, but you are crazy if you are trying to predict the sales of a brand new Apple product. There is no basis for comparison.
You see, hackers tend to measure a company by the level of innovation that happens there. Instead of where is it, they tend to ask what is it? What are you working on? High-performance hardware? Graph algorithms? Video encoding? CRUD? Consider the first sentence on a major ad agency’s website:
Interpublic Group is a global provider of advertising and marketing services. Through our 42,400 employees in all major world markets, …
This text is positioned dead center on a huge picture of the company’s spacious Tokyo office, with leather chairs and lots of glass. Contrast this with some job postings on Hacker News:
Help build a game influenced by Minecraft, SimCity, The Sims, and Lego (YC S11)
Come change the way data is used. Chartio (YC S10) is hiring.
GoCardless (YC S11) hiring Linux sysadmins in London
It’s all about the technology. These listings tend to have big gorgeous pictures of the product or the team. Many of the headlines are so product-focused that they don’t even mention the name of the company, let alone the size of their office. It’s a different way of thinking.
Prestige and market cap and office space are lagging indicators of innovation. They tell you that innovation has occurred awhile back. Like a dog following a scent, they don’t tell you where it is, they tell you where it’s been.
I am too young to remember the first rise. I will let someone who was there tell you how it happened, because it’s an incredible story.
But I clearly and distinctly remember the second rise. The “click” moment for me was the Intel transition. In 2003, Steve Jobs put up this slide:
Hail the mighty G5. Followed by about 20 minutes of diagrams, charts, demos, and benchmarks proving it was faster than the G4.
In 2005, just 24 months later, Jobs said this at that same conference:
I stood up two years ago in front of you, and I promised you [the best computers we can build]. And I haven’t been able to deliver that to you. … But these aren’t even the most important reasons. The most important reasons are that as we look ahead… we can envision some amazing products that we want to build for you. And we don’t know how to build them with the future PowerPC roadmap. When we look at Intel, they’ve got great performance, yes. [But] just as important as performance is power consumption… and that tells us what we have to do.
It occurred to me then that Jobs was willing to eat crow, to destroy the whole ecosystem, just to build a better product. In spite of the arrogance ascribed to him and/or the Reality Distortion Field. The Altivec demos were a good marketing gimmick, but innovation is the real story. They’ll devalue the entire product line, obsolete old units, and force everyone to port their software, just to get a slightly better CPU.
That’s a hard call to make even for products that are deadpooled. Contrast this announcement with the one retiring the HP Touchpad line: buried deep in a press release that actually included the words “Other Announcements” in its title, HP said:
In addition, HP reported that it plans to announce that it will discontinue operations for webOS devices, specifically the TouchPad and webOS phones. HP will continue to explore options to optimize the value of webOS software going forward.
See the difference? No apology, no explanation, no technical rationale, no transition, no product roadmap. And it actually uses the phrases “reported that it plans to announce” and “optimize the value … going forward” which I can only assume are the prose equivalent of including a big fat shot of the Aeron chair in the glass office. We paid big bucks for this press release.
Of course, Jobs’ comments have taken on recently an even bigger prescience than anyone imagined back then: back in 2005, they were worried about “amazing products” that they can only build on the basis of performance-per-watt. Intel didn’t ultimately give them all they needed, but it’s astonishing to me how forward-looking that strategy was. In 2012, there are big tech manufacturers out there who still haven’t figured that out.
As soon as I saw this announcement, I knew that something big was happening at Apple. So I bought one of the very first Intel Macbook models that shipped. One Macbook at a time, Apple started taking over the developer conferences, started to displace the Linux-equipped ThinkPad as the de facto weapon. By about 2007, it became clear to the hacker community that Apple was on a rocketship trajectory. Where that trajectory was going exactly, and what it would mean for the future were unknown. But the fact that it was a big deal was clear. Innovation was happening at a computer manufacturer in a way that it hadn’t happened since the IBM PC was first released.
Fast-forward to 2012. Apple is the world’s largest company. We use words like “non-Apple” to describe entire markets, because otherwise it just wouldn’t be fair. Time to party, right? For the first time in seven years, I am concerned. The concern started with a blog post by John Gruber describing Jobs’ succession:
A new Apple CEO would need credibility and the ability to instantly instill confidence with two highly disparate groups: Apple’s rank-and-file employees, and Wall Street. The only candidates who could satisfy both factions, to any degree, already work at Apple.
You read that right: John Gruber is evaluating CEOs by how well they can impress people. Gruber goes on to say:
The knock against [Tim Cook] is that he’s an operations and finance guy, not a product design guy. Ideally Apple would find someone just like Steve Jobs, but there exists no such person.
Yes, “ideally” Apple would continue to have a focus around building products, but that’s just not feasible. See why this is worrying? If you believe, as I do, that success is a lagging indicator for innovation, and Apple is not picking its next CEO based on product innovation, danger, Will Robinson. If continued innovation is the key component of success, then the fate of the company rests in the hands of Jonathan Ive and Scott Forrstall. Neither of which had enough backing from the board to be named CEO. Hmm, I wonder what happened the last time Apple had a non-product guy at the helm?
To be clear, $300b is a lot of runway. I’m sure that Jobs left many innovative products in the pipeline, and Ive and Forrstall will be able to keep innovating for a long time. Apple has no real competition. (The hacker consensus is that the innovation at Google stopped, shortly before or concurrent with, the death of Jobs.) I’m saying that the acceleration seems to have changed sign. The use of the word “runway” means that the clock is ticking. And I fear, now, that it is. Paul Graham, who is more or less the defacto spokesperson for hackers, wrote:
I was talking recently to someone who knew Apple well, and I asked him if the people now running the company would be able to keep creating new things the way Apple had under Steve Jobs. His answer was simply “no.” I already feared that would be the answer. I asked more to see how he’d qualify it. But he didn’t qualify it at all. No, there will be no more great new stuff beyond whatever’s currently in the pipeline. Apple’s revenues may continue to rise for a long time, but as Microsoft shows, revenue is a lagging indicator in the technology business.
Somewhere along the way, I think we have slipped into a delusion: that Apple is a big company. They aren’t. They’re a very small company which just happens to have a $300b valuation. If you study them up close, if you talk to employees, past and present, you get the distinct impression that they are quite tiny. My clearest memory of the Jobsian Apple is hanging out with the engineers–playing trivia games, singing songs, inside jokes, and challenging the graybeards who worked on Lisa. It’s a lot more like a family vacation or a pirate ship than a technology company. Tim Cook doesn’t make a very good pirate. In fact, the ship might just be drydocked.
We’ve started from the wrong premise: we’ve started with the size of Apple as the fixed constant. Apple is large, so therefore we must find a large-company CEO. If we can’t find one who is a world-class innovator, oh well.
No. The first criteria is to pick the best innovator, and then if that person can’t handle a company the size of Apple, we downsize Apple to keep it healthy. Apple has been large only recently, but it has been innovative forever. The priorities are completely backwards. Of course, they are forwards if you are optimizing for impressing the investors with your Aeron chair and the glass skyscraper. Gruber is completely right when he points out that doing the right thing here wasn’t feasible. That doesn’t change the fact that they are selling the golden goose–sacrificing the real innovation in exchange for that picture of the Aeron chair inside the glass skyscraper office to slap on the quarterly report.
We will not get another visionary like Jobs for a long time. But I do not think all hope is lost. Presumably, the new innovation is happening in some small garage, just as Apple once was.
But suppose we were to discover another Jobs. Where would this person be hiding? What would they be working on? As for products, this person would be working on everything. Jobs was probably only able to execute a tiny fraction of the ideas he had given the technology and resources available to him. We would expect such a person to have bold ideas, of toppling industries and changing the way people use technology, insanely arrogant strategies for changing the way new technology is created.
Where would we expect to find this person? In a garage somewhere. Perhaps, if far enough along, having converted a house into the empire headquarters. Living somewhere in SV. This person would have to have an indisputable commitment to making great products. And they would have to be a world-class communicator. Someone who can inspire others to greatness, to attract the best and brightest to the world’s most important problems.
I think we’ve got at least one person who performs quite well by this yardstick: Paul Graham. pg’s numerous Requests for Startups are unbelievably bold, and discuss everything from destroying entire industries to transforming the way people interact with technology. Y Combinator is headquartered in a tiny office and all the startups it funds are forced to find their own space. It’s like building office space one garage at a time.
And of course, pg’s communication skills are truly world-class. His essays have inspired the greatest minds in software development, and HN is the de facto home page for hackers. He’s had some growing pains, but there is literally no single person with more reach to inspire more software developers than pg does.
And he distilled the Jobsian mantra in a way that even Jobs failed to do: make something people want.
Of course, there are many ways in which pg is different. For a “Next Jobs”, people are expecting a well-spoken presenter in a turtleneck who ruthlessly negotiates and builds a closed egotistical empire with a secret list of future products. Perhaps that is what makes the plan so brilliant: pg is alike in precisely those ways that matter and unlike in precisely those ways which don’t. They’ll never expect it.
That’s why it took me so long to see it–Y Combinator is much more like a federation than an empire. The creator of the Maps app (if there was a single creator) may forever remain a mystery, but the people behind Dropbox are Arash and Drew. You don’t have the Y Combinator logo slapped over everything to clue you in, nor is pg some ruthless guy who gets involved in your day-to-day ops and yells at you for a mockup that’s two pixels to the left. It doesn’t fit the stereotype–and it doesn’t have to. It may have been what Jobs did, but not what made it work–the important part was creating a culture in which innovation is encouraged to grow. That’s what pg does. He does it differently, but he does it.
And I think in a few key ways perhaps he does it better. pg has realized what Jobs could not–that a “one size fits all” working environment will always exclude good people. There are bright hackers who will refuse to work for Apple for whatever reason–the DRM, the closed ecosystem, the secrecy, they don’t get along with Jobs, the chairs are too squeaky, they’re on a project they hate. You can have these problems with some particular YCombinator company, but you can’t hate all of them, because they’re autonomous and independent and each have a distinct character.
One way to look at it is to say that Jobs pushed hard on the “product” side of the equation, the demand side. Make something people want. But I think that pg, while being completely true to that demand-side mantra, has also done a lot of thinking about the supply side. About how the teams that build things are structured. About how to attract the right development talent, about how to get the working environment right, about how to extract more productivity. For example:
Another thing you notice when you see animals in the wild is that each species thrives in groups of a certain size. A herd of impalas might have 100 adults; baboons maybe 20; lions rarely 10. Humans also seem designed to work in groups, and what I’ve read about hunter-gatherers accords with research on organizations and my own experience to suggest roughly what the ideal size is: groups of 8 work well; by 20 they’re getting hard to manage; and a group of 50 is really unwieldy.
And what you see in Y Combinator is a solution to this problem; a way of scaling horizontally. No single company can span the incredible breadth of products that Y Combinator funds, nor could any hierarchically structured organization hope to tug in all of these directions at once. But a with a federated organization, it now becomes possible for a couple of people in an unimposing Mountain View office to steer far many more projects towards success than Jobs ever did. What they lack in unified vision, I think they make up in productivity.
I believe that Paul Graham may actually transform the process of writing software. The default way to get software written today is to hire, but I do not believe it is the right way. For starters, everyone acknowledges that hiring programmers is horribly broken. And you have a huge cacophony of voices that explain to you, you need to present a programming challenge, you need to give them test projects, write better job postings, hire based on interest, and look for smart go-getters. And yet, the hiring problem remains intractable. There’s a new article on how to hire every week on HN.
I am starting to think that hiring itself might be irreparably broken; that the way forward is not to ask different questions or present a different coding challenge or cast a different magical rite, but simply not to hire. You wait for a startup to come along, and you acquire them. Hiring is essentially a special case of acquiring: it’s acquiring a team of cardinality 1 and a product of cardinality 0. pg is proposing a generalized model: acquire a team of cardinality M and a product of cardinality N. It’s a safer bet, and to the extent that startups are more efficient than the companies that acquire them, the team and product you inherit will be better than trying to construct either one in-house.
It’s a revolutionary way of getting software written, and pg is at the forefront of that revolution. He is the world expert in small software teams; there is literally no person that exists with more data or time invested in thinking about small software teams than pg has. So saying that small team acquisition is the solution to hiring is equivalent to saying that pg himself is the solution to hiring.
I can defend this idea in greater detail with particular reference to the iOS market, which I have studied intimately. We’re reaching a tipping point: essentially the hiring pool has been exhausted. I do not mean this simply in the general “hard to find good people” hyperbole. I mean, actually, the hiring pool is now zero. You cannot hire an iOS developer anymore for any amount of money. The senior people who write iOS software seem to prefer to work on their own projects or do smallish freelancing gigs rather than get a day job, and the people who have day jobs seem to either be uninterested in learning iOS, or, once they learn iOS, quit.
The people who have felt the rumble first are the recruiters, whose livelihood depends entirely on the number of people they can hire. Back in 2009, I wouldn’t even get calls, because they wanted people with multiplatform experience, and I just do iOS. Then in 2010, they started calling with full-time offers, but I was happy as a contractor. Then in 2011, they started saying “I know you don’t want a full-time offer, but do you know anyone who does?” Now they’ve broken down and just ask me to contract on projects, because they’re out of options. Unfortunately now I’m booked solid.
Lest you think I’m some special case, let me give you some numbers. I’m fairly active in the local iOS scene (or at least I was before I got completely booked). Of the developers I know, less than a third of them are a bona fide employee, and less than 10% have an employee number higher than 5. So just by posting a job ad, you’ve excluded two-thirds of qualified applicants, and by being a company large enough to pay people to post job ads, you’ve excluded 90%. Hiring is broken.
Some people say that it is all a bubble and we will return to the old way of doing things soon, but I am not convinced. You have new graduates entering the workforce now as freelancers, who have never known the 9-to-5, never been to a job fair. You have these huge mobile development shops that are essentially, when you strip away the fluff, a sales department for independent contractors, for startups funding their projects. I think this might be the new way. I think we will start to see more of it, not less.
And pg is at the eye of the storm, leading the charge. And if it is a revolution, it is a revolution comparable with any product that Jobs ever introduced.