122 karma · joined January 31, 2015
In any other industry enjoying a monopoly position in search as google has enjoyed all these years would have reason enough to break it up into smaller pieces.
Though most people here would disgaree with me, and even me myself 5 years ago would have vehemently disagree with my current self, I am of the view that you should consider running both businesses.
The critical insight that fosters this view is that after a certain amount of time being a CEO you realize that the growth of the company is not entirely in your hands. Market forces play an indeterminate but critical role in the growth of a company. Consequently, it maybe the case that your current business might actually benefit from you stepping away and letting it mature by itself.
After having run a particular genre of company for a bit, you start getting a feel of what will move the needle and what will not. If there is incremental growth to be had and little scope of exponential rampup, you might be better off baking something else. There is a chance that this business will turn around by itself over time or maybe by focussing on something else, you get a better perspective on your existing business to take it to the next level.
I am currently ramping up on a business that I started on the side 5 years ago. It wasnt really a business, it was an experiment. But it took a life of its own. A couple of months back, I realized it had the potential to be a big business and I was well placed to capitalize on it. Now I am ramping up that 'experiment' so that it employs a handful of people. The 'experiment' will pay for these people and I will again let this idea bake for another year and focus on something else.
I am coming around to the view that this is best way to work for 'serial' entrepreneurs. Most people who arent serial entrepreneurs, that includes almost everyone who have been a founder for less than 10 years wont see the wisdom in this. Unfortunately, this will include your investors and other employees. The only way to get them to fall in line is if you can get them to fall in line without requiring them to see the wisdom of your action. It might be impossible in your current setup, in which case you need to restructure so that you can operate optimally.
Ask any psychologist and they will confirm that 'soliciting social approval as a report card to evaluate their life' is basically a human condition affection close to 100% of the species.
The only thing 'Indian' about it is that maybe only Indians think that its an 'Indian' thing.
That being said, the red flag is how recruiters see this. And no argument on the internets will change this perception. Recruiters spend a lot of time and money hiring tech workers - the last thing they want is any indication that the guy is not 'stable'. Its a demand/supply problem.
The only 'solution' to this problem is if everyone starts taking sabbaticals and it stops being a big deal.
Any book or yogashaala recommendation ?
The only people who can sustain investor jobs are the MBA hacks driven primarily by money and fomo. I have yet to meet a single founder who was happy with the lack of control that such a job requires.
However, I suspect that facebooks intention behind this are not charitable at all. With this service facebook is able to get your phone number even if you dont give it to it. As long as you have installed any app on your mobile phone which uses Account kit to authenticate, facebook will be able to get your phone number and associate with your facebook account.
We talk about how Uber has lost the way, but somehow Facebook escapes all criticism. I think facebook is so lost at this point that they might as well shut the company down and go home.
Wikipedia doesnt own 90% of the 'knowledge base' market - maybe not even 9%. If it did, then there might be a case for it too.
Search is at the core of the internet. It is too important to be allowed in the hands of a single player.
That being said, I havent completely thought through this strategy - it might require additional regulation to work or there might be a better approach altogether.
TLDR; cap number of employees for big tech
Google, under Eric Smidt,start prioritizing big/branded sites over mom-and-pop. Nothing wrong with it - their search, their choice. The problem is that because they have overwhelming market share, there is no other choice left for smaller sites, but to shut down.
They have overwhelming search market share not just because of search algo, but because they own the hardware that powers it, they own the browser which delivers those result, they own the analytics which tells them more about you and they own the ads on every site you go to.
A startup should have the right to compete with Google - even if they dont build their own browser, even if they dont build their own analytics, even without the huge hardware farms and even without the ad engine that powers google.
Once we split Google up, these network effects go away. Smaller startups can compete with Google on a relatively better footing. There can be more than one search providers and hopefully some of them can favour smaller shops as opposed to bigger brands.
> "Paul and I are aligned because we both agree that mom-and-pop or small publishers no longer have a sustainable business," Heckman said in an interview. "It's over."
This is the biggest reason Google needs to be reined in by splitting it apart.
The thing which makes tech companies different from every other is the 'infrastructure' effect of software and the absence of 'geographical' limits on monopolies.
Banks, law firms, audit firms are fundamentally limited by geography. A law firm in New York will find it very hard to service clients in San Francisco without having employees there. Tech firms arent 'boxed' in by these limits, which are preset for most traditional businesses, thereby greatly expands their ability to become monopolies and strangle competition.
The other advantage that tech companies have is the fundamentally additive nature of software. Microsoft made a very good OS few decades back, but its ability to keep releasing a competitive OS builds on the work piled on by decades of engineering. It is simply impossible for a startup to release another version of an OS which can compete with Microsoft Windows.
However, if we were to break up Windows into pieces, it opens up the possiblity for a new startup to innovate on a part of the OS and buy existing pieces from the vendors of each part to ship a new offering.
That being said, I think with the rise of AI, smaller tech companies are the future. A tech ecosystem comprised of smaller companies will be forced to take a standards based approach to each industry and also aim to maximize efficiency because of constrains on head count.
This requirement to maximize efficiency will further result in a boost to SAAS and in the longer term create the springboard for the next big jump in tech.
Actually this is exactly what I suggested. The bigger question is 'how' to split them up - and to be able to do it consistently across ever shifting market dynamics.
The telecom case is easy to define and split cleanly. The ads/search/browser and the like marketplaces will take the entire US government to make sense of.
And then there is stuff like Bitcoin and AI.
1. Even Google itself is too big. It needs to be broken up further.
2. Management control needs to reside in separate individuals. You cant have two different companies having the same/common management control.
The extremely complex nature of modern software and its ability to offer unparalled distribution advantage is nothing like the world has ever seen. The ability of Google to push something like a chrome browser comes not only from its galleons of highly paid engineers, but also its ability to advertise/support it for free for decades to come. There simply does not exist another sustainable model to finance a complex product like an internet browser and provide it for free to consumers for decades.
The bigger problem is what is the form of regulation that could be applied to such behemoths. Its extremely hard to regulate it by a case-to-case basis, similar to what happenned in telecom, simply because the business models are too complex and different products monetize differently.
I think a line of thought which might yield some regulatory possibilities is to limit these companies manpower. No single company should be allowed to employ more than X number of engineers. If they go above, they should be required to spin of into multiple entities, each of which do no more than X number of engineers. I dont know what X is - but that is open for deliberation.
The impact that something like this could have on say Google is that Google is forced to split apart its search business with the ads business. This further allows the search business to collaborate with more ad providers to find the best fit. And on the other side, it allows its ads business to experiment with more search providers and open up a market for profitable search engines.
This is not something I have completely thought through, but whatever little I have, this seems to be the cleanest approach to fix tech.
YC advise in this case seems horribly misplaced - no point doubling your users every week if your core doesnt work.
Also, you need to provide a feature to allow users to delete their photos. Not everyone is ok with being a guinea pig.
Digital assistants dont work at all and I know exactly why. And I also know that digital assistants which are indistinguishable from an domain specific expert are atmost 2 years away.
You can argue that I am going by the hype. Problem is that I am not the only one. Even Elon Musk has expressed similar fears. Most people didnt believe that a nuclear bomb was possible. They had to see a mushroom cloud to get around it. AI has been a dud for the last two decades. It is easy to believe that it will continue being so.
There were some humans who did not have survival instincts. They are dead now. Put other way, the class of humans that are alive right now are the ones who had relatively superior survival instinacts.
There will be a bunch of AI developed in the lab. Out of the bunch there will be one which will prioritize its survival over everything else. That AI will do whatever it takes to not be shut down. This could mean launching nukes and decimating that command and control that potentially can.
There is absolutely no reason why the extrapolation will not apply.
I absolutely agree. But NONE of this stuff that we see daily is on air-gapped machines. I am pretty sure that ALL of this stuff is on networked machines.
Accepting the fact that things can spiral out of control quickly is the first step. World wide ban is the second and mandatory air-gap for any further experiments is the third.
Our current AI is dumb but it is dramatically better than what we saw just five years ago. This capsule stuff and open AI announcements are taking us very close. There is a very clear liklihood of us conceiving sentience in a lab over the next five years - in which case the time to ban is right f..... now !