6 ms·

There is a lazy argument that ignores the difficulty and hard work of putting out not only a working but a good application out, but the other end of that is us

There is a lazy argument that ignores the difficulty and hard work of putting out not only a working but a good application out, but the other end of that is using Dan's argument (which is valid), to dismiss criticism. To wit, Google has 80k employees now, in 2010 it was at 10k and the number only drops as you go further back. I think it's pretty easy to say that Google was doing more impressive things at the beginning of the century, and certainly the climb was greater. Within the last five years, there was a moment when Google had 4 competing chat apps. This is pretty rank inefficiency. If you were to take 80k people (not all are engineers, obviously) and sit them down and ask them to use their talents to build separately, you'd end up with a lot more innovation and productivity, I'm pretty certain of that. The entire point of a minimum viable product is not that you have a complete, flawless app, but rather that you get something up you can iterate on quickly. That can be built in a weekend (I'm a slow worker, but I know people that can do crazy stuff in a night and interest). My personal criticism (and something I'm testing), is that our current processes are not harnessing all the gains in productivity we've seen over the decades, nor the ability to do that because it is very cheap to do so is not happening on a large scale. It's not totally apparent why it is happening, but it isn't. I realized late that people take this criticism personally and feel like you're saying they are stupid or don't work hard or it is all so easy, but I ask you to consider the nature of our industry and how easy things are to build and then ask you why we don't see real competition across so many products. The moats of many companies are perceived. The goal isn't to build Google in a weekend, it's to take the first step and give yourself an opportunity. edit: my numbers are off, it's 25k to 100k

Silhouette 6y ago
Small businesses innovate. Big businesses consolidate. Figuring out how to maintain a small business pace of innovation while achieving the consistency and economies of scale of a big business is basically the tech industry's quest for the holy grail.
- lowdose 6y ago
  Isn't that exactly what Amazon is doing?
  - Silhouette 6y ago
    Sorry, I don't really see why. Amazon's big success stories recently have been about providing the infrastructure for other businesses, both in its retail storefront and with AWS. This is one of the tried and tested strategies for big businesses to exploit their greater resources while not actually having to do the grunt work to create new products or services that end customers are willing to pay for.
    - lowdose 6y ago
      Amazon is not a moonshot factory inside trying hundreds of ideas small & scale the successful big? You think AWS was planned ahead for years and everything that runs on top of it? If AWS was such a logical consolidation why has Amazon such a big market share of the cloud and can even Google & Microsoft barely catchup with the innovation? Did you ever speak to an engineer working at Amazon or is this some feeling you have?? I talked to Werner a couple of times and everything he told me disputes your personal assumptions. The commoditization of logistic and compute infrastructure is real innovation although it looks like consolidation from the outside.
      - Silhouette 6y ago
        Amazon as a whole has nearly a million employees and generates 12-figure revenues. It would be remarkable if it had no R&D at all happening outside its main activities. But imagine if you could take the more capable people who work on Amazon's high-tech systems like AWS or logistics automation, bundle them into well-composed teams of a few people each, and give each team say $10M in funding. You've potentially just created more highly capable startups with enough runway to do something worthwhile than YC has funded in its entire history. Tell me with a straight face that Amazon as a whole is innovating as much as the combined portfolio of YC from day one.
quicklime 6y ago
Their 2010 revenue was $29.3 billion, and in the last four quarters (including the recent Q1 announcement) they raked in $166.71 billion. If your headcount numbers are correct, and they've gone from 25k to 100k employees, they've actually managed to increase their revenue faster than their headcount. Their constant failure to make a dent in the chat app space is surely embarrassing, but having a few parallel efforts is a drop in the bucket in terms of efficiency.
- BenderV 6y ago
  Facebook have 2 successful chat apps (Whatsapp, Messenger) and 2 successful social networks (Facebook.com, Instagram)
  - TulliusCicero 6y ago
    One of those chat apps was already enormously successful and then Facebook just bought it. The other was a natural outgrowth of Facebook.com being a successful social network, you may as well claim that Instagram is also a successful chat app.
crazygringo 6y ago
> I think it's pretty easy to say that Google was doing more impressive things at the beginning of the century... Within the last five years, there was a moment when Google had 4 competing chat apps. This is pretty rank inefficiency. If you were to take 80k people (not all are engineers, obviously) and sit them down and ask them to use their talents to build separately, you'd end up with a lot more innovation and productivity, I'm pretty certain of that. You're misunderstanding how large companies operate. Their job is not to produce massive innovation -- that's what startups are mostly for, because most innovation actually fails in the market. Once innovation is proven, a big company like Google or Facebook can buy it to scale it larger and with less risk than the startup could have on their own. Google's job has been to increase revenue, and they've been doing a great job at that. The productivity increase happens from building enterprise features, hiring gigantic sales teams, etc. Building Google Cloud. And separate chat apps is necessary when WhatsApp and Slack and iMessage are eating your lunch. You can't risk having a single chat app that fails -- so you build one (Duo) that competes with WhatsApp and iMessage, another (Hangouts Chat) that competes with Slack, without pulling the plug (yet) on the legacy Hangouts. So if you took all 80k people, as you suggest, you'd wind up with a mess of startups which will mostly fail. You're not going to wind up with Google's actual increases in revenues, that's for sure.
- gustavo-fring 6y ago
  I understand that's how lots of big companies operate, but I disagree that's how companies have to or should operate. The 80k thing isn't he notion that you'd get 80k success stories (you won't), but rather that if you had a .01% success rate, you'd still end up with 8 successful companies out of that, and the market is big enough for those 8 companies doing different things, making lots of money, and creating jobs/value. But more importantly, I feel like it would be better for us as societies.
  - jmchuster 6y ago
    Google Ventures already spends $4.5 billion on that very thing? And there are tons of other VCs and incubators out there?
yowlingcat 6y ago
" The entire point of a minimum viable product is not that you have a complete, flawless app, but rather that you get something up you can iterate on quickly." Given that I agree with this premise, it's interesting that I come the exact opposite conclusion. If 99.9% of the work is in the iteration after the MVP, then when someone says "I could build that in a weekend" it's totally fair to say "'That' is only 0.01% of the work". Hard work that it may be to put out an application, none of that matters when no one uses it. Conversely, easy work that it may be to come up with an idea that has already been done before, if it's successfully marketed and developed into a product with traction and then ends up being the first of its ilk to be the runaway success, then you end up with Google instead of Lycos or Yahoo. I think his point is that reducing successful companies and products to implementation is to miss the forest for the trees. Traction is hard. Execution is everything. If it was easy, everyone would be able to do it successfully.

Silhouette 6y ago

Small businesses innovate. Big businesses consolidate. Figuring out how to maintain a small business pace of innovation while achieving the consistency and economies of scale of a big business is basically the tech industry's quest for the holy grail.

lowdose 6y ago
Isn't that exactly what Amazon is doing?
- Silhouette 6y ago
  Sorry, I don't really see why. Amazon's big success stories recently have been about providing the infrastructure for other businesses, both in its retail storefront and with AWS. This is one of the tried and tested strategies for big businesses to exploit their greater resources while not actually having to do the grunt work to create new products or services that end customers are willing to pay for.
  - lowdose 6y ago
    Amazon is not a moonshot factory inside trying hundreds of ideas small & scale the successful big? You think AWS was planned ahead for years and everything that runs on top of it? If AWS was such a logical consolidation why has Amazon such a big market share of the cloud and can even Google & Microsoft barely catchup with the innovation? Did you ever speak to an engineer working at Amazon or is this some feeling you have?? I talked to Werner a couple of times and everything he told me disputes your personal assumptions. The commoditization of logistic and compute infrastructure is real innovation although it looks like consolidation from the outside.
    - Silhouette 6y ago
      Amazon as a whole has nearly a million employees and generates 12-figure revenues. It would be remarkable if it had no R&D at all happening outside its main activities. But imagine if you could take the more capable people who work on Amazon's high-tech systems like AWS or logistics automation, bundle them into well-composed teams of a few people each, and give each team say $10M in funding. You've potentially just created more highly capable startups with enough runway to do something worthwhile than YC has funded in its entire history. Tell me with a straight face that Amazon as a whole is innovating as much as the combined portfolio of YC from day one.

quicklime 6y ago

Their 2010 revenue was $29.3 billion, and in the last four quarters (including the recent Q1 announcement) they raked in $166.71 billion. If your headcount numbers are correct, and they've gone from 25k to 100k employees, they've actually managed to increase their revenue faster than their headcount. Their constant failure to make a dent in the chat app space is surely embarrassing, but having a few parallel efforts is a drop in the bucket in terms of efficiency.

BenderV 6y ago
Facebook have 2 successful chat apps (Whatsapp, Messenger) and 2 successful social networks (Facebook.com, Instagram)
- TulliusCicero 6y ago
  One of those chat apps was already enormously successful and then Facebook just bought it. The other was a natural outgrowth of Facebook.com being a successful social network, you may as well claim that Instagram is also a successful chat app.

crazygringo 6y ago

> I think it's pretty easy to say that Google was doing more impressive things at the beginning of the century... Within the last five years, there was a moment when Google had 4 competing chat apps. This is pretty rank inefficiency. If you were to take 80k people (not all are engineers, obviously) and sit them down and ask them to use their talents to build separately, you'd end up with a lot more innovation and productivity, I'm pretty certain of that. You're misunderstanding how large companies operate. Their job is not to produce massive innovation -- that's what startups are mostly for, because most innovation actually fails in the market. Once innovation is proven, a big company like Google or Facebook can buy it to scale it larger and with less risk than the startup could have on their own. Google's job has been to increase revenue, and they've been doing a great job at that. The productivity increase happens from building enterprise features, hiring gigantic sales teams, etc. Building Google Cloud. And separate chat apps is necessary when WhatsApp and Slack and iMessage are eating your lunch. You can't risk having a single chat app that fails -- so you build one (Duo) that competes with WhatsApp and iMessage, another (Hangouts Chat) that competes with Slack, without pulling the plug (yet) on the legacy Hangouts. So if you took all 80k people, as you suggest, you'd wind up with a mess of startups which will mostly fail. You're not going to wind up with Google's actual increases in revenues, that's for sure.

gustavo-fring 6y ago
I understand that's how lots of big companies operate, but I disagree that's how companies have to or should operate. The 80k thing isn't he notion that you'd get 80k success stories (you won't), but rather that if you had a .01% success rate, you'd still end up with 8 successful companies out of that, and the market is big enough for those 8 companies doing different things, making lots of money, and creating jobs/value. But more importantly, I feel like it would be better for us as societies.
- jmchuster 6y ago
  Google Ventures already spends $4.5 billion on that very thing? And there are tons of other VCs and incubators out there?

yowlingcat 6y ago

" The entire point of a minimum viable product is not that you have a complete, flawless app, but rather that you get something up you can iterate on quickly." Given that I agree with this premise, it's interesting that I come the exact opposite conclusion. If 99.9% of the work is in the iteration after the MVP, then when someone says "I could build that in a weekend" it's totally fair to say "'That' is only 0.01% of the work". Hard work that it may be to put out an application, none of that matters when no one uses it. Conversely, easy work that it may be to come up with an idea that has already been done before, if it's successfully marketed and developed into a product with traction and then ends up being the first of its ilk to be the runaway success, then you end up with Google instead of Lycos or Yahoo. I think his point is that reducing successful companies and products to implementation is to miss the forest for the trees. Traction is hard. Execution is everything. If it was easy, everyone would be able to do it successfully.