3 ms·

Two stories spring to mind. The first is I worked at Google when the whole "real names" saga of Google Plus was going on. The worst part about that was if some

by cletus 7y ago

Two stories spring to mind. The first is I worked at Google when the whole "real names" saga of Google Plus was going on. The worst part about that was if some algorithm decided your name wasn't real, it could block your entire Google account. You'd lose access to GMail, Drive, everything. I don't know what bright spark PMs, lawyers and executives signed off on that one but it was insanity. As a result I would strongly recommend against any family or friends use any G+ features in case the Real Names bot zapped their accounts. This was ultimately fixed but it should never have shipped that way. Ever. The fact that disputing a charge with you credit card provider like this could result in you losing your GMail, Drive, Android developer account, etc means you should absolutely under no circumstances use that service, ever, period, no exceptions. Don't get me wrong: I like GMail and Drive. I use them all the time. Thing is, I want to keep using them so they get firewalled from every other service that I use. The second is something Eric Schmidt said a lot. He didn't say it first but he's well known for saying it, and that is that more revenue is the solution to every problem. That's both true and not true. It's true in a high margin business (like search). If you look at Google financials 10+ years ago there was a point where the profit per employee was nearly $1m. It's less now but it's still pretty significant. Low margin businesses are not in its DNA. Let me repeat that and make it clear: Google simply cannot compete in a low margin business. Personally I learned this the hard way as I worked on Google Fiber until that was quietly shelved (by "advancing our amazing bet" no less). Fiber services are capital intensive. They require strict financial discipline and planning. There are far more moving parts to this kind of business than many people realize. Comcast, Verizon and AT&T have gotten very good at this sort of business. Google didn't (and IMHO never will). I put Google Fi firmly in this camp. I also put making hardware at scale in this camp (compared to, say, Apple who is second to none when it comes to logistics, procurement, financing and supply chain management). Google's DNA is hard technical problems at scale with no UI like an almost blank page with a text box and a search button. The infrastructure behind that is unbelievably good. But telecommunications? Tied to my Google account? Sorry but that's a hard pass for me.

Gibbon1 7y ago
> Let me repeat that and make it clear: Google simply cannot compete in a low margin business. That reminds me of Intel. The addiction to high margin microprocessors and a captive customer base had two effects. a) Intel doesn't know how to play with customers that have a choice. b) Customers with a choice avoid designing in Intel solutions like the plague. Over the long haul Intel is a much smaller company than they would be otherwise. Here is a good one. In the late 80's Intel fucked over Panasonic badly enough that Panasonic put Intel in their prohibited vendors list. I worked for a group that managed to get Panasonic in as a customer. And then Intel fucked them again. And back on the list they went. Google has the same mindset that Intel does. They can't process the ramifications of their business practices on future business opportunities.
- cletus 7y ago
  > They can't process the ramifications of their business practices on future business opportunities. Here's where that really hurts them (IMHO). Google, as a whole, has a very short attention span. A lot of things at Google began as bottom-up engineering-led projects, the poster child for which is probably GMail. This seems to work really well for things that are engineering problems. I think it works incredibly badly for a bunch of other things though. We, as engineers, tend to get bored. There are all sorts of quotes for this in software development like after you finish the first 90% you need to finish the second 90% (I forget who said this). I see project after project come out of Google that might be promising but it gets abandoned as the talent moves on to the next hot thing until finally it's killed. The cost of this, as you say, is paid by future projects. The classic counterexample to me is Apple Pay (vs Google Checkout/Google Wallet/Google Pay/Android Pay/Whatever it's called now). Every few weeks Apple announces some new bank or merchant or country is launched. Over years now they've slowly been building out this ecosystem. Google just doesn't have the commitment, focus and attention for this schlep [1], which is why they'll lose. Apple does. Apple has their failures too (Ping anyone?) but it's fair to say Apple really does put more wood behind fewer arrows [2]. [1] http://www.paulgraham.com/startupideas.html http://www.paulgraham.com/startupideas.html [2] https://www.techradar.com/news/internet/page-more-wood-behind-fewer-arrows-driving-google-success-979307 https://www.techradar.com/news/internet/page-more-wood-behin...
  - HillaryBriss 7y ago
    > after you finish the first 90% you need to finish the second 90% (I forget who said this). I don't know who said it either, but I heard an alternate phrasing: "The first 50% of the project takes 90% of the time. But the second 50% takes the other 90% of the time."
    - cletus 7y ago
      I probably should've just Googled it as it was easy enough to find. According to Wikipedia [1], it was Tom Cargill of Bell Labs and it's called the 90-90 rule: > The first 90 percent of the code accounts for the first 90 percent of the development time. The remaining 10 percent of the code accounts for the other 90 percent of the development time [1] https://en.wikipedia.org/wiki/Ninety-ninety_rule https://en.wikipedia.org/wiki/Ninety-ninety_rule

by cletus 7y ago

Gibbon1 7y ago

> Let me repeat that and make it clear: Google simply cannot compete in a low margin business. That reminds me of Intel. The addiction to high margin microprocessors and a captive customer base had two effects. a) Intel doesn't know how to play with customers that have a choice. b) Customers with a choice avoid designing in Intel solutions like the plague. Over the long haul Intel is a much smaller company than they would be otherwise. Here is a good one. In the late 80's Intel fucked over Panasonic badly enough that Panasonic put Intel in their prohibited vendors list. I worked for a group that managed to get Panasonic in as a customer. And then Intel fucked them again. And back on the list they went. Google has the same mindset that Intel does. They can't process the ramifications of their business practices on future business opportunities.

cletus 7y ago
> They can't process the ramifications of their business practices on future business opportunities. Here's where that really hurts them (IMHO). Google, as a whole, has a very short attention span. A lot of things at Google began as bottom-up engineering-led projects, the poster child for which is probably GMail. This seems to work really well for things that are engineering problems. I think it works incredibly badly for a bunch of other things though. We, as engineers, tend to get bored. There are all sorts of quotes for this in software development like after you finish the first 90% you need to finish the second 90% (I forget who said this). I see project after project come out of Google that might be promising but it gets abandoned as the talent moves on to the next hot thing until finally it's killed. The cost of this, as you say, is paid by future projects. The classic counterexample to me is Apple Pay (vs Google Checkout/Google Wallet/Google Pay/Android Pay/Whatever it's called now). Every few weeks Apple announces some new bank or merchant or country is launched. Over years now they've slowly been building out this ecosystem. Google just doesn't have the commitment, focus and attention for this schlep [1], which is why they'll lose. Apple does. Apple has their failures too (Ping anyone?) but it's fair to say Apple really does put more wood behind fewer arrows [2]. [1] http://www.paulgraham.com/startupideas.html http://www.paulgraham.com/startupideas.html [2] https://www.techradar.com/news/internet/page-more-wood-behind-fewer-arrows-driving-google-success-979307 https://www.techradar.com/news/internet/page-more-wood-behin...
- HillaryBriss 7y ago
  > after you finish the first 90% you need to finish the second 90% (I forget who said this). I don't know who said it either, but I heard an alternate phrasing: "The first 50% of the project takes 90% of the time. But the second 50% takes the other 90% of the time."
  - cletus 7y ago
    I probably should've just Googled it as it was easy enough to find. According to Wikipedia [1], it was Tom Cargill of Bell Labs and it's called the 90-90 rule: > The first 90 percent of the code accounts for the first 90 percent of the development time. The remaining 10 percent of the code accounts for the other 90 percent of the development time [1] https://en.wikipedia.org/wiki/Ninety-ninety_rule https://en.wikipedia.org/wiki/Ninety-ninety_rule