Hacker News Reader

seems really poorly defined and argued

Here are the author's examples:

>McDonalds and Burger King

McDonald's is far larger than BK. Also Wendy's is larger than BK according to this source: https://www.qsrmagazine.com/content/qsr50-2017-top-50-chart

>GM and Ford

Another non-example. We just gonna pretend toyota doesn't exist?

>Nike and Reebok

Again, wtf? Even when reebok stood alone, there was still Adidas in the picture. Not a duality

Or indeed VW Group - which, as least as of last year, was the biggest car company:

https://www.bbc.co.uk/news/business-38793253

You should try to understand the point and the context of these examples. I would suggest you read Al Ries books.

These examples were picked in the early 1990s, and are specific to the US market. Things change.

For example the US car market has long been largely dominated by GM and Ford with Toyota only recently breaking that (and VW is non-existent). GM and Ford are still the two largest manufacturers: https://knoema.com/infographics/floslle/top-vehicle-manufact...

The point is that there is usually very few dominating brands in a market, with the top 2 (approximately) yielding the power and the others following and filling the gaps.

> These examples were picked in the early 1990s, and are specific to the US market. Things change.

...like whether that claim is well founded or not.

In the tech sector this has often reduced to 1 single brand instead of 2.

The worst mistake we engineers tend to make is to dismiss the 'wisdom' of professional marketers and branding experts...

It’s incumbent on you to make your point with contemporary examples or heavily explain why old ones are needed (and even then it severely weakens the point). I too was extremely confused with your examples.

Sure there should be charity in reading but the default assumption is not that your examples are 26 years old. Many people here were either children or not even born yet.

It's odd to see the traditional industry of the "big three" automakers being used as evidence for the "law of duality"...

It's not odd when you look at the link I posted that shows that the 3rd of the "Big Three" was much smaller than the other 2 in terms of volume and brand.

In 1990, GM had 35%, Ford 23%, and Chrysler 12% of the US market. That doesn't seem like an overwhelming dominance by GM and Ford.

This "law of duality" is not very convincing. He does not offer criteria where it applies not does he argue why it would apply for the examples given. How does his adoption curve apply to coke?

Especially in the cloud market, I see rather more players in the future as the underlying technologies become commercially available or even commodity.

In the other side customers value support and services, which does not scale well, and smaller players have an advantage.

As others have noted, hype from random marketing book != actual laws of anything.

If I was starting a startup today, I’d choose Google Cloud. They’re a bit cheaper than AWS, they have all the big types of cloud services you need, and they tend to be more simple/plug-and-play than AWS, requiring less Ops expertise to manage. GKE is awesome, Big Query is easier to manage than Redshift, PubSub easier to both manage and use than Kinesis, if you ever want a truly global database Google has that while AWS doesn’t, etc. AWS has a larger product catalogue, and is more configurable for big enterprises with huge Ops/security teams, but for a startup that has mostly dev and little Ops expertise, I’d choose Google Cloud.

" as far as I always understood, there is a rule [1] that says there can always be only two leaders in a single space"

The '22-laws-of-marketing' are not 'laws' - they're just called that by profs trying to hustle a book.

It's a great book that everyone, especially Engineers should read, but it's not necessarily factual.

There can be many 'leaders' it depends on the industry.

Google is no slouch, they have a differentiated offer that may appeal to different kinds of orgs, particularly startups.

To your own point: 'developer friendly' ... Google may be a choice for devs building 'cloud first' solutions, and the primary basis for could based companies. Remember that AWS and Azure have a lot of customers just porting over old-school software onto cloud instances. They need virtual instances in quantity and classic, straight forward easy ways to manage them. Not fancy new cloud gear.

Cloud is big, and there's a lot of growth, moreover, as more services go cloud and we move away from legacy installations ... Google might even possibly have an advantage.

1. Google will do this anyway, because it's not going to run on AWS or azure, so why not keep some skin in the game?

2. Isn't k8s a play to reinvent this space?

Kubernetes also run on the competing clouds, so it isn't that much of an advantage, unless they add GCE specific features.

Well, they do. For example you can configure your node pool to automatically upgrade its members so you don't have to worry about kernel patches. They have a proxy service for their Cloud SQL product so you don't have to fiddle with configuration, its just another k8s service name. Of course Amazon could do these things but they simply haven't. I think their biggest advantage though is they abstract away the master - you just have a node pool and don't have to worry about the masters. With Amazon you will pay $300 a month just to have a HA master, before you launch even one node in your pool.

Also once you start getting into k8s you will find so much more information in the community with specific instructions for GKE; its by far the best environment to learn in for that reason.

Google falls down in so many other areas though that it probably doesn't matter how good GKE is. The biggest annoyance to me is their Postgres offering still has no point in time recovery.

edit to add: One other big advantage I forgot to mention is that launching a GKE cluster takes only a couple of minutes. I have a Terraform configuration that will launch a GKE cluster and run my deployments and my app is up and running after less than 5 minutes. This makes it much more attractive to do things like spin up a new cluster for each pull request.

EKS pricing is $0.20/hour, or ~ $144/month. Not free, but also not $300.

I will agree that EKS startup is slow, but given the way it needs to run, I don’t think many customers will want to run more than a few clusters per VPC.

$144 a month gets you one master. As I said, an HA master is twice that. If you want to compare to GKE, you are starting at $288 a month.

The only master for EKS is an HA master (to be precise a 3 AZ master).

Not sure where these numbers are coming from but EKS has 3 masters by default. You don't get HA by running 2 clusters (I assume that's where you got $288). $144 gets you 3 masters in 3 AZs.

1. Because it is a very different thing to have an internally managed cloud than it is to offer this as a product with the massive support required for enterprise, something that Google has never put much resource into.

Amazon and MS have vast support networks because enterprise wants to feel like they have some control and feedback on their systems.

Also, unless you can innovate in the space (or go really cheap), why would I ever choose a smaller player over a larger one?

This is another example of the problem these rich companies have where cash is not a problem but identity and direction is.

I completely agree with this.

But, also in think Google does have some advantages. Out of the box, their NLP algorithms seem much better than AWS’s offering. They also offer custom search as an api. I haven’t used it, but bigquery seems very good. I do agree these are probably not enough to have an enterprise shift. Maybe if they specialize on a couple industries that won’t choose amazon (fedex, ups, target, Walmart, streaming/media). It does seem like google has made strong inroads in finance as well.

There are technical reasons.

Azure is doing quite well leveraging Microsoft's existing enterprise relationships. I'm not sure how much success folks are having "lifting and shifting" their Microsoft infrastructure into the cloud, some, maybe?

However, everyone that I've heard of that understands cloud native type architectures that has attempted to deploy into Azure is looking to leave.

>However, everyone that I've heard of that understands cloud native type architectures that has attempted to deploy into Azure is looking to leave.

Extremely unsurprising, but I'd be interested to read more.

Kind of ignored this because it's a long list of stuff. I don't really want to repeat stuff that I've "heard". Friend of mine does consulting for a few big orgs, and he tells me that things are bad enough that many of his customers are looking to move in spite of being heavily subsidized by Microsoft. He mentioned numbers like >$1m in credits. Unfortunately, my friend isn't very technical so when I press him on details, I get very little.

However, that correlates pretty closely with my own personal experience. I've worked with a fair bit of the Azure products and it feels to me very much like a work in progress.

The identity management stuff has been a moving target since day one. That is, how to authorize against the service itself. This is apparent in many of the Python libraries where this is documented.

The Python libraries that I've seen look like they were produced by interns. I'm performance with ADL is pretty rough.

My main goal with Azure was to deploy Kubernetes clusters there. The list of problems is very long and has been documented ad nauseum in many Medium articles. In the 14 or so months I've been working with Azure, it hasn't improved as quickly as I'm accustomed to with AWS and GCP.

A year ago, I shifted the majority of my production infrastructure from AWS to GCP. By comparison to Azure, I was pleasantly surprised with the quality of the products from GCP. My company uses GSuite which made identity management and RBAC a breeze. The GKE product is way, way better than anything else, and there are a handful of things that I like better than AWS.

I wouldn't encourage anyone to shift from AWS unless they were planning a large Kubernetes implementation. I would encourage folks to try GCP and especially GKE to get a sense of the differences.

There's nothing about Azure that I've discovered yet that would cause me to recommend it to anyone. It's possible that if a company had a heavier investment in MS products that it might be worthwhile, but that's not the case for mine. We only ever use MSSQL occasionally for development purposes and I've found that I can spin up an instance of that quite nicely in k8s with the Microsoft authored helm chart.

k8s feels much more like an effort to commoditize this space through standardization. It's a pretty common business strategy [1].

[1] https://www.gwern.net/Complement

I’m not sure I agree with the rule of two. Industries typically consolidate over time. Also, anti-trust rules typically prevent a rule of one. Several industries have had multiple competitors thrive though. Banking would be a good example of this where Citi, JPM, Wells Fargo, and Bank of America are competing in a landscape where many more competed 15 years ago. Automotives have consolidated, but there’s still more than two. I would also throw IBM into the MSFT and Oracle 90’s enterprise comparison. I think one of the biggest advantages of Google cloud as a business is the fact that most retailers will refuse to use AWS and more retailers are building out their technical abilities.

"Where would this possibly leave Google? All I can see is that, since it would be impossible for them to get a decent position in the cloud market, they will have to completely reinvent this space and create their own, new market which makes the cloud as we know it obsolete."

How naive would it be to say there's such potential in Firebase? It seems so different from what's out there now, and Google's putting a lot of effort and resources into it now too (including sunsetting existing products like Google Analytics Mobile in favor of the Firebase versions.)

and when is firebase going to be sunset for the newer shinier thing? google has a huge problem with sunsetting services that work for something marginally better

I've also mentioned in another comment but in the market where I'm located (Eastern Europe) the Google Cloud sales-people are totally absent while MS has been very, very insistent in pushing Azure down people's throats. I think Google was expecting their cloud solution to be similar to how AdWords + Search worked for them, i.e. that they would need close to no sales-people down in the trenches because the product would almost sell by itself based on its technical merits alone, but apparently (as you well mention) the different cloud implementations are not that, well, different between them so it all comes down to which company was first in the market (AWS) and which company has got the best and most aggressive sales-people (MS).

so it all comes down to which company was first in the market (AWS) and which company has got the best and most aggressive sales-people (MS).

That is not it at all. What typically happens is an organisation will run a pilot project and see how it goes. On Azure it’s easy to speak to a real person, and they will be knowledgeable and take you seriously, and be empowered and willing to help you. On Google even if you manage to speak to anyone, they’ll treat you like an idiot. Guess who that organisation is going to go to for bigger and bigger projects?

My thoughts are that I don't know anyone that's adopted Azure at scale that are happy with the service.

N=1, YMMV, etc.

The rule might feel true as a consequence of Zipf's law that says the second biggest is roughly half as big, and the third biggest is 1/3 as big as #1, etc. So because of a power law, the #1 and #2 are big enough that many people will just dismiss the rest.

> there is a rule [1] that says there can always be only two leaders in a single space

The problem with this “rule” is it’s a tautology. Any ordered list of N > 1 will have two leaders. (The better question, how are market shares distributed, is difficult to precisely answer given the inherent subjectivity of delineating markets.)

Google cloud seems to be a better fit for startups while azure is for larger non-tech firms. I can’t imagine a new startup ever choosing azure over gcloud or aws and i can’t imagine some random all-microsoft insurance company choosing gcloud over azure.

Eventually All internet company will love and move to gcp. i.e. snap or Netflix or Salesforce or slack etc. All enterprise company will move to internet company. Google wins.

The law of duality just seems to be Zipf's law with an arbitrary cutoff at #2...

There's several aspects here.

I quite agree that if Google does not get into the top 2 it will likely be irrelevant in that market.

The other aspect, as discussed in the article, is that this is a completely new market and way of doing business for Google, which seems to create a clash of cultures. If that is indeed the case it may be extremely difficult for Google to overcome.

"The 22 Immutable Laws of Marketing" is a must-read, by the way, as is the one on branding.