Live data from Hacker News

Steve's Google Platform rant

plus.google.com

231–240 of 418 posts

Re: Steve's Google Platform rant

#231
post #56

Earlier quoted context omitted.

I've thought a lot about this. In part because, you can't generally say anything bad about Amazon without being attacked. "You must be disgruntled" etc. I only made my post because I had the cover of Steve Yegge saying very similar things (Though I went into more detail.) Most people believe Amazon's press releases. In 2006, they said that AWS powered Amazon.com. It was a flat out lie. But how could I prove it? Fortu…

AWS was pirate operation That's amazing to me. I thought such things were never done in software anymore because: why not leave the company and do the same thing in your own startup? Same hard work, high risk, etc, but with giant upside. Was it because of something wrong with Seattle's startup culture?

There is no effective Seattle startup culture. I worked in the Seattle area for 7 years, and startups are not really a big thing. Going to work at microsoft is a big thing. You make more money there anyways.

A lot of this is due to the lack of a VC infrastructure I think. no sand hill road there.

Re: Steve's Google Platform rant

#232

Earlier quoted context omitted.

I interviewed at Amazon and got an offer recently, but I have to admit, it did kind of sound like this kind of job. I really liked the offer stage, where I asked for 4 weeks vacation (what I have now) instead of their standard 2 weeks. "That's not negotiable! It wouldn't be fair if you got better benefits just because you're better at negotiating than others on your team!" True. Offer rejected.

Europe and Australia have forever ruined me with their standard 4 weeks of vacation time. I can't fathom ever going back to anything less.

The legal minimum in the UK is now 5.6 weeks; That's still amongst the lowest in Europe. I simply cannot comprehend how even the best paid workers in America expect less holiday than the lowest-paid European labourers.

Re: Steve's Google Platform rant

#233
post #170

It now 404's so I've posted it here: Stevey's Google Platforms Rant I was at Amazon for about six and a half years, and now I've been at Google for that long. One thing that struck me immediately about the two companies -- an impression that has been reinforced almost daily -- is that Amazon does everything wrong, and Google does everything right. Sure, it's a sweeping generalization, but a surprisingly accurate one.…

Although the etiquette of posting it here is debatable this post is gold, it is not (totally) an ordinary disgruntled coder rant. It's worth reading even for the small list of problems that he gives that Amazon faced when trying to go to an all-services architecture. But if I had to summarize the main point, I would choose: "The Golden Rule of Platforms, "Eat Your Own Dogfood", can be rephrased as "Start with a Platf…

I agree with you in principal but I think practically this isn't always possible.

Sometimes you don't know if a product is worth being an entire platform until it's been thoroughly field tested. It's too expensive to assume that everything is.

Perhaps in Google's case there could have been more foresight, though.

Re: Steve's Google Platform rant

#234
post #56

Earlier quoted context omitted.

I've thought a lot about this. In part because, you can't generally say anything bad about Amazon without being attacked. "You must be disgruntled" etc. I only made my post because I had the cover of Steve Yegge saying very similar things (Though I went into more detail.) Most people believe Amazon's press releases. In 2006, they said that AWS powered Amazon.com. It was a flat out lie. But how could I prove it? Fortu…

AWS was pirate operation That's amazing to me. I thought such things were never done in software anymore because: why not leave the company and do the same thing in your own startup? Same hard work, high risk, etc, but with giant upside. Was it because of something wrong with Seattle's startup culture?

A startup doesn't have Amazon's network of datacenters or hardware resources.

(I've never worked at Amazon, but I'd also heard that AWS was a small skunkworks project that basically got cover from Amazon's CTO, Werner Vogels, who protected and nurtured it until it was too big to kill.)

Re: Steve's Google Platform rant

#235

Earlier quoted context omitted.

On the contrary, the way Amazon treats their employees does harm them substantially, they just haven't paid the price visibly to the public (or quite as hard as they will, eventually). Attrition at Amazon is at horrific rates. I know the actual number, though I'm pretty sure that'd violate my NDA to reveal. It's high. It's really high. Guess a really truly terrible number. It's probably higher than that. The stream o…

"On the contrary, the way Amazon treats their employees does harm them substantially, they just haven't paid the price visibly to the public (or quite as hard as they will, eventually)." If it hasn't hit them yet, it might after the rant. Given the Dilbert Employer From Hell publicity generated by the rant and the subsequent discussion, I'd say that they might have trouble filling in for the people that leave. Heck,…

Locally in Seattle, Amazon has a really bad rep. Only on places like hackernews where everyone is excited about AWS (despite its many, huge, annoying flaws, remember when ebs when down - AGAIN?) do people not realize this.

Amazon has lost a lot of key tech talent, it's been happening since about 2006 when the economy picked up a lot more.

Re: Steve's Google Platform rant

#236

Hello All, I was an intern at Amazon this summer and they extended a full time offer. I read Steve's rant with great interest. Since many people in comments have confirmed the points raised by him about Amazon, I am not feeling good right now :( I still have 18 days to accept the offer. I am currently interviewing with Microsoft. I have applied to Facebook just now. Sadly, I screwed up my Google phone screening last…

I was kind of in a similar position a couple years ago. I didn't do an internship at Amazon, but I was considering joining full-time, and was really unsure of whether to accept (I had great offers from a couple startups, which I was mostly focused on at the time, but a friend at Amazon was really pushing for me to join).

In the end, I decided to go with Amazon, thinking like you are that I could always change jobs if I didn't like it.

And yeah, I hated it. Absolutely hated it so much that I quit two months after joining. The problem was that the other startups I was interested had filled the position I'd been looking at (I'm more of a data analyst than a software engineer, and data analyst positions are a little more rare), and now every other place I'd interview at would think I was a total flake. I'd interview somewhere, and I'd have to spend 30 minutes explaining to multiple people why I left Amazon so soon and how they'd know I wouldn't do that again at their place.

I'd have great technical interviews, and I would get told this, but people would be worried about what kind of employee I was simply because of this one mistake I made joining Amazon.

So yeah. Hopefully you'll have a much better experience than me, so that even if you don't like Amazon you can always change jobs, but just throwing my own experience out there as another data point.

Re: Steve's Google Platform rant

#237
post #59

Earlier quoted context omitted.

Amazon internal propagand emphasizes that Amazon is a startup. "It's day one!" is a phrase I heard way too many times. They have "door desks" -- which actually cost them more than regular off the shelf furniture would-- but they pretend like they're being frugal to perpetuate the perception that they're a startup. (When I found out about the door desks costs, I was sworn to secrecy, it is very important for morale.)…

I have an acid test for any place which claims to be a startup. When someone proclaims this, ask them to answer any two of the following questions: - What's my name? (Covering your badge, naturally) - What am I working on? - Where am I from? If the bigwigs can't answer that, I'm sorry, you've moved past actual startup-dom and you're just in la-la land where you just think you are one. Unfortunately, I only thought of…

Google never actually claims they're a startup, they just say they want to be like a startup.

There're varying degrees of success at this, ranging from "if I squint really hard, I could almost see it" to "Yeah...I don't think so." But I give 'em props for trying, as most companies with 30k employees don't even make the effort.

BTW, I could go all the way up to the VP level within the Search bigwigs and they'd be able to answer this.

Re: Steve's Google Platform rant

#238
The fact that he didn't mean to post this is proof enough that G+ sucks. And he's spot on about everything he said. Google doesn't get platforms at all. Look what happened to Blogger, arguably their most successful platform (which they didn't even build).

Re: Steve's Google Platform rant

#239

It now 404's so I've posted it here: Stevey's Google Platforms Rant I was at Amazon for about six and a half years, and now I've been at Google for that long. One thing that struck me immediately about the two companies -- an impression that has been reinforced almost daily -- is that Amazon does everything wrong, and Google does everything right. Sure, it's a sweeping generalization, but a surprisingly accurate one.…

Its out of context, but please bear with me. Is it ok if amazon has an internal library(hypothetical) that is a collection of all its online books that they sell, can they have access to it , i mean normal guys have to pay for just getting a chapter out of a big book. For Amazon guys is there any concession , can they check on some cool book just like that?

That is not the sort of library he is talking about, I, unfortunately, may not give details, but rest assured that is not what is being referred to, at all.

Re: Steve's Google Platform rant

#240
There is one thing that Google has done well that no other company (Amazon, Microsoft, Facebook) seems to understand. Hire experts. They call this hire Ph.D.'s, but that's slightly inaccurate because having a Ph.D. does not make you an expert. They understand that building systems at large scale requires people who have a deep understanding of distributed systems that goes much beyond "My code is on SourceForge" mentality, or "Git is better than SVN because it is a distributed repository".

To this day, I am still shocked as to how many devs have no clue on what I'm talking about, yet they are in charge of Internet-scale systems. Here's a list of symptoms, I've heard over the years:

- I'll put something quick together.

- I implemented Paxos last night.

- I found an optimization in the two-phase commit protocol.

In my opinion, being expert means becoming humble and doubtful about your code when implementing large-scale systems. If your code runs on thousands of machines and serves 100K+ people and you think you rock as a developer/architect then you're doing something wrong.

Facebook doesn't get this. Look at their systems. They barely work. Good thing it doesn't matter. Yet. But it will eventually.

Amazon gets a little of this. Bringing Werner showed signs that they started to get it. They are still in this mix where a small group of people gets it and continues to bring in experts and push amazing things out. We'll see how long this will last.

Microsoft clearly doesn't get it. But that's ok, because they have no Internet-scale systems anyway. They built MSR which is capable of building such systems, but they make sure MSR remains isolated from their code. MSR seems happy to have no role in the company and to continue to publish amazing research.

Post reply on HN