Who amongst us hasn't used "Oops I hit reply-all by mistake" as a cover story for a good broadcast rant?
The dangers of social platforms have been illustrated yet again, when a Google staffer made public a long and detailed rant against his employer on Google+. The thesis of the post, attributed to Steve Yegge, is that Amazon can build platforms and expose services far better than Google: “We don’t understand platforms. We don’t …
Is the write of the article really not familiar with Steve's history of previous rants? I used to read them religiously, back in his Amazon days.
Stevey's Drunken Blog Rants™
I read a couple and they are too coherent to read like drunken.
They are also funny only for a given set of values of funny. Ditto for insightful.
For example in his Weak/Strong type rant he misses the obvious fact that nearly all big "Weak" typed systems written in Ruby, PHP and especially Perl operate off database backends (usually SQL). This means that they are actually "Hard" typed. In fact they are "harder" than a lot of stuff written in C++ or Java and + persistence.
So much for insight.
Bless you and your childlike hope and naivety. The strong typing that SQL offers is an impediment to swift development, in the same way that in Yeggeland security is an impediment to releasing products. Happily, you can just define every database with 50 columns of varchar(8000) and let all the clever stuff be done in the application! Better yet, completely redefining the schema can be done without waking up the database monkeys. Win-win.
This was posted to RISKS Digest back in 1994, a 1993 memo attributed to Tom Davis.
It's basically an overview of what's wrong with IRIX 5.1 performance, and what needs to be done to fix it. It includes gems such as "The complexity of our system software has surpassed the ability of average SGI programmers to understand it. And perhaps not just average programmers."
Accidental disclosure? I don't think it was an accident, and I think the most likely outcome will be the rolling of his head, though not right away. It is possible the Google will decide to fight the growing EVIL within and make some constructive changes, but I think not, for two reasons:
(1) My own experience with the Google says that the Google is increasingly EVIL and the Googlers don't care.
(2) He is inside, has a much better view, and apparently decided this was his best tactic, which is evidence that any normal internal corrective mechanisms have already been destroyed.
Disclaimers: (1) I used to trust and even like the Google, so now I have the enthusiasm of a convert to the opposite viewpoint. (2) I can be regarded as working for one of the Google's competitors, though these days that simply means I use computers at work and I don't actually work for the Google.
I'd actually be surprised if he gets the sack-- Google knew exactly who they were hiring and had to expect this kind of rant being published. After all, Steve Yegge is internet famous. If they do fire him, I'm sure he won't have trouble finding alternate occupation-- the guy is smart, opinionated, and frequently right.
Well, I am somewhat of a fan of Google's range of products, but as a user of social media and a person who uses different social media platforms for different types of interaction I have to say that I completely understand his rant about Google+.
The fact that Google+ had no APIs at launch confused me, the fact that it still only has one public API worries me and the fact that no one is really putting much effort into integrating Google+ into other things makes me think that it could fail... betamax style. AI say betamax because it is potentially a better way of interacting than Facebook, it has more flexibility than Twitter and yet because it is still an Island of content I struggle to integrate it as part of my lifestyle.
If they opened up and made it a platform instead of an application it might go somewhere, but significant momentum has already been lost.
Fire him. But not for the "Oops" - everyone makes mistakes - but for what he writes.
What he lauds about Google is that it gives the Oompla-loompas generous terms. Those same well respected and rewarded staff then turn around and write crap products (or in the case of Android allegedly forget to do due diligence on the code and applicable patents).
He berates Bezos for an obsession about the Amazon web site (one of the most successful in the world). Owners of a successful product have a tendency to keep doing the successful things in the face of criticism. Strange, that. But he doesn't compare or contrast the understandable obsession Larry and Sergey have for the Google search algorithms.
These same, down-trodden, poorly paid, disrespected Amazon employees, apparently using crap tools and haphazard standards have gone on to create a hugely successful cloud computing platform on top of the most successful retail site. Those haphazard standards and crap tools created a consistent and easy to use set of APIs for AWS that I use *every* day.
My take from the rant is that Google staff are overpaid, that the organization is too centralized and controlling so does not have the breadth of ideas and internal criticism to maintain itself in the future.
By contrast, Amazon's more haphazard organization is focused on some core essentials - the web site and software services - which has resulted in a more flexible organization. And I think that based on this rant I'd be more likely to hire an Amazon employee over an ex-Oomla-loompa.
I think Google stockholder ought to have some concern about the sentiments expressed.
Amazon makes it money from selling stuff. It has a large established distribution infrastructure for shifting physical goods as well as a large IT infrastructure to catalogue its inventory and collect payment for sales.
For Amazon, reselling its IT infrastructure is a win-win: it's an additional source of income and reduces the cost of running it just for Amazon's core business. Add to that the additional revenue from reselling its reseller infrastructure (payment services and fulfilment). It's core business isn't threatened - it's now so big that noone else is going to be rushing out to build the same scale of warehousing and only Apple has the same level of business to offer to suppliers of downloadable content.
Nearly all of Google's revenue comes from putting adverts in front of eyeballs and then watching what those eyeballs do. As soon as you abstract Google's services into APIs you remove the eyeballs and the revenues that go with them. What's more, the services for which Google charges are sofware applications which can pretty quickly be emulated and distributed across a cloud service - the only thing that distinguishes them is that they're from Google. Allow 3rd parties to build similar apps on top of the Google infrastructure and you're lending your brand to the competition.
That's not to say that Google's apps couldn't benefit from some extensibility, but that's about APIs that consume stuff from 3rd parties, not APIs that allow 3rd parties to consume stuff from Google.
Facebook isn't successful because of its constellation of products, it's successful because of four or five core in-house features which were executed well and along with some luck allowed it to reach critical mass.
In fact, Facebook's constellation of products is one of the prime reasons I closed my account.
I had read the Google+ part of it, but found the rest more interesting. It is surprising that internally Googlers like him are most concerned that they aren't adequately delivering a platform. I tend to see Google's larger problems revolving around legal issues related to its de facto monopoly on search/web advertising and Android, as well as a systemic need to know at least as much if not more about its users than Facebook does. The latter is by far the worst. Not long ago, I read a thoughtful article by a developer who pointed out the benefits of users paying for software or content, namely that there was no need for the company to bombard them with advertising or mine their personal habits. In this regard, Amazon has a few problems as well and is willingly heaping on more by selling the Fire at cost. But as Yegge alluded to, Amazon doesn't have a profit model predicated entirely on one type of revenue source - and let's face it, selling books, power tools and music is somewhat useful to society. Advertising is corrupting.
The problem is really broader than the programmer's view that Google doesn't deliver a platform. What Google doesn't have - even with Android, Google Office, Gmail and Google+ - is an ecosystem (whether narrow like Apple's or broad like Microsoft's) that provides a variety of revenue sources. Apple gets money from hardware sales (which are split between ioS devices and Macs), music and other media, and iOS apps. Microsoft gets money from hardware (Xbox, input devices), Windows, Office, Server, SQL Server, developer tools, consulting and hosting. Amazon also is a far cry from the online bookseller. By and large, Google is still a one-trick pony and you see it more and more when trying to do searches that advertising is skewing the results. And it's what you don't see that might be more concerning. Google has admitted and explains to users in the fine print of its agreements that every email, ever document, every picture you send through its servers is subject to analysis. Coupled with the patterns it can glean from the searches you do (whether public or private, if done close enough in time), click-throughs and locational data, Google can know a lot about it you. Of course what it knows is always less complete than if the details were directly provided by you, but like an erroneous credit report, that could be worse than if you just provided a complete profile on your own.
My point is, Google has to stoop to these things because it has to keep feeding a business model that not only narrow but prone to corruption. Its competitors have their own faults but at least have had the vision to actually create things that people buy (or in Amazon's case, a store where you can buy them) so they have a more varied and direct revenue model. Perhaps the object lesson for consumers is to think about paying up front instead of in a roundabout way. Either way, you're gonna have to serve somebody.
Biting the hand that feeds IT © 1998–2019