The Joel on Software Discussion Group (CLOSED)

A place to discuss Joel on Software. Now closed.

This community works best when people use their real names. Please register for a free account.

Other Groups:
Joel on Software
Business of Software
Design of Software (CLOSED)
.NET Questions (CLOSED)
TechInterview.org
CityDesk
FogBugz
Fog Creek Copilot


The Old Forum


Your hosts:
Albert D. Kallal
Li-Fan Chen
Stephen Jones

Is the Magical Fairy-tale For Google Engineers about to End?

So we all know about what a fairy tale place Google is supposed to be for software engineers. A coding cowboy's wet-dream. All smart people, and very little hands-on management. Kind of like Microsoft back in the day.

It is my impression that with more and more people working there, and as more and more projects are being pursued the quality control is starting to slip. The fact that they don't seem to care as much for experienced engineers and project managers, as for high-IQ coding cowboys seems to be exhibiting itself in a sinking quality to the Google beta products.

I've had pretty crappy experiences using the Google gbase API and the Google co-op service (I go into more detail on my blog if you care: http://cdiggins.com/2006/12/02/google-custom-search-engine-brilliant-idea-poor-execution/ )

So what do you think? Can Google continue merrily down the yellow brick road, or do you think there are going to have to accept the boring reality and neccessity of quality assurance and proper project management?

What I am trying to say, is that I think the Google approach to software development is not sustainable at a large scale and will collapse in on itself unless they make some big changes.
Christopher Diggins Send private email
Saturday, December 02, 2006
 
 
Can you give an example of a company Google's size that has an alternative engineering process and gets better results?

I'm no google fanboy, but nothing comes to mind.
_
Saturday, December 02, 2006
 
 
search seems to be a good product. taking the overture ad model, which they mocked when it was introduced, was a smart move.

most of the rest is also ran, derivative, incremental, or broken:  google base, froogle, gmail, blogger, google maps, orkut.
PPC Tony
Saturday, December 02, 2006
 
 
Also, you might be confused about how Google operates.

Search and ads are run like a machine and the engineers on those projects get paid about double everyone else. They also get double the workload and triple the stress and pressure. Think "investment banking."

Non-engineering is like any old company, with average salaries and "The Office" type environment, but with Google perks like free lunch, puppies, nerf guns, hand jobs, dignity replacement therapy, etc.

The rest of engineering is basically glorified R&D with pretty good salaries and the looser management style you may be imagining.

No idea about sales and execs, wasn't exposed to them.
_
Saturday, December 02, 2006
 
 
"dignity replacement therapy"

I'm keeping that one to use later.
Matt Conrad Send private email
Saturday, December 02, 2006
 
 
"Can you give an example of a company Google's size that has an alternative engineering process and gets better results?"

Gosh no. I am just throwing stones. ;-)

Masybe they will be able adapt to their new size, and relatively recent obligation to create profit and growth for shareholders.

There will definitely have to be a significant shift in approach, and it should be interesting to see what they do.
Christopher Diggins Send private email
Saturday, December 02, 2006
 
 
"search seems to be a good product. taking the overture ad model, which they mocked when it was introduced, was a smart move."

I think calling it a "smart move" is unfairly assigning credit to the Google founders. In the beginning IIUC there wasn't any planning with regards to how they were going to be profitable. Their business plan was essentially:

1) build a better search engine
2) ???
3) proft
Christopher Diggins Send private email
Saturday, December 02, 2006
 
 
Forgive me, I was caught up short by the OP.  An engineer's "*wet* dream"?  What's with the crudeness?
Kyralessa Send private email
Saturday, December 02, 2006
 
 
Christopher,

Exactly my point. They had a nice search engine and no idea how to make money. Goto.com introduced paid search placement. Google mocks them briefly, then wakes up and copies the idea.  Smart move. No credit for the idea.
PPC Tony
Saturday, December 02, 2006
 
 
"Forgive me, I was caught up short by the OP.  An engineer's "*wet* dream"?  What's with the crudeness?"

I'm sorry for being crude. I'd correct it if I could.
Christopher Diggins Send private email
Sunday, December 03, 2006
 
 
"Forgive me, I was caught up short by the OP.  An engineer's "*wet* dream"?  What's with the crudeness?"

Why so uptight?
Steve Prefontaine
Sunday, December 03, 2006
 
 
I almost splooged on my keyboard when I heard I got an offer from Google. But then after working there a few months, I realized it was truly a don't-drop-the-soap-in-the-shower kind of place to work.
typical internet discourse
Sunday, December 03, 2006
 
 
"Goto.com introduced paid search placement. Google mocks them briefly, then wakes up and copies the idea."

Since when did Google have paid search placement?  I thought they were pretty rigorous about making a clear distinction between ads and search results.
Iago
Sunday, December 03, 2006
 
 
Well, no matter how cool they think they are, if they continue to hire noobs based on IQ and the stature of the school they attended, they will continue to have problems.

Google has been lucky so far, in that they haven't had to offer true customer service. When that day comes, they will be in for a rude awakening, because the noobs will have no idea how to deal with customers, or to produce software even Skooter can run. And until then, they are working in a kind of unrealistic software dev bubble.
MBJ Send private email
Monday, December 04, 2006
 
 
What kind of customer service does a web site need to provide? For most of the users the services are FREE so I'm pretty sure the "customers" are getting a lot more than they pay for, even if things malfunction from time to time.

Their paying customers, people buying advertising, are purchasing a completely nebulous service with results that can't be controlled by google anyway. You can purchase all the keywords you want, but that doesn't mean anyone will click on them.

They will never be in for a "rude awakening" because they already have all the money in the universe, and just keep making more. They will probably eventually fade away into lameness like Microsoft is starting to, but that's going to take decades and even Microsoft seems to be doing OK.
clams
Monday, December 04, 2006
 
 
> So what do you think? Can Google continue merrily down the yellow brick road, or do you think there are going to have to accept the boring reality and neccessity of quality assurance and proper project management?

>> "Can you give an example of a company Google's size that has an alternative engineering process and gets better results?"
> Gosh no. I am just throwing stones. ;-)

Then why would Google replace their current approach with a more traditional one if evidence suggests that traditional practices do not achieve better results?
Bean
Monday, December 04, 2006
 
 
"When that day comes, they will be in for a rude awakening, because the noobs will have no idea how to deal with customers, or to produce software even Skooter can run."

Like clam pointed out, Google's customers are the advertisers.  "Skooter" is a user.  Just like with ad-supported broadcast TV, you're not the customer, you're the product.
Drew K
Monday, December 04, 2006
 
 
I'm interested to see how things turn out for Google.

Their development model is to try to turn developers into entrepreneurs.  Reputedly, they provide stock incentives to developers to create successful products and then let developers self-organize to come up with both the ideas and implementations for new products.

I'm skeptical.  At best, their development model may work for them but I'm convinced that it wouldn't work at most places.  Employees just aren't entrepreneurs.  Even though they may be smart, hard-working, insightful and "get things done", employees don't usually come up with good businesses on their own.  They often just want to reinvent the same products they've seen at other companies or work on things that are idealistic but have questionable commercial appeal.  Being specialists, they often don't adapt to being generalists and doing whatever odd jobs need to be done in order to bring a product (company) together.

But, even more, when you make these kinds of environments, the place can easily split into tribes (a.k.a. high school cliques, prison gangs).  I think that the anonymous post above hints at this.  I've seen various degrees of it at other companies and in my personal experience.  Instead of self-organizing to come up with new products, people self-organize to have something to do, have somebody to hang out with and have a group to protect them.  Companies have to be on guard for this behavior; it is destructive when unchecked.  Friendly economic competition is one thing; unchecked social and political competition is quite another.

In my opinion, traditional command-and-control management may not be much fun but works better and is more effective.

Will Google be able to pull this off?

It's possible but, even if they do, I don't expect many companies to follow their lead nor do I expect impressive results.
Daniel Howard Send private email
Monday, December 04, 2006
 
 
""Forgive me, I was caught up short by the OP.  An engineer's "*wet* dream"?  What's with the crudeness?"

Why so uptight?"

Not to put words in Kyralessa's mouth, but perhaps she was annoyed by the OP's not-so-subtle implication that all engineers are male.
anon y mouse Send private email
Monday, December 04, 2006
 
 
+1 to - (yes, that's his name apparently.)

As I've said before, Google really wants to become the next ABC/Disney/ESPN. Or perhaps the HBO network is a better analogy.

They don't necessarily want to create content per se, but they want to be the first place you go to, to look for content. Everything else is considered a loss leader.

Let's use my favorite example, Star Wars. Let's say you're a Star Wars fanatic. You can go to Google search and look for stuff (which is obvious). You can also join SW discussion forums (GMail, Blogs), shop for SW merchandise (Froogle), find SW wallpapers, screensavers and images (Images), download SW videos (Youtube), find conventions and other fans (Earth, Maps), get news (News), reminders (Alerts), read books (the book scanning and searching thing), share stuff (Desktop), etc etc, ad nausem.

Basically, you can go to one site, type in Star Wars, and it instantly becomes a portal for everything Star Wars related. But like ABC/Disney/ESPN, they don't provide the content. They just make it easy to get, and they charge for the advertising superimposed.

With respect to the original poster's question, yes, Google might tighten up the "non-engineering" side depending on how valued each loss leader is, but no, I don't expect them to develop a Microsoft style mentality towards profit overnight.
TheDavid
Monday, December 04, 2006
 
 
I agree with theDavid however for that strategy to work Google must keep the main business - search - relevant and functional.  Sadly that is debatable now.  Googles results are less and less useful than they used to be.  They need a :nobogs and :nolinkfarms option.

I like gmail and google is still useful, I just think their current dominance is going to slip away from them because they're not addressing issues with their main offering.
Anon
Tuesday, December 05, 2006
 
 
>What kind of customer service does a web site need to provide?

They are more than search today. They have a spreadsheet, e-mails, and a lot more. If they want their customer base to expand beyond people like us, they need to offer real support.
MBJ Send private email
Wednesday, December 06, 2006
 
 

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics
 
Powered by FogBugz