This is not exactly breaking news, but once upon a time there was this company called Netscape that thought it could beat Microsoft. 'Jane Galt' reviews the episode, in response to a write-up by Steven den Beste. They both make the argument that Netscape should not have been so brazen about being out to create a new computing platform.
If they wanted to be the next Microsoft, Netscape should have looked more closely at how Bill Gates did it -- he kept his mouth shut until it was too late for IBM to respond.
Second, it is not clear that Netscape's key backers cared whether or not Netscape won the platform war. If you're Jim Clark or Kleiner-Perkins, you can make a lot of money by convincing Wall Street that you are going to be the next Microsoft, even if you have no way of achieving that objective. So you "moon Microsoft" to pump up the stock, and then sell it. I think that there was a lot of that going on with Netscape.
Which brings me to what I think is the biggest myth about Netscape, which is that they were a great engineering company. They were a crappy engineering company. Their web server software, which had the biggest potential for earning revenue, was so bad that it got no traction even though for a couple of years there was no meaningful competition. The Netscape web site did not even use Netscape's proprietary scripting language--the Netscape webmasters could not deal with all the bugs. The company would not even eat its own dogfood.
Netscape was what economist Burton Malkiel used to call a "Castle in the Sky." In the early editions of his book A Random Walk Down Wall Street, Malkiel used that term to describe a stock that fires the investor's imagination. Netscape was conceived, designed, and executed as a "castle in the sky." There was no foundation.
The Wall Streeters that Eliot Spitzer and other crusaders are going after are petty criminals. The multi-billion-dollar con artists, like John Doerr and other venture capitalists who created the castles in the sky, are untouched.
Well said, Mr. Kling. Also: Why is James Barksdale considered savvy in some quarters, instead of the laughingstock he actually is?
Posted by timks on November 13, 2003 02:28 PM | Permalink to CommentTim,
Probably because James is rich. So he lost a lot of other people's money. He still got his. Sounds like he was savvier than I am.
Some Guy
I'd find this to be an interesting perspective if I didn't disagree with your technical assessment of the Netscape Enterprise Server software. This software is still in use, and I have worked with it on a variety of projects throughout my career, including a major sports entertainment property that I am currently working on. (I am in no way affiliated with the team that made the software, and I have never worked for Netscape.) If the software is robust and stable enough to serve CNN.com, I would venture to say it is not crappy. As a Web Application server, it was not a very good platform, but it works fine with some of the major Web application software that is currently available.
Posted by Robert Occhialini on November 16, 2003 09:26 PM | Permalink to Comment"They were a crappy engineering company."
I'm not sure why you say this, you don't elaborate as to why specifically. The Netscape engineers were some of the brightest and most talented people in the industry, many of them still are. I'm not talking about the "rockstar overnight millionaires." I'm talking about scores of everyday Joes and Janes. They had good product support, they had good channel support and they had decent customer support. No, they didn't lead the industry in any of those categories, including product development but, they were no worse than Sun, whom they learned most it from.
"Their web server software, which had the biggest potential for earning revenue, was so bad that it got no traction even though for a couple of years there was no meaningful competition."
It was the only meaningful, professionally supported enterprise web server software available for many years. IIS was unstable and insecure, buggy garbage. Apache was mostly for the hosting crowd and hackers. At one point, (1995-1999) nearly 90% of the largest 500 web sites in the world were running it. In fact, far more websites than we realize still use Netscape's Enterprise Server. I agree with Robert's statements above and I also am not affiliated with NS in any way.
"The Netscape web site did not even use Netscape's proprietary scripting language--the Netscape webmasters could not deal with all the bugs. The company would not even eat its own dogfood."
What are you referring to? Livescript? Which obviously became Javascript? That's the only scripting language that Netscape developed that I can think of... This is far too much of a blanket statement.
You are right, they did loose the platform war but, not at the end-user level as many people mistakenly believe. That war was was lost the minute the Microsoft engineers grabbed the source code for Mosaic and compiled IE 1.0. They lost the war at the server level. Which was a war they could have won, were it not for AOL's ill-conceived, ill-regarded and greedy "land grab" in 1999.
At that time, the Netscape server products moved entirely under Sun's guidance and control, who had little interest in putting the necessary resources behind capturing more web server market share. Many customers and channel partners simply called it quits at the inception of iPlanet scheme. Had they won that battle, at the server level, they would have won the war forever, making them the perfect candidate for purchase by Microsoft... the ultimate indication of victory over Microsoft.
Finally, as for the greed, stock price pumping and other non-sense. They did nothing more, for better or, worse... then the majority of their "Valley" counterparts. Most of whom, I suspect, are patiently waiting to do it all over again. I empathize with your sentiment now as strongly as I did while it was happening. The MBA's profited, while the technical people perished, it was a sad sight indeed which we've seen before and we'll see again.
Posted by Dean Paxton on November 17, 2003 11:25 AM | Permalink to Comment"it is not clear that Netscape's key backers cared whether or not Netscape won the platform war. If you're Jim Clark or Kleiner-Perkins, you can make a lot of money by convincing Wall Street that you are going to be the next Microsoft, even if you have no way of achieving that objective. So you "moon Microsoft" to pump up the stock, and then sell it. I think that there was a lot of that going on with Netscape."
...
The Wall Streeters that Eliot Spitzer and other crusaders are going after are petty criminals. The multi-billion-dollar con artists, like John Doerr and other venture capitalists who created the castles in the sky, are untouched."
How come Clark and Doerr are still regarded as demi-gods, then?
I would ask Arnold Kling what his evidence is for asserting that Netscape was a crappy engineering company and its web server crappy. It is easy to fling allegations -- what's your evidence?
Posted by Tim on November 18, 2003 01:52 AM | Permalink to CommentThe Netscape people were quite candid about how bad their software was. At one time the browser contained five different HTML parsers because they didn't write modular code that could be re-used. So every time they wanted to handle HTML they had to start from scratch. This, in a program whose *purpose* is to parse and render HTML.
Posted by Dave Winer on November 18, 2003 09:04 AM | Permalink to CommentTim asks, where is my evidence?
My evidence is that I tried to use the server to power a commercial web site, www.homefair.com. Scripts written in Netscape's Livescript caused the server to crash constantly. They could never diagnose the problem.
I was very active in the server discussion boards. You had to *pay* to join them, and the Netscape staff refused to participate. If I had saved the reports from those discussion boards, that would constitute a mountain of evidence.
I speak from bitter experience. Very bitter. The bitterest.
Netscape's server was so bad that we made a rapid switch to the JavaWebServer from Sun, which meant rewriting all our applications and using a product that Sun said was not really meant for production use. It worked fine.
Posted by Arnold Kling on November 18, 2003 04:51 PM | Permalink to CommentI would tend to agree that Netscape was, at end of their life, a crappy engineering company, but I would modify that statement by saying that it wasn't totally the engineers fault. They had a lot of great engineers, just bad management and lack of process. I also think the "crappy" appearance was more real on the enterprise/server side than on the consumer/browser side of the world.
I worked in Netscape Professional Services for a couple of years as a contractor. It never ceased to amaze my how screwed up the relationships with their strategic accounts could get. I think the only reason they worked at all with Netscape was because, at the time, they had to. Netscape was an innovator at that point of time (95/96 time period), and a lot of major corporations started jumping on the Internet bandwagon.
It was really a combination of factors that brought about Netscapes demise:
1) Promises made by Netscape sales executives to accounts that weren't realistic and couldn't be kept;
2) Netscape engineering/product development releasing buggy, inadequately-tested software products ahead of their time (Netscape was a victim of their own mantra of "Internet time");
3) Sloppy handling of accounts by professional services in which they had no/poor follow-up of custom development done for clients (I wished I had a penny for every unhappy customer);
4) Burnout of Andreesen and other key, founding engineers, starting in late '96.
5) Some bad acquisitions, in retrospect, like Kivasoft.
Netscape did make some feeble attempts in '97 to turn things around, such as getting Andreesen more involved again and focusing on quality and customer support more, but by that time it was "too little, too late"; they're reputation had already been badly damaged by that point. Netscape was fast out of the gate and had "first mover advantage" (talk about dot com buzzwords), but they never really pulled it together as a company, and never really made the transition from unruly, but exciting and energetic, start-up to well-managed large corporation.
In the end, the mighty ogre which ate up Netscape was not Microsoft's unfair business practices (which a lot of Netscape executives would like you to believe), but Netscape's own inability to sell, develop and deliver to customers their own products. In other words, it was Netscape's own inability to execute which tripped it up; by contrast, Microsoft was a much better managed company internally, even if its products did suck.
I think the only thing ex-CEO Jim Barksdale was good at was public relations. The acquisition of Netscape by AOL in '98 for 4 billion dollars, was an extremely good exit stratgey for Netscape, allowing them to save face (as opposed to an ugly crash and burn which was about to happen). By getting swallowed by a much bigger company, the real problems inside Netscape never became too visible. The only people who knew Netscape's real problems was Netscape themselves and their many, unhappy customers that they left in the lurch. Eventually, their Enterprise products languished from a lack of support and marketing, and soon disappeared from usage.
These days, Netscape is really nothing more than a portal, with it's Enterprise business dead and it's browser completely irrelevant. If you use Windows, you most likely use IE as your browser, and if you use a Mac, you probably use either IE or Safari.
In all fairness, Netscape did play a very important role in the development and popularity of the Internet, and their early engineering team was outstanding, and they should be given credit for that.
I agree that the problem at Netscape was one of process, not of lack of engineering talent. Managing a software development process that involves multiple people and multiple projects is *really hard*
I think that Netscape had some good visionaries, and some good coders. It had no management process at all. However, they faked it really well, getting a couple of Harvard profs to write a fawning book about them.
Posted by Arnold Kling on November 25, 2003 11:02 AM | Permalink to CommentI personally think, that just the huge capital behind Microsoft is the problem, and of course a bit lack of management in the team of Netscape, but that's just my 3 cents - have a good start into 2004!
Posted by Dave on December 29, 2003 04:03 PM | Permalink to CommentHow can Netscape can be the next microsoft when they cant write good browser at all. Well, MS cant do it either :o)
Posted by Petr on July 7, 2004 03:20 PM | Permalink to Comment
I remember downloading the netscape source when it became available.. what a mess, like a lot of code out there. why doesn't the business press ever make an issue of what a company's code looks like, is it because the code just isn't available? I seriously believe that any reasonably intelligent person (ok, maybe someone who's written some script) should be able to look at a snippet of source code, say a single function or a class interface, and tell you what that piece of code does.
Posted by greg on November 11, 2003 07:58 PM | Permalink to Comment