Marc Andreessen:'Every Single Idea From The Dotcom Era Was Correct' | Hacker News Reader

Marc Andreessen:'Every Single Idea From The Dotcom Era Was Correct'

statspotting.com

15 pointsnpguy23 comments

I just wish that the guy who did the browser did a better job. I agree with Alan Kay when he said "the Internet was done so well that most people think of it as a natural resource", and "The Web, in comparison, is a joke. The Web was done by amateurs."[1]

Here's an idea: Let's fix the web. I'm not talking about Web 3.0. Let's actually fix the web. Let's get rid of the browser. Let's stop building stuff on top of HTML, CSS, JavaScript, all of this so-called "technology". Let's fix the foundation first. And let's try to get it right this time.

I'm tired of my computer with 4gb of ram, 2 ghz processor slowing to a crawl whenever I open the browser, to the point where I'm reminded of typing over a 2400 baud connection.

Anybody know if it's possible to get the old Xerox Star OS running in an emulator? That seems like it might be a good place to start.

[1]http://www.drdobbs.com/architecture-and-design/interview-wit...

"I just wish that the guy who did the browser did a better job."

Amen! Last night I watched an "expert" explain how the thirty lines of Node.js that reinvents a SQL Join statement on the server just to get a some simple data to a browser on a phone.

Biatch please! The web still hasn't replicated the desktop apps we had 20 years ago. It's busted when you have to glue stacks of APIs with 4 or 5 separate programming languages together just to get a simple app to work. It's enough to make me run from phone apps and stick to the CLI and "normal" desktop GUI.

What browser are you using, and what websites are you visiting? Because one of the two is the cause of your slowing-to-a-crawl, and if you want to vote with your wallet you should start with that before throwing away the entire web ecosystem.

It's not necessarily just one of them that's responsible for the poor performance. It could easily be both. Running bad JavaScript using a bad interpreter leads to awful performance.

Having been part of that era I think the biggest problem was the lack of broadband for home users. We'd be sitting in our offices with a T1 line and only a small fraction of folks were even using 56k dial up modems, so the only folks looking at sites were other folks in offices with T1 connections. What's funny is that broadband started to really take off right after the dot.com bust around 2002 which was a gap of a year or two.

Makes one think about ideas that are ahead of their time now, but will make sense when Gigabit broadband is widespread.

You could apply this to other areas in tech too, e.g. car OSes, which are new and not great (like 2002 phones versus 2012 phones), but in 10 years will be a key selling point in cars - over the air firmware updates, every car has it's own IP address, high-speed mobile broadband as standard, detects if you're too drunk and dials a cab, auto-dials emergency services if you crash (some of these are already in development, there are probably lots more features that are hard to imagine just like with phones and tablets).

basically, dial-up modems made the internet access a on/off cumbersome service. broadband routers and wifi provided an always-on internet at homes. this was the main actor for post-2003 boost, i guess.

Imagine only having one landline in your house before the era of cell phones: You have to use that one line to go on the internet, so if anyone calls you they get a busy signal. Yes if you were a geek you'd get that extra phone line like I did, but most normal people didn't do that. Having a 24/7 live net connection changes everything.

Mobile is a pretty big thing, too. PDAs kind of sucked, but even the original Paypal (device to device financial transactions) makes sense now. Mobile devices are better, and more importantly, carriers have less power over the ecosystem (due to the iPhone breaking their power, and Android expanding the opportunity afterward).

That is such a simple but powerful observation. Like, you can have the greatest product on earth but if you cannot be available at the point of sale you have nothing.

You know you are a senior computer scientist when people refer to events you witnessed first-hand using the words "that era".

pmarca is much more correct on this than fredwilson. I think there were some dotcom ideas which were inherently bad even skill (beanz/flooz...a fully closed, non-convertible currency for the web is silly; a more open system is reasonable, which is what causes the limited success of bitcoin so far).

But the ideas were rarely the cause of failure of those companies; it was the execution. Webvan was an awesome idea; they just seriously fucked up execution by overbuilding on capital equipment (warehouses) and expanding to areas which were inherently unsuitable for early launch (anywhere where anyone rich enough to be a good customer had a car and drove daily anyway).

Tesco (http://www.tesco.com/) has applied the Webvan strategy, except in reverse - start with a chain of stores, and add the online and home delivery element later. They have the brand loyalty, the consumer attention and the infrastructure - so adding the online element is an obvious next step. Seems to be working out well for them.

(Tesco also tried to expand to the US but failed: http://www.guardian.co.uk/business/2012/dec/09/fresh-not-eas...)

I think the "pay a higher per-order cost to have a human pick items from the shelves at the grocery" (which was also the strategy of Safeway) is the best way to enter the market.

Webvan could have justified a big warehouse in markets like SFBA once they had demand.

I didn't realize Tesco had folded; I thought they were doing a partnership with Safeway or something. I guess I didn't watch US grocery stores much 2003-2010.

(I wish I hadn't lost the 30 webvan crates I'd accumulated in the whole Sealand/HavenCo thing. They were awesome. I also bought a few of the Napster servers at auction and used them for a few years.)

If that's the case, can a founder who failed in the dotcom era repeats his idea and might be successful?

If we can find an example of this, it would be awesome.

Dodgeball/Foursquare is the canonical example. dens is relentless. (dodgeball wasn't so much a failure, but was a smaller success than 4sq to date)

I hope the things I did/failed at during dotcom (a datahaven beyond the reach of nation-states, and electronic cash) end up being things I can do successfully in the future.

Great example. demonstrates the fact that if you keep solving problems in one domain / area, one day you will solve a really painful problem profitably

Maybe -- it's not clear that 4sq is a winner, but if there were any single person who I would bet on coming up with a profitable mobile location service, it would be him. If not 4sq, whatever he does next.

As far as I can tell, Foursquare has basically pivoted to become a Yelp competitor. I was excited about it initially because I thought they were going to do cool things to enable real-world socialization, but there's really none of that now. It's all about the places, not the people.

It's going to take a completely new approach to make location-based social networking a world changer like Facebook was.

From what I've read, ingress might be version 0.01 of that. I don't have an android device to try it out though.

In my opinion it hugely depends on idea implementation. Good team with bad idea can turn it in a profitable business and the opposite.

I'm with Wilson on this, and I hate this sort of rhetoric in general, so I'm frustrated to see Hacker News eating it up time and time again. It's not thought-provoking to make claims that everyone knows are plain absurd. I understand Andreessen may not literally mean this and instead has some underlying points about innovations in the Dotcom era, but then he should just say them straight.

All this is good for is making some easy views with an "inciting" blogspam title - which the authors were clearly relying on, as the article doesn't have much substance beyond the quotes.

Reply on news.ycombinator.com