From the article:

But free web services are not like free software. If your free software project suddenly gets popular, you gain resources: testers, developers and people willing to pitch in. If your free website takes off, you lose resources.

Perhaps thats a sign the webapp-for-things-that-don't-need-to-be-webapps model has flaws.

Nahhh. That can't be it.

Maybe nothing should be web apps... maybe the pipe is for content and services, and native code is for layout and interaction...

So are you willing to have native code pushed to you with no control over it? Because that's the major logistical advantage of webapps that has Enterprises trying to use them.

Maybe when NaCL gets better...

Untrusted native code isn't strictly necessary. There is no need for locally-run software to be in any particular form, so long as it doesn't depend on the developer's server to function.

Java bytecode is excellent precedent for what is broadly possible. There are problems with it, but most of what's missing for the end user is a better degree of convenience and control.

I think iOS software is a better example. There's no reason they can't be hosted in a Safari page, and they're rich while still being sandboxed.

Actually, the iOS model isn't that great IMO. Rather than relying on verifiably safe bytecode, Apple uses a combination of TrustedBSD-based technology on the phone, and human-directed testing in their labs, in an attempt to keep native code from doing things they don't want it to.

I think this is backwards for what they're trying to achieve, and I think that's part of the reason their app review process is such a pain in the ass.

But that argument is also flawed--free software in web applications is possible through other forms of revenue generation. Advertisements, services related to the product, in-app purchases....none of those options are precluded simply because the company chooses to respect their users. In fact, those options become a lot more prevalent, since there's no longer any payment for the service. But since the users saved that money, they will probably pay more for the extras--and they'll certainly be happy that they're allowed to change the product. This will also benefit the original site, since they get the changes made by everyone else.

The terminology is not good.

"Free website" often means "hugely profitable website"

That really depends on how far you stretch the term "often".

I think the number of unprofitable free websites vastly outnumber the "hugely profitable free websites".

Apologies yes you're right.

Often in terms of websites that exist on the internet.

Rarely in terms of ones that are mentioned on hackernews.

I think the spirit of what he is saying is pretty clear (free service), but sometimes he could use better terminology.

I was also confused at first because I usually think of the free software movement not like free beer but free like free speech.