Asa Dotzler: Firefox and more

You know when they react that fast and that aggressively, it means only one thing.

They feel really endangered.

This is indeed great proof Prism is going in the right direction.

I suppose Mike Chambers will disagree with this post as well, because you write "correctly pointed out that Adobe's Air is a proprietary end-run around the Web". How is trying to introduce an alternative for WinForms or .Net applications, in the guise of AIR applications, an "end-run around the Web". Should desktop applications be forbidden to use and integrate web technologies in some way? Or am I missing some point here?

BTW, I understand and agree with all other sentences in this post :)

I wish anonymous would have left his name because I agree. We're not trying to end-run around the web and I think especially in the context of AIR it's being disingenuous to imply that. AIR is about building desktop applications, something you've *never* been able to do with "the web". But we love the web, and we see value in all of the web technologies out there, so we wanted to enable all of those peoples to take existing applications and skills and bring them to the desktop.

I also don't think it's fair to equate a browser with the web, but that's an argument for later. I think Prism is great, I think it validates a desktop approach. I think more functionality is better, but in this case, people get a lot of good choices.

=Ryan
rstewart@adobe.com

Well, the comment quoted is taken out of context. it was in response to this comment:

--
I think the test should be, will “AIR” applications work with any standards compliant browser?
--

From the Prism article, it appears that Mozilla will be adding features that won't work in all browsers.

The original prism article suggested that they would be moving closer to the desktop:

http://labs.mozilla.com/2007/10/prism/

--
And while Prism focuses on how web apps can integrate into the desktop experience, we’re also working to increase the capabilities of those apps by adding functionality to the Web itself, such as providing support for offline data storage and access to 3D graphics hardware.
--

(Now, I don't have a problem either way, but the poster of the original thread was suggesting that that was the acid test of whether a runtime was trying to "replace the web".)

Also, If you read the original blog post, it was not knocking Prism or an attempt to spread FUD. I as posted in my comment:

--
but my post wasn’t about how people were reacting to Prism, but rather, about Mozilla’s characterization of Adobe’s efforts as trying to “replace the web”. I found this odd, as we havent done anything to indicate that, and there is a lot of overlap between Prism and AIR.

There are a lot of comments in the thread, which is a positive thing, but very few of them adress the post. I understand why someone might choose a 100% open source project vs one that is not completely open source. Again, I was questioning mozilla’s characterization of Adobe’s intentions around Adobe AIR. A characterization they made without any explanation.
--

Btw, I may be misinformed (I only have what has been posted by Mozilla), but I can promise you that I am not trying to spread any FUD (I don't work for Microsoft... ;).

mike chambers

mesh@adobe.com

>From the Prism article, it appears that Mozilla will be adding features that >won't work in all browsers.

Here is what we said in the announcement:

"Prism isn’t a new platform, it’s simply the web platform integrated into the desktop experience. Web developers don’t have to target it separately, because any application that can run in a modern standards-compliant web browser can run in Prism."

So, you are right in that we didn't say "all browsers" (which I guess would include browsers from the past as well). However, as we add features to Prism, those features will be pushed through standards bodies, and they will be uplifted into the next release of Firefox, as well as other standards compliant Web browsers. For instance, Webkit just announced client side database storage http://webkit.org/blog/126/webkit-does-html5-client-side-database-storage/

My impression is that the canvas3d stuff isn't as far along with the standards groups as offline storage is, but the overall point is any capabilities we give application developers in Prism will also apply to the Web itself. We probably should have elaborated on this point a little more to make sure there wasn't too much confusion after the post.

One technicality is that a feature may arrive in Prism before it arrives in Firefox because Firefox gets a new version every year or so, and Prism is a labs experiment that doesn't have a formally defined ship schedule.

Mike,

I think the point is that Prism doesn't push Flash+Flex. While it's true that Adobe AIR can run standard HTML+Javascript, the intent behind AIR seems to be to lure developers to Flash+Flex. Even for HTML+Javascript, there are all kinds of non-standard libraries made available in Adobe AIR that would take some discipline to avoid using.

Prism is a great effort and it is more pure from the openness and freedom point of view.

Adobe AIR is an interesting technology. Adobe AIR is not open source. Neither is Flash. Flash is not standard. Many people hate Flash. Flash is what brought us these annoying advertisements that you pretty much never see anymore. So there are some negative association there. Also Adobe is known for being hostile to customers. I speak as an Adobe customer now. Adobe is a large corp behemoth. While I am absolutely positive there are nice people and great developers working inside of it, somewhere deep in the bowels of Adobe, Adobe is still just another unwieldy corp and nothing more. Adobe AIR is in a tough spot.

"My impression is that the canvas3d stuff isn't as far along with the standards groups as offline storage is..."

Let's not get ahead of ourselves. *None* of the "HTML5" stuff (including DOM Storage, Canvas, etc.) has cleared *any* standards process yet - not even in a preliminary form. It's been taken up by the W3C's HTML Working Group as the base of their next version of the HTML spec, but it's still under debate and refinement in that body and for all we know the final product will look very different from the original WHATWG "HTML5" spec.

In that sense none of it is "standards based" as there is no formal standard to base it upon -- just a specification ginned up by the WHATWG, aka Mozilla-Opera-Apple. I'm a little concerned therefore to see it showing up in browsers, as it seems like a "preemptive strike" to keep the working group from diverging too far from the WHATWG spec ("You can't change it on us now! We have X million people out there using it the way it is!"). That feels too much like the bad old days when browser vendors stuffed all sorts of new features into their products, and then waited for the rest of the industry to play catch-up.

(If it's not "standardized", though, it is at least open, in the sense that anyone who wants to can download the WHATWG "HTML5" specification and build their own implementation of it. Which is more than you can say for AIR.)

Jason,

Your suggestion that the WHATWG isn't a standards process is laughable. The W3C doesn't own the Web. The W3C has been AWOL on HTML for years and just now returned to the party. The browser vendors moved on and it's time for the W3C to play catch-up. When the W3C's HTML WG has earned back its reputation by good deeds over time, maybe then they can claim ownership of HTML without being laughed at.

- A

I don't know. I, too, find it a little dubious that Mozilla is preaching about standards when they're part of the standards-making body and railing against proprietary software vendors. Seems disingenuous.

And if features added to Prism pass through the "standards" body and get adopted by Webkit, doesn't that also help out Adobe in the long run?

"The W3C doesn't own the Web. The W3C has been AWOL on HTML for years and just now returned to the party."

I don't claim that the W3C "owns the Web", and I certainly don't disagree that the W3C has been shamefully absent in keeping its core specification up to date. Of course it has.

I do claim that when we talk about "Web standards" to date we have generally been talking about standards that have been processed and ratified by a major standards body, such as the W3C or the IETF, not just an informal coalition of vendors.

The distinction is similar to that which exists today in Web feed formats. There is an open specification for RSS 2.0 that anyone can download and implement. But Atom is generally considered the "standards based" option, because it has been formalized by the IETF. And we can see the value in that formalization from the fact that RSS 2.0 has begun to splinter, with Dave Winer and the "RSS Advisory Board" -- an ad hoc standards committee -- publishing competing, incompatible specs under the name "RSS 2.0".

We've seen what happens when vendors rush to implement early drafts of a spec already. The first browsers that supported CSS had wildly different implementations because the vendors couldn't wait for the standard to be 'set' before shipping. That caused incompatibilities that Web authors are _still_ having to deal with today, coming up on 10 years later.

What I fear is that we are going to see a similar outcome with vendors rushing to ship product based on preliminary versions of the HTML5 spec. And finding out that the answer is "WHATWG r00lz, W3C dr00lz" doesn't make that fear go away.

Just my $0.02, for whatever it's worth (my guess is not much, but you never know)...

Asa, will you be a the technical plenary in Boston. Next Week.
I wish you would, because I think there are topics that would be worth discussing more than both hidden behind a keyboard :)