Re: Reboot: Strategic goals for GNOME

[Jim Gettys <jg freedesktop org>]

The point I was trying to make was  that HTML 5 (or more formally some 
of the API's for javascript for accessing local storage), among other 
things, enables offline use of web  applications.  Think google gears 
use in google calendar and gmail or google air.  Note gears was just 
formally abandoned by Google in favor of standards (a good thing, IMHO).

It is becoming feasible to build applications with those technologies 
that you *can* take with you.  In this sense, they become no different 
than software we currently install in conventional ways based on GTK+; 
just more convenient.  This is part of the inflection point I believe is 
about to arrive.

I've seen little discussion on how gnome should be thinking of becoming 
part of an much larger ecosystem of applications that I believe will 
form due to this capability.

And yes, the end state may be that desktop environments do become 
entirely browser based; this may or may not be a "good thing" (either 
technically, or on software freedom grounds), depending on how the 
process plays out.

And without thinking this possible trend through, the probability of 
influencing this trend in "good" directions for the social good is 
greatly diminished.

Let me give you a concrete example common practice in javascript 
programs is to "obfuscate" the code
1) as a mechanism to "protect IP" (RMS, please don't jump on me here: 
I'm parroting how the companies involved are thinking about it), -
2) but also to improve compression of the loading of such programs 
initially. People like Google work *hard* on latency and understand 
every byte counts (among many other things: go look at the google talks 
by their engineers on the topic).

Right now, these are two disincentives for the source code to be 
available at all.

As a solution to 2), Gnome (and/or the FSF) could work in the web 
community to standardize mechanisms and code for making such source 
available.  So long as solutions to 2) do not exist, we're in a much 
poorer position; free and open source code should not work *worse* than 
proprietary, IMHO.

I'm concerned to have not seen this sort of strategic issue discussed 
widely.
			- Jim




Stormy Peters wrote:
> On Thu, Mar 4, 2010 at 6:46 AM, Richard Stallman <rms gnu org 
> <mailto:rms gnu org>> wrote:
>
>        The combination of technologies going under the name "HTML 5" have
>        made/are making web technology based applications finally competitive
>        with those built using conventional toolkits such as Qt, GTK+,
>     and the
>        Windows and Mac equivalents.
>
>     If everything gets done inside or through your browser, it would make
>     toolkits such as GTK and desktop environments such as GNOME obsolete,
>     except as platforms for a browser.
>
>
> I personally use a lot of web apps and I still need desktop and GNOME 
> type technologies both on my desktop and my phone. (And my netbook.) We 
> need to develop GNOME in ways that encourages developers of web apps to 
> consider it, either as building blocks or as a way to have an app 
> integrate with their service.
>
> We need to work on things like:
>
> a) Making our apps more web friendly, like Tomboy is doing with Snowy.
>
> b) Integrating with web apps in ways that make sense. Like Diego was 
> talking about with Facebook. Like things the Telepathy and Empathy teams 
> have been working on.
>
> c) Think about developing our own "free" web alternatives like identi.ca 
> <http://identi.ca> did. I'd especially like to see an open alternative 
> to Dropbox/Ubuntu One. But there are lots and lots of web apps that 
> people use regularly that could use alternatives. Social networking, 
> finances, managing collections, ...
>
> Stormy