Uuuuuuh ?
By glazou on Monday 23 March 2009, 19:18 - Standards - Permalink
Yep. I just read last Stallman's article and I'm lost. Unless I completely missed something, this idea is totally stupid. So, let me try summarize his point : the web pages a user browses load/use JavaScript code. Free Software philosophy applied to the Web should provide the user with a way first to know the license of the JavaScript code a page is using/loading, second to replace that JavaScript code with a piece of his/her own. Fine. Great.
Now, the author of the web page modifies his/her page. Or even revamps it entirely. The old JavaScript code becomes irrelevant here but the new one keeps the URL of the old one. The user-modified version of the JavaScript code does not work because it's made for a page that is completely different. The Web page is then not browsable in a normal way and the whole user experience is trashed.
Can someone explain me what's good here in RMS's idea because I fail to see that part...
Comments
Why do you hate free software, Glazou? Did RMS not send you the Free Software Song CD you asked for?
@Matt
If RMS did like Woz and danced, I missed it.
BTW, what RMS proposed is different from Greasemonkey; GM adds scripts to the page, do not replace the existing ones.
So, hands up who wants to go to a site, and have their browser warn them that the source code isn't available?
Keeping in mind, of course, that there's no way of distinguishing a site that keeps thing secret from one that just doesn't declare it's status in a manner recognized by this hypothetical browser feature...
And yes, the idea of being able to drop in replacement scripts that conform to free-software ideals - that's just wacked in the head. Is he really suggesting that a site like GMail (or any other complex web-app) could be made to work with 3rd-party 'clean' javascript replacing the original?
@Daniel: You haven’t seen RMS dancing? http://www.youtube.com/watch?v=7C6r...
I don't really know if I like this or not yet...
But a solution could be a Firefox extension which would download new versions of a free version of gmail (for instance) ? Like APT in Debian...
But I feel something's not right: gmail, for instance, isn't just Javascript; it's both HTML and Javascript; I don't feel that changing the JS into free JS replacement would make gmail free.
I think he simply fails to understand that "JavaScript" isn't a self-contained "application". The client-side part isn't very meaningful by itself. And even if JavaScript is "all open and free", it can be easily written in such a way that the server side determines what this JavaScript will do. So this idea only makes sense if you view the entire web application as a whole and mark the web application as "free". But I don't think it gets much less crazy then.
> Can someone explain me what's good here in RMS's idea because I fail to see that part...
Actually, it's all about RMS' philosophical consistency. Software Freedom is the top priority for RMS, while usability and user experience come after.
You can disagree with RMS, but you can't dispute his consistency!
It seems that the best alternative would be use open-source email readers, such as Thunderbird.
Or unplugging your network pin :
No Non-free JS !!
I don't really see the problem. In fact, mashups are already sending the web in this direction. It's a fundamental principal that users can modify web content to suit their needs.
I increase text size and correct colors for many sites using CSS. For example, try reading this very blog with a white on black default theme!
I browse without script enabled, if I could substitute local javascript and enable this on a per site basis, I could make "web 2.0" applications work for me. It's no different than allowing users to override CSS and don't certain browsers already ship with JS hacks for common web sites?
Then we get to the question of distributing modified or derived code. To decry what stallman is saying is simply to miss the point. My CSS hacks are unlikely to be useful to anyone else, an alternate js interface to this weeks cool web app could be useful to others.
The problem here is assuming RMS is operating in the real world. He isn't, and hasn't been for a long time. It's a situation where you have a fantastic organization with a useful goal that makes a lot of progress, does great things, but then the game changes, and rather than adapting to the new environment, the organization's leader has ossified and refuses to adjust. RMS doesn't understand he's not fighting the same battle as 30 years ago, and won't adjust course. The movement has made incredible headway, but now, he needs to adjust or step aside, and he'll do neither. The movement now moves without him, and he doesn't like this.
He has admitted that he doesn't even use the web like ANYONE else. Someone sends him a link, he emails it to a daemon that wgets it, tars it up, and sends it back to him and he views it locally. The entire concept of AJAX is lost on him because he's even more asynchronous than the code.
RMS is irrelevant in today's web. He's a religious extremist who is incapable of compromise.
"He has admitted that he doesn't even use the web like ANYONE else".
Which doesn't make his point any less valid. Existing browsers and extensions don't provide enough control over scripting for advanced users. Mashups and offline apps may be the practical real world drivers in bringing the issue of "software freedom" (ick) for client side web components to the fore.
It's easy to dismiss Stallman as a crank but there's no denying his track record. Revisit "The Right to Read" for example.