[Greasemonkey] Question: what would happen if Greasemonkey sentthe ID of every applicable user script with each request?

Julien Couvreur julien.couvreur at gmail.com
Sat Apr 30 11:19:27 EDT 2005

On 4/30/05, Ben Hyde <bhyde at pobox.com> wrote:
> Revealing grease monkey configuration to the site is contrary to monkey
> nature.  These scripts stand outside the contracted part of the
> exchange.  There are no interoperability problems for grease monkey
> scripts, by definition.  

What about considering the greasemonkey "nature" as being a way to
evolve the web?
Consider it an experimental ground for the long tail of features. 

>From one of my previous post on this thread: "I think the most
interesting thing that publishers can do with the information is: look
up the scripts that are being used on their sites and learn from them,
possibly making them easier or integrating these features in their
website. They would also learn how popular each script is."

Advertising the GM scripts being run on a page is mostly not about
stability. If 50% of a site's users were using a given user script,
maybe that shows a problem in the site or cool feature that is worth
Maybe they would also implement it in a less hacky way than GM scripts
do, leading to a bit more semantic markup (which in turn helps future
GM scripts). You create a virtuous cycle for those who play along.
Those who resist the evolution of the web wouldn't receive the
benefits of that information, and won't really leverage the
experimentations of the community.

Instead of using these headers to stabilize the UI, I'm thinking these
GM headers have the potential to create a virtuous feedback loop to
change the web faster.
What if Apple found out that 40% of their iPod users use some
workaround to some of the basic features/limitations? What do you
think they'd do for iPod vNext?
Maybe websites would find ways to create a more modular infrastructure
for supporting features needed by only 10% of their users.

Why not at least attempt it? Greasemonkey is about experimentation. If
that experiment fails, then let's just remove the feature. The best
way to proove this feature is useless it to implement it and "fail
fast". We could even set a criteria for what constitutes a failure

That's one thing I didn't like about the "fork threats": they put an
unusual pressure on a relatively small feature (that doesn't
revolutionize the architecture or principles of GM) than can easily be
reverted if need be. This kind of reaction is like a self-appointed
veto power, which stiffens the brainstorming and the debate. If you're
against that feature why not just reply with (-1) vote?!
We're only at version 0.3 of Greasemonkey! And Greasemonkey is not
Linux (see the politics going on over there about BitKeeper and lots
of other stuff). GM is still being defined and experimented with...

Anyways, this thread is becoming too serious :-P
I suspect the vetoers are not going to change their minds and Aaron is
going to give up like he said. Greasemonkey will still be great.
Sorry for such long email. I'll try to give my soapbox a rest for a
couple days ;-)


More information about the Greasemonkey mailing list