In the week that IBM'ers rejoiced over the announcement of official support for their using Firefox and the amazing Mark Pilgrim published his latest free book Dive into Greasemonkey, I was passed an industry analyst report, you know the sort that big companies pay big money for their insight, which, says something along the lines: "Greasemonkey has absolutely no place on corporate computers [snip] it opens you up to exploits, performance problems, and support chaos.". I found this really depressing. Firstly as any good cook knows you're more likely to cut yourself using blunt knives. Secondly, I wonder what really is the difference between a GM script and a bookmarklet such as flickreplacr. Guess mnot maybe onto something with his ideas for a Greasemonkey proxy if it puts corporate thought-police minds at rest, but in essence GM is about personal computing and that is always going to be in contention with those responsible for protecting the herd - "you do realise it's not your computer, it's the company's".
I pity websites who get greasemonkey scripts written for them. Yes, the scripts can potentially provide some useful information about what some people want to do with the site but there is a bigger, uglier danger - people using other people's scripts.
Now when a website wants to update itself it has to be very careful about how many of its users are using a script it didn't write and may not even know about.
I bet that the vast majority of people who use greasemonkey will be using other people's scripts and when the website changes the scripts will break, in any number of unpredictable ways.
The irony here is that the more popular greasemonkey gets the less innovation and change websites will be able to bring. It's ironic but greasemonkey more or less undoes the main benefit of the web for most data providers - the ability to quickly develop and change software without having to worry about client side deployment issues. With greasemonkey client side deployment issues come back with a vengence.
I have trouble imagining that the benefit of greasemonkey is worth the cost for the majority of normal, non-programming, users.
All that having been said, for geeks, greasemonkey is clearly pretty cool. But that's because when their greasemonkey script toasts a new version of the site the geek can understand the problem and take appropriate action.
And yes, I do realize the potential here for a versioning system but I strongly doubt it would work in practice. Sites get tweaked all the time and trying to express that in a useful version tag that doesn't just get changed every day is a non-trivial problem. Semantic markup could potentially help but the scope for misunderstanding is awesome.
It's not true about cooks cutting themselves on blunt knives. Unexpectedly sharp ones are the most likely culprits.
Andy: fair comment!
Yaron: i think your excellent comments deserve a separate blog entry of its own ..