In a rare case of actually doing something I said I would, I’ve started to write a utility for elevating certain words and phrases in the web browser to your attention, by highlighting them and providing a tool-tip explanation for why they were highlighted. It’s still in the occasionally-blow-up-your-webpage-and-crash-the-browser phase of development, but is showing promise nonetheless.
I have my own reasons for wanting this utility (“LOOK AT THIS WORD! SUBJECT IT TO SCRUTINY OR IT WILL BE YOUR UNDOING!”) but thought I would throw it out to LW to see if there are any use-cases I might not have considered.
On a related note, is there a reason why Less Wrong, and seemingly no other website, would suffer a catastrophic memory leak when I try and append a JSON-P script element to the DOM? It doesn’t report any security policy conflicts; it just dies.
Ooh, ooh, thought of another cluster of danger phrases (inspired by a recent Yvain blog post, I forget which): “studies have shown”, “studies show”, “studies find”, and any other vague claim that something’s corroborated by multiple scientific studies which the writer mysteriously can’t be bothered to reference properly, or even give a clear description of.
Not without breaking everything horribly (including my debugging tools) in a non-negligible number of cases (including Less Wrong).
I did put together a little bookmarklet example, but since it doubles as an “occasionally mangle your page and possibly make this tab explode in a shower of RAM” button, I decided not to share it until I’ve isolated and fixed this particular problem.
In a rare case of actually doing something I said I would, I’ve started to write a utility for elevating certain words and phrases in the web browser to your attention, by highlighting them and providing a tool-tip explanation for why they were highlighted. It’s still in the occasionally-blow-up-your-webpage-and-crash-the-browser phase of development, but is showing promise nonetheless.
I have my own reasons for wanting this utility (“LOOK AT THIS WORD! SUBJECT IT TO SCRUTINY OR IT WILL BE YOUR UNDOING!”) but thought I would throw it out to LW to see if there are any use-cases I might not have considered.
On a related note, is there a reason why Less Wrong, and seemingly no other website, would suffer a catastrophic memory leak when I try and append a JSON-P script element to the DOM? It doesn’t report any security policy conflicts; it just dies.
Ooh, ooh, thought of another cluster of danger phrases (inspired by a recent Yvain blog post, I forget which): “studies have shown”, “studies show”, “studies find”, and any other vague claim that something’s corroborated by multiple scientific studies which the writer mysteriously can’t be bothered to reference properly, or even give a clear description of.
So is this script somewhere we can try it?
Not without breaking everything horribly (including my debugging tools) in a non-negligible number of cases (including Less Wrong).
I did put together a little bookmarklet example, but since it doubles as an “occasionally mangle your page and possibly make this tab explode in a shower of RAM” button, I decided not to share it until I’ve isolated and fixed this particular problem.