[Hunch]

Picked up a copy of Jeremy Keith's Bulletproof Ajax because I'd heard it wasn't bad and I thought I'd see what his take on it was.

Found a stonking error on page 26.

Text reads as follows:

It's important that something happens inside the curly braces to change the condition so that it eventually evaluates to false. Otherwise the loop will carry on forever. Here, the alert statement will be looped 5 times:

var i = 0;
while ( i < 5 ) {
alert (x);
x++;
}

Talk about ironically (and erroneously) illustrating your very point.

[MikeMackay]

Ha ha ha, well found. There's usually an errata page for the book where people can submit any errors found in it, but I can't find one for this book!

- Mike

[sub]

Excellent. I'll have to rub his face in it when I see him.
He couldn't recommend the book highly enough when we first met!

[Hunch]

Yeah I found his errata page just before I posted, but it was blank, so I sent him a report. I'm (sort of) a Sussex boy myself actually - went to 6th form in Brighton back in the day.

[iblastoff]

ehhh. i would say 100% of all technical text in books have errors/typos in them.

[pgo]

Written by a human being afterall. Great book.

[Mr Fred]

so whats going on then?

[pgo]

Typos.

[combat sheep]

how cool are you to find the mistake, maybe you should write a book called even more bullet proof ajax

[herkalees]

"bullet proofer ajax"

[Hunch]

Originally Posted by Mr Fred

so whats going on then?

It's a bit of code called a while loop. Basically it does the same operation over and over until a particular condition is satisfied, and it stops looping. In this case when a variable called 'i' (a stored value) reaches 5, the loop will exit and the program will continue. However in this case, the value of i never increases and always stays at its initial value of zero (he keeps adding to another variable called 'x' instead). Basically this code will keep causing the browser to produce endless popup alert boxes forever. Although I haven't tried it, it would probably annoy someone who tried to run it because alert boxes tend to control the mouse focus on OSs like Windows, so it might be difficult to kill.

The irony is that in the paragraph leading up to it, he warns about how mistakes in loops can make them endlessly repeat and then goes and makes that very same mistake.

[Bill Posters]

it would probably annoy someone who tried to run it

…which Keith, presumably didn't.

I understand deadlines and all that malarkey, but I always find it odd that tech authors don't run every scrap of code they print/recommend before committing it to the book.

[Hunch]

I suppose he probably thought something that simple wouldn't need checking.

[haku]

Most likely.

My boss tasked me with learning to make yahoo widgets this week. So I downloaded the developers kit and was going through the reference manual, and found a few errors. Maybe Yahoo should take that offer from microsoft - it looks like they cant afford to hire proofreaders!

[Hunch]

Occasional proofreading errors I can cope with. I actually did a certain amount of freelance proofreading when I first came to Japan and it is very difficult to spot every single last mistake.

But when an author specifically issues advice about a problem, even going so far as to describe it as "important" and then fucks up and makes the very same mistake less than an inch further down the page, it just makes him look stupid.