[Tutor] Re: fnord

[Rob]

> Use your web browser and browse the archives.  When you find an
> interesting post that you want to discuss with the author, highlight
> and copy the address from the page onto your system's clipboard.  Now
> fire up your mail client and paste it in the "to" field.  Oops, that
> didn't work, you now must de-munge the text manually.  (hence
> "annoying")
>
> Once I made just such a copy-n-paste error no less that 3 times in a
> row (on the same address, after my message bounced).  Not only that,
> but the address didn't come off a web page and had been munged by hand
> in the first place.
>

I don't blame you for finding the experience annoying, but in most cases I
would consider this an acceptable compromise to make in order to attain the
desired end result.

> | What if not all addresses were munged the same way? A munging
> application
> | could have several different mung styles built in, and randomly
> choose which
> | one to apply for each email address to be munged. That would be a little
> | trickier to code around, I'd wager.
>
> Nah, just use a regex to identify which of the finite (and small, more
> than likely) number of munging styles were used.  Or just run the text
> through each de-munger and see which one(s) yield an address
> afterwards.  Then spam all the addresses you got back and remove the
> bad ones from your list afterwards.  If you can programmatically
> create it (and the output is comprehensible) then someone else can
> programmatically comprehend it.
>

Certainly. Hence my phrasing "a little trickier to code around". I suppose
we could also make the archives only available via an authentication
process.

Rob