keith said:
>   There a two kind of users.
>   1) the nerds here on the list with their pet technologies.
>   2) the average joe that wants to volunteer.

well, actually, keith, we shouldn't consider "the nerds here"
to be "users"...  we're just nerds with our pet technologies...

and further, most p.g. "users" do _not_ "want to volunteer".

they just read the books, and that's all they care about, and
i'm fine with those users; i'm glad they found those books...

a very rare and special few _do_ "want to volunteer", and
-- due to that p.g. banner -- most of them end up at d.p.,
where they are coddled by the unofficial "welcoming staff"
but abused badly by the workflow, and eventually go away.

the extremely small percentage who stay generally grow to
feel more allegiance to d.p. -- it's stockholm syndrome --
than to the now-more-distant-to-them project gutenberg.

but, by and large, the most common type of "volunteer" is
the person who happens to notice a mistake in a p.g. e-text
and decides to be a good citizen and make an error-report...

usually this is just a short-lived feeling that leaves quickly.
"i thought i should do it, and i thought i would, but i forgot."

if the person does persist, however, they will usually try to
learn how to make their error-report.  some never find out.

but there are others who _do_ find out!  the procedure is to
send an e-mail to a @pglaf address.  (it sometimes bounces;
in that case i would assume the person finally just gives up.)

if the report does get received, they get an automated reply,
saying "thanks for your report, we got it, and we'll act on it."

then, maybe years later, the error will be corrected.  or not.

as you can see, there's lots of little hang-ups along the way.

but, to me, the _worst_ is the vague nature of the instruction.

remember, this is a person who _wants_ to do the right thing.
plus they have to have been highly motivated to have gotten
this far along in the process.  so the least you can do for 'em
is to give _specific_ instructions on how they should proceed.

as it is, they have to provide their own answers for questions
that will invariably arise in their mind, such as "what should
the e-mail say?", "how should i format it", and similar issues.

i suspect that the _intention_ was to convey that "it's all good"
and "do it however you like", but that's simply wrong-headed.

what people want in this situation is simple clear instruction.

they want to fill out a form, so they know they did it "right",
and so they don't have to spend any time thinking about it...

remember, these people _want_ to do the right thing, _but_
their motivation is fleeting, so make things _easy_ for them.

e-mail is not "easy".  composing an e-mail is not easy either.
especially if you don't have the slightest idea what to include.

moreover, and just as important (if not more), everyone will
compose their e-mail in their own unique way, which means
that the person on the p.g. end has to parse that uniqueness.

37 people will find 37 different ways to say "this period here
should be a comma", meaning you'll have to _parse_ all 37,
mentally reworking all of them as "change period to comma".

that "mental reworking" becomes draining after a few hours.

the kicker is that it's all pretty much _unnecessary_...

this is how project gutenberg _should_ do error-reports...

there should be a web-page specifically for reporting errors.

it should have a form on it which solicits the necessary info...

>   book-number: ______  or title: ________________________
>   current incorrect line: ________________________________
>   proposed correct line: ________________________________
>   any other relevant notes: _____________________________
>   your e-mail address (for confirmation): _______________

every e-book should have a link to this page.

heck, every _page_ in every e-book could have a link to it.

all the info you need, without tiresome mental reworking.

the output from this form could even be used by an app
for convenient checking and then acceptance/rejection of
the proposed correction, with auto-updating of the files...

indeed, i wrote up such a system in a post to this listserve,
years back, and offered to write the code to implement it...

but -- of course -- the write-up and the offer were ignored.

now, years later, al laments that "error-reports are hard"...

it's only hard because the powers-that-be are super-stupid.

***

does anyone here know how to make a web-form for p.g.?

there must be _someone_.  if not, i will volunteer to do it...

***

p.g. fails to capture the efforts of people who want to help
by reporting an error in the e-book which they are reading.

further, the failure to reinforce such motivation means it is
usually extinguished, a second curse compounding the first.

-bowerbird

p.s.  i put such error-reporting in my "continuous proofing"
system way back as early as 2005, and probably even sooner.
to report the error on this page, use the form at the bottom:
>   http://z-m-l.com/go/mabie/mabiep003.html