Blunt Tools (was: cgf does not want private email about cygwin)

[Ken Collins]

> You always want to err on the side of caution when it comes to the 
> interpretation of the tone.  Doing so helps keep one from reading too 
> much (negative) into a message being read.

By the same token, it's advisable to err on the side of friendliness and
helpfulness when expressing something. I'd like to see more of that. It
makes for better reading.

> And who would be answering questions on a list full of newbies? ;-)  

Actually, the newbies would be. As they work through the stupid problems,
if they're encouraged, they may post back simple howtos from the
perspective of someone seeing the system for the first time.  If you
discourage people from joining in early, you lose that stuff. I know I
could put together a pretty good intro on setting up tcsh, in that regard.
As people solve individual problems, they gain expertise in specific areas
without knowing much about the system as a whole. 

> understand the resources available to make the most of Cygwin.  I'd be
> hard-pressed to find allot of negativity in this type of response but,
> like I said, everyone interprets things differently and you can't please
> everybody all the time.

It's not so much a matter of pleasing but discouraging. Do you really want
to risk discourage people interested in Cygwin?

> Personally, I think you're reading too much into the "strict"ness of the
> list.  Given the number of posts and the number of responses containing
> real information that address the posted queries, I'm not sure what 
> constitutes this "strict" nature you refer to.  

The ratio of RTFM replies versus full responses to questions constitutes
the "strict" nature. I find it high given the unique nature of the system.
It's like grafting a fourth leg on a three-legged dog. There are bound to
be a lot of questions, especially if people are used to seeing three-legged
dogs.

> What constitutes a "non-answer" for you?  Is a pointer to the place where
> the answer is a "non-answer"?  I've seen very few global pointers on this
> list.  Most point to specific email archive messages, FAQ entries, or
> user guide chapters.  If you're referring, the "search google" responses,
> then I am still confused by your comment.  How is it that pointing the 
> original poster to the method of finding the answer a bad thing?  Are you
> assuming that the responder has the answer at their finger-tips and is
> just punishing the original poster for not being in the same boat?  I 
> think you're assuming too much here.  People direct folks to the search 
> engine for two very good reasons:

Okay, imagine if you reply to every entry level message about a given topic
with a Google search, aside from the first response to the message.
Ultimately, when you finally do the Google search, half the messages will
point to the same search you just did. When you find the original message,
it will be old, and it possibly won't pertain to what you've asked, unless
you fully trust that the person who directed you there with one line and no
comment other than "look it up" has followed the link themself. 

If the reply pointed to a specific message with a specific answer, that
would be significantly more helpful. If the person replying is going to go
to that much trouble, why not just cut and paste into the reply, and
provide the text for everyone else who's curious? Instead of killing the
discussion, you leave it open for comment, in case the reply isn't exactly
relevant.

I would probably have less of a gripe with the Google search approach if
urlview didn't SEGV on me. It's what I primarily consider a non-answer.

Also, imagine if, instead of replying to your sincere question, I sent one
line with a long URL linking to a series of complaints I've had about
mailing lists in the last year? It doesn't promote dialogue, to say the
least.

>    1. They don't recall the specifics of the answer to the poster's 
>       question but know its in the archives.  Its not the responders
>       responsibility to look up the actual message for the original 
>       poster, although that is always an option.

Why not just ignore the post and wait until someone with the energy to look
up the actual message chimes in instead of preempting that response?

>    2. Getting to know how to use the archives and even of its existence
>       is a very good thing.  Its a powerful resource which gets real time
>       responses and provides allot of background.  Its important for 
>       everyone to know about it and to use it to make the community a
>       productive one.

True. I'd like to see the archive contain actual responses, instead of
links to links. I'd like to read discussions of problems instead of jumping
out of mutt and into a long list of Re:s in IE. 

> These are the reasons that people are pointed to searches.  It's not 
> some patronizing response meant to make the poster feel stupid.  It's the
> responders best effort to help the poster find that which has been 
> requested.

Perhaps. It hasn't entirely been coming off that way to me, and that's why
I'm mentioning it. If it's unintended, I'd assume that people would
appreciate the notice from someone who follows the project and who cares
about the quality of the dialogue. 

> Actually, I find preambles like "Sorry for this newbie question but I 
> looked at the FAQ and mail archives and didn't find the answer" pretty
> useful.  At least for me, I then know whether the person has made an 
> attempt to find any existing information on the topic of question.  Even
> if they've clearly missed an obvious pointer (like in the FAQ), I
> personally am more likely to prod them more directly to the information
> they'd find there.  In other words, if I know someone has tried to help
> themselves and failed for some reason, I can then be more specific with 
> the suggestions I have, because I know what they've tried and what didn't
> work for them.  But to each his own...

It's unfortunate when people have to apologize for asking questions on a
mailing list. I'll add the preamble to my sig :)

Ken Collins

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple