From: Randy Kramer (rhkramer@fast.net)
Date: Fri Dec 20 2002 - 09:28:32 EST
I just want to be on record as saying I disagree with what's written in 
the FAQ, specifically this:
<quote>
 Why not set the Reply-To: to automatically go to the mailing list?
If the Reply-To address is automatically set to go to the mailing list 
there is a risk that people will accidentally send mail to the many 
many people subscribed to the list, the alternative may mildly 
inconvient to some but it is the right thing to do and is the best 
anwser for the most people. Some of the more advanced email clients 
have features that allow you to indicate that certain email address are 
mailing lists and your email program will automatically change the 
address so that Reply-To will go to the list. 
</quote>
Clearly, I'll abide by it (what choice to I have ;-) ), and won't make 
an extensive argument if it is truly the consensus of the leadership of 
this project, but I do think it's totally wrong:
   * A mailing list is a place for "public" participation / 
collaboration -- by default your replies should be to the "public".  
(You chose to join the list -- no one forced you.)
   * Three recent posts were sent to someone privately affecting a 
subject I brought up on the public forum.  It wasn't until I forwarded 
one of those private replies back to the list that the right person 
(Hub -- who knew how to solve the problem) realized the problem and 
addressed it (voting was not enabled for "AbiWord" future).
   * More generally, when someone doesn't want something to go to the 
list, I may become suspicious of their motives.  Are those replies 
confidential??  On an open source project??   Are they criticisms that 
are best done in private??   
Aside: I think that off-list criticism sometimes is appropriate to a 
certain extent, but if, for example some undesired behavior is 
mentioned on the list but only criticised off list, a perception can 
arise that the undesired behavior is acceptable.  Maybe that means in 
some cases that if a behavior is criticised off list, a post should be 
made to the list something of the nature "I responded to Mr. _______ 
off list to explain that his suggestion is not the kind of thing we 
want to encourage on this project".
Aside: I guess there is room for confidentiality on an open source 
project -- I think it should be the rare exception rather than the 
rule, and people that want it kept confidential should be responsible 
to keep it confidential -- I know there is one open source project that 
I won't remember (don't even remember what the product category is) 
where the product is dual licensed and sold under one name and open 
sourced under another name with the intention of keeping many of the 
purchasers unaware that the product is available for free as open 
source.  (It's not as underhanded as I've made it sound -- the concern 
is that a certain class of customer does not trust open source, so they 
are happier kept unaware.  (I learned this from a post on the os-bus 
mail list (IIRC), and no one raised an outcry that this was in any 
sense a violation of open source policy or ethics.))
Anyway, one of my Wish List items is to create (or find, or modify) an 
email client where I control what reply to does -- one that I can set   
to reply to the list regardless of the list settings.  (I suppose there 
can be some problems -- for lists that do that by default, I'll have to 
set it differently so duplicates are not created.)
Randy Kramer
PS: I'm certain Alan is right, this subject has been discussed before.  
Although I am not suggesting that we have another debate on the subject 
now, it certainly can be considered again some time in the future.
On Thursday 19 December 2002 08:42 pm, Mikey wrote:
> On Thu, 2002-12-19 at 20:17, Alan Horkan wrote:
> > Definately Not.
> >
> > And i am fairly sure i explained this in the FAQ.  This has been
> > discussed several times before.
>
> My bad, haven't checked the FAQs in month or two.
>
> Thanks!
> Mikey
This archive was generated by hypermail 2.1.4 : Fri Dec 20 2002 - 09:26:07 EST