Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: discussion list

Dear Yves
 
> I am not sure that the choice of distribution lists software is the best.

Certainly I have not made an exhaustive survey of what is available.
Listproc was recommended by John Westbrook, who runs it at Rutgers for the
mmCIF discussion lists (and I think it's also used for the imgcif list at
Brookhaven). What's most important (from my point of view) is that it should
be relatively easy for the system administrator to manage, and for list
owners and subscribers to understand. So far, I'm fairly happy with this
software, but I think it does need some experience before one understands
all its features.

> I have now five years of practice with lists and I found that most scientists 
> (except in the computer field) are not familiar with the mechanism. Nowadays 
> the listproc should be hidden.
> 
> Most scientists find rather difficult to register by sending a message to a 
> first addresse (listproc) and then to discuss using a second addresse (epc-l 
> in the present case).

I understand this. However, there is an advantage in having a single
listproc (in Chester) managing multiple lists, in that the principle needs
only to be mastered once, and then there is a common interface to one or
several lists that an individual might want to subscribe to.

The UK community runs a centralised discussion list service (look at
http://www.mailbase.ac.uk/), and this has proved very popular across many
fields of interest. I have been tempted to suggest that we simply use this
service, but I think the lists it supports are supposed to have memberships
that are predominantly, or largely, British. However, we might borrow some
ideas from their documentation to provide easily navigable web pages for
potential users of the IUCr lists. (It's interesting that Mailbase requires
a staff of eight!)
 
> In many cases I have to register or delete the people myself. You will also 
> find a difficulty with protected lists such as this one. In many places the 
> From: field which is used to identify people and recover their e-mail address
> is not properly set (this happens especially with agents on micros such as 
> Eudora), thus when they send messages from different machines most messages 
> are rejected!

It's important that the list owners understand how to manage the list, and can
register and delete subscribers. 

The problems with configuration of mail headers are real, but there is a
need for some sort of validation with private lists. Listproc uses the
"From " (not "From: ") field, which I think is the property of the envelope
rather than the message body, and hence is less fragile. It also has the
ability to register multiple aliases for subscribers, but I agree that this
may still involve too much labour. (It's a pity we don't still have a
25-member EPC - that would surely make a very useful testbed for this
list :-)

> My present software is from the same generation as the one you are using and
> seems outdated. We did not changed it but we will do it when we will find the
> time. I know one called sympa developped in France which is in the public
> domain and which includes a Web interface for registration.

I shall try to evaluate the sympa package. Of course, there is a balance
between novelty and stability - "out of date" but functional can be
better than "state of the art" but bug-ridden. We did think about interactive
web conferencing tools such as HyperNews, but they didn't seem to be
sufficiently easy to install, simple to manage or efficient to use. But
perhaps they will rapidly improve. Certainly I shall keep an open mind.

> The present software seems to add an additional burden: what is the use of the
> password and do you believe that most people will answer as requested?

I agree with this; it may be possible to conceal the password functionality.
The password is designed for interactive contact with the listproc daemon,
but it would probably load the server nachine too heavily to permit
unrestricted interactive sessions.

> About archives definitively they must be accessible through the Web for public
> lists. What I did was to create the archive directories under our anonymous
> ftp so that a pointer in a Web page allows to access the archives. Using a
> software to transform them into HTML pages is certainly a better solution and
> Brian should definitively use it.

Do you like the web archives? I think the layout is attractive; but it
doesn't have a search function, which the listproc's own archiving system
does support. One could, of course, WAIS-index the html archives. It all
depends on how much we think the community would benefit from such features.

> Welcome messages should include the following:
>
> - Ethics
>   + Scientific object of the list
>   + Style of discussion: be polite1
> - Technics
>   + How to unsubscribe
>   + How to retrieve the help information

I see the IUCr presenting an integrated email/web service, so that the
welcome message to the list should contain a detailed explanation of the
scientific purpose of the list, and perhaps only some pointers to web pages
explaining the general principles of netiquette.

I have a couple of examples of good documentation, such as the mailbase
setup I mentioned above, and some pages at UC Davis (e.g.
http://ir.ucdavis.edu/faqs/listproq-faq.html). The UC Davis pages are a good
example of what I would like to put together, and it might be worthwhile
talking to them to see if they're willing to share their experiences. It's
not clear whether they're running the public-domain listproc, or the
commercial version now sold by CREN (at $2000).

> Attached is a typical message (in french)

Thanks - I'll try to build up a collection of ways of doing this.

Regards 
Brian

Reply to: [list | sender only]
International Union of Crystallography

Scientific Union Member of the International Science Council (admitted 1947). Member of CODATA, the ISC Committee on Data. Partner with UNESCO, the United Nations Educational, Scientific and Cultural Organization in the International Year of Crystallography 2014.

International Science Council Scientific Freedom Policy

The IUCr observes the basic policy of non-discrimination and affirms the right and freedom of scientists to associate in international scientific activity without regard to such factors as ethnic origin, religion, citizenship, language, political stance, gender, sex or age, in accordance with the Statutes of the International Council for Science.

Useful links