Thomas Krichel wrote: 

>  I think we could introduce a distinction of two types
>  of resources. A "dated" resource would aim to remain as
>  it was at a particular time. A "timely" resource would
>  be subject to change.


>  Such a distinction would help to clarify the difference
>  between e-journal and preprint collection. An ejournal
>  (remember "jour" in French is day)  would be a collection of "dated"
>  material. If for example a typo would be found in a papers, this should
>  not be corrected ex post. If the problem is severe, a correction should
>  appear in a subsequent issue. A preprint collection, on the
>  other hand would be about "timely" resources only.

Would it not be possible to import some ideas and techniques from the
world of technical manuals?   These are documents that by their nature
must be corrected if found to be wrong. A typical technique is
for a manual to be issued in a loose leaf binder.  Amendments are then
issued by means of replacement pages.  The replacement pages will have
revision marks showing which chapters, sections and paragraphs have
been revised.  It is then possible for the owner of the manual to keep
all replaced pages and have a complete history of the document.

However, that is past technology.  We can do better on the Web.

Another way of describing what Thomas refers to as "dated" is "frozen".
I would see no fundamental difficulty in revising even frozen material
PROVIDED that each revision is clearly marked with the date and details
available to readers.  

Technically this could be done with clickable revision marks with links
to details of the revisions.

Each "paper" (whether in an e-journal or elsewhere)
might then consist of two sets of pages: 
1) the Current Text,
2) a Revision Record holding full details of all changes 
   (including deleted and modified text).  

An approach such as that could handle material of various degrees of
fluidity, not just frozen material.

Obviously there would have to be rules about the types of revisions
acceptable for different types of material. (Only typo corrections
permitted in frozen material, no changes at all in deep frozen?)

And finally, can we really expect to continue with the concept of
"issues" of an e-journal?  The traditional form of a journal is
heavily condition by the nature and restrictions of the physical  
processes of production and distribution.  

We are now using technology that allows material to be published
as soon as it is ready, why not just do so?

It would still be possible for editors to give us their
words of wisdom at regular intervals and to provide lists of
"what's new" since whenever, plus plans for the future.


Peter

=============================================================
Peter W Duncanson		E-mail:	[log in to unmask]
Principal Analyst		Fax:	(+44) 1232 230592
Computing Services		Tel: 	(+44) 1232 335375 
Queen's University,  Belfast BT7 1NN
=============================================================



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%