Mathias Bauer wrote:
Peter Hillier-Brook wrote:
[snip]
Well and good so far, but the response to the issue was surprising
in its MS-like quality - it's not a bug, it's a feature!
I wouldn't tell that MS-like quality. Did you ever get a response
from MS at all? :-)
Mathias, I'm sorry I forgot to put a "smiley" on that remark. In a
previous life many years ago I developed operating system and data
communication software, and a standard joke was, "it's not a bug, it's
a feature" when confronted with a problem. In reply to your question,
yes I did get a reply when I reported a bug in a 'C' compiler in the
early 1980s, but never since:-)
OK, seriously now.
From a *technical* POV the current behavior *is* a feature, though it
is not adequate for the use case that didn't exist when the feature
was designed. At that point in time it appeared logical that versions
belong to the source document, not to the copy and so they are
intentionally removed.
I think the confusion is with the term, "copy". I appreciate that the
converter is implemented via the "Save as" function, and that is where
the problem lies, but when converting a document to a new file format I
would expect to convert the complete document and not to view it as a
copy. Using "Save as" seems to have been a mis-judgement of the true
requirement: either that or a short cut that went wrong.
If you think about document converters you are right if you call this
design wrong - I agree. But it's still a feature change, not a bug
fix from a technical POV. This classification into
"feature/enhancement" and "defect" does not tell anything about the
severity of the isse, it's only a technical classification. You could
also call it a "design bug", but that category doesn't exist in IZ,
and IMHO "enhancement/feature" is a better match to this non existing
category than "defect".
Accepting the foregoing, is my issue now an RFE, or do I have to start
again and make a specific request?
But I don't think that this classification is the problem, if the
"feature change" was accepted for OOo2.0 you wouldn't care, would
you? OTOH even if you call it a "defect" this doesn't give it a
higher probability to become fixed for OOo2.0 because it doesn't make
the fix easier and less risky.
Let me be a little more clear about where I'm coming from. This is NOT a
personal problem for me - I can obviously open and re-save each version
to a separate document, "touch" the files with the appropriate date and
get round the problem that way. What I'm trying to do is address this
issue from a corporate perspective for the ultimate benefit of Star and
Open Office.
My last employment was with the nice people in Neu Perlach and my
colleagues and I made extensive use of versioning so as to maintain an
audit trail when working with customers who, shall we say, occasionally
suffered memory loss that could have been to our detriment. By keeping,
in one file, every version of a contractual document we were able to
manage the situation.
Upgrading to a new software version with a new file format - guess who -
could have been difficult if the versions had been discarded during the
upgrade process. Obviously it was an internal discipline for us to
create a new transmittable copy of the document without the versioning
data and this did not prove to be a problem.
The problem is that you can't just revert the behavior and always
copy versions when a document copy is created by "SaveAs".
Agreed, and this is not the issue, as described earlier.
[snip]
It is definitely too late to create a new design for this feature in
the OOo2.0 release. If a workaround is possible instead that allows
to customize the behavior via API is currently under investigation.
That doesn't make a feature redesign obsolete but can avoid at least
a data loss in the document converter use case. Wether we can do
something for OOo2.0 depends on the feasibility of the mentioned
workaround.
I would urge the development of a work around, for the corporate market
requirements mentioned earlier, if at all possible, but in any case I
would hope that you accept my reasoning?
Best wishes.
Peter HB
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
