On 07/20/2016 02:19 AM, Scott Kostyshak wrote:
On Tue, Jul 19, 2016 at 12:50:19PM +0200, Michael Berger wrote:
Hi all, this is to share my experience with other users and asking for help
from the developers. ;-)
I have quite a number of LyX documents made in Mageia5 using LyX 2.1;
TexLive 2013; classicthesis-LyX-v4.1, most of them with the module
'Linguistics' loaded and using different bibliography styles.
After upgrading to LyX 2.2.0; TexLive2016; classicthesis-LyX-v4.2 and
loading those older documents I found the following when trying to export to
PDF:
I don't know much about this stuff, but in general you do not want to
upgrade several critical components at once. Ideally you should upgrade
one system (e.g. LyX), test, if everything goes well, upgrade another
system (e.g. Koma-Script), test, etc. And whenever you upgrade, you
should always make sure you can easily downgrade. If you can't easily go
back to how you had it before, take a step back and ask yourself "do I
really need to upgrade"? You will learn through frustration that
upgrades break things. This will always be true. But since you are
learning more about Linux you will learn how to create a robust workflow
and how to easily go back to how you had things before.
Sounds like a frustrating issue. I hope you find a solution and don't
have to redo all of the documents!
Best of luck,
Scott
Thanks Scott,
good comment, very basic but nevertheless quite true: there is no such
thing like a general recipe or one single reason if things don't work as
expected after such radical upgrades.
To be frank, I am surprised that by now all but two my of old documents
(Miede's classicthesis) do compile after changing/adapting just one or
two details - most documents compiled off the cuff!
And for these two last documents I found my way: a combination of
copy&paste and/or redoing other parts.
And yes, I could always go back (there were situations when I thought to
just forget about this whole dam.. idea) without loosing a bit because
this exercise is done on a separate experimental computer while all my
work is still there and save on another machine.
Forgetting the frustration and thinking of the benefits and learning
effects I have no regrets :-D.
Cheers!
Michael
