Στις 10-07-2006, ημέρα Δευ, και ώρα 20:54 +0200, ο/η Erwin Rol έγραψε: > On Mon, 2006-07-10 at 20:26 +0200, Nicolas Mailhot wrote: > > Le lundi 10 juillet 2006 à 13:44 -0400, Benjy Grogan a écrit : > > > On 7/10/06, Nicolas Mailhot <nicolas.mailhot@xxxxxxxxxxx> wrote: > > > > > > The sad thing is people don't care about what you're asking - if they > > > > did word processing would not have rolled over DTP in the last decade. > > > > In a word processing context all you have is a text flow (as in > > > > free-flow). > > > > > > People do care. I've heard this complaint before. It's not a concern > > > when a 27 page document becomes a 28 page document, but when you want > > > to have 1 page and suddenly that 1 page has become 2 pages and you're > > > searching for what OS it last fit properly on, then it gets to be a > > > nuisance. > > > > Well, they don't care enough to buy DTP products, which means word is > > the only game in town, and writer is just emulating it. > > > > > I'd be happy with a set of fonts that work the same on all OSes, and > > > then I'd stick with those. Times New Roman seems like a hassle now. > > > > Even with a single common font you'd have problems : > > - some other parts of the formatting will have fuzzy definition > > interpreted slightly differently over time by different software > > versions. > > - every system won't interpret the same font the same way (currently > > Fedora won't use the bytecode interpreter because of patent concerns, > > Windows will) > > - fonts are not static : they include instructions for the rendering > > engines, and as rendering engines get smarter font designers include > > more complex instructions, which mean display and print approach > > progressively the font designer ideal, but the size of a given text > > string will change as a result. > > those are all technical reasons, here we (with we I mean software > developers, and that includes me) are trying again to find a excuse to > not make what the user wants. I would dare to bet with you that most > users expect their ODF (or DOC) documents to look the same even if they > open them 5 years later with a new version of OOo or Word (of course for > some documents they might say "oh well strange that it is now 26 instead > of 25 pages, but of well, probably just a little bug in my new version > of Word"). > > > The page change the poster is complaining about is due to Fedora > > honoring some font settings it ignored before. > > What ever the reason is, the software's way of working changed in a bad > way, so this "fix" did not fix anything it broke things. > > > Right now the only game in town if you don't use a DTP-like product with > > transparent fit-to-frame scaling is to reserve enough blank space on the > > page to account to the slight rendering variations between office > > suites. > > Not an option for multi-page documents. BTW differently rendered > multi-page documents can be very confusing too, for example think about > a meeting where 3 ppl open a document, one on a mac, one on windows one > on XP, and than say; "Now all look at page 25 where you will see ...." > good luck finding out where the text is that is shown on page 25 by one > of the ppl. > > For me, not rendering documents correct in a word processing application > is a _fatal_ bug. Changing the way old documents are rendered is even > more fatal. Copying this behavior from Microsoft (if Word even has that > behavior which i would not dare to bet about!) is not a good thing to > do. > > But since i have no idea how to fix it, I will not continue to bug ppl > with it. And sounding by the arguments on why it is displaying documents > incorrect it seems OOo is just broken by design, so there is nothing > easy to fix in the first place. It's difficult to say without seeing some sample documents. It's common that some documents are written in such a way that makes them fragile to slight variations of system settings. For example, there are documents that are written without specifying styles; to make the pagination they use blank lines instead of page breaks. Having some minimal sample docs would be great to pinpoint issues. Simos -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
