On 30 April 2012 02:54, Dennis E. Hamilton <dennis.hamil...@acm.org> wrote:
> It's tricky to find these because they are not shown or linked on the page > itself, even though they have been uploaded as attachments. > That's because I didn't have time and I don't want to spend time if it is not going to be useful. A lot of pressure in my real work at the moment. I notice that the vector-graphic versions don't render so well when > displayed with IE9. The idea is not necessarily to use these directly but to produce the raster versions from them at whatever size and resolution is necessary. If Drew has vector versions of the images the simplest solution is to upload those but since they are not in the sources I'm assuming he hasn't. > In the transition vectorized logos [1], - the diffused shadow and > transition background render as solid with no diffusion or transparency. > For Ian's version, the shadow under the ball is not diffuse but the > lighting on the ball is closer to the expected form (in IE9). > Chromium on Linux does not render these perfectly either. Puts hard lines around the gulls and light area. > > In all views of Ian's versions, the shadow under the ball is as if the > ball is suspended above the surface rather than resting on it. That was deliberate :-). I liked the look of it but easy to change. I'm certainly not a professional artist, I'm just trying to establish that we use vectors and it is not a big job to do it. The really important thing is to get definitive vectors from which all other images are derived. > Also, the shadow has a hard outline and is a single tone in some viewers, > but is diffused in Google Chrome. I think this is again the way different svg implementations handle things. In Inkscape the line width is defined as 0. It would be useful to test these as OOo 3.4 rc1 imports to Draw - I haven't got 3.4 installed so I can't test it. In LibO the import doesn't work at least not with any reasonable fidelity. > The reflection at the top of the orb is unreal, as if the surface is > different, rather than there simply being a diffuse light source above or > behind the orb. In this case, IE9 renders it better. Chrome renders it as > if it was a hatch cover on the orb! (an edge shows around the lighter area > on the top part of the orb. > Quite happy for anyone to edit it better, but I wouldn't worry too much about the svg rendering direct in a browser at this stage. It's more about having a consistent reference. Let's decide on that and then get the details right. Compare with rasterized versions of the orb, such as the Approved Logo > large version at section 5 on the page itself, < > https://cwiki.apache.org/confluence/display/OOOUSERS/Logo+Proposals>. I > think the lighting on that orb is unnatural too, though not so distracting. > It is also the lighting of the orb that goes back to OpenOffice.org 3.3.0, > unnatural or not. > > It may be that those effects are available in vector graphics only by > using 3D modeling of the orb and its lighting. Likewise for the defocusing > that works in the transitional rasterized logo, [1]. It is going to be a > challenge to find an SVG that renders consistently. It may be necessary to > pick a tool where it works as needed and scaled raster versions are made > where the export preserves the appearance. (Even a screen capture from a > rendering of the vector graphic could be used, if the vector-graphic > rendering is successful.) > I think we should choose either AOO Draw.odg or svg for the definitive images and then produce png and jpgs as required from them. If the reference is at least consistent we have controlled the variables for subsequent copies. There is always a degree of subjectivity about anything "artistic" :-). If we go down the odg route from AOO we are at least eating our own dog food. Main disadvantage is if that then limits it to odg on AOO to be sure of it rendering consistently. If we use svg on Inkscape it is supporting the W3C standard which even though differently implemented now on different browsers has probably the best chance of converging to something consistent as HTML5 gets a grip. Using odg might have the advantage of providing a focus for svg import export filters. At least they should work accurately with our logos :-). Perhaps they will in RC1 if so it would be a selling point over LibO ;-). That's all I can tell by a comparative inspection of the rendered images. > > - Dennis > > -----Original Message----- > From: Ian Lynch [mailto:ianrly...@gmail.com] > Sent: Sunday, April 29, 2012 15:57 > To: ooo-dev@incubator.apache.org > Subject: Re: Logo of AOO in SVG? > > On 29 April 2012 23:27, Claudio Filho <filh...@gmail.com> wrote: > > > Hi > > > > 2012/4/29 Ian Lynch <ianrly...@gmail.com>: > > > Just to say I have uploaded more logo candidates including with the > text > > > for 3.4 Release Candidate 1 and some versions incorporating the Apache > > > feather. Any comments improvements, redesigns welcome. > > > > > > Ian, have some wrong thing in your logo. The shadow and reflex are > > strange. > > > How strange? If you want to improve them just download, improve and > reupload. I'm off to bed now. End of week end. > > Please, make your idea with my logo[1] or with logo made by > > Michael Acevedo[2], that is perfect! > > [1] > > http://cwiki.apache.org/confluence/download/attachments/27834483/aoo.svg > > [2] > > > http://cwiki.apache.org/confluence/download/attachments/27834483/AOO+Logo+SVG.svg > > > @Michael > > Very nice work! If i had found this file before, could be saved a lot > > of work. ;-) > > > > Best, > > Claudio > > -- > Ian > > Ofqual Accredited IT Qualifications (The Schools ITQ) > > www.theINGOTs.org +44 (0)1827 305940 > > The Learning Machine Limited, Reg Office, 36 Ashby Road, Tamworth, > Staffordshire, B79 8AQ. Reg No: 05560797, Registered in England and > Wales. > > -- Ian Ofqual Accredited IT Qualifications (The Schools ITQ) www.theINGOTs.org +44 (0)1827 305940 The Learning Machine Limited, Reg Office, 36 Ashby Road, Tamworth, Staffordshire, B79 8AQ. Reg No: 05560797, Registered in England and Wales.
