LibTiff Mailing List
TIFF and LibTiff Mailing List Archive
August 2006
Previous Thread
Next Thread
Previous by Thread
Next by Thread
Previous by Date
Next by Date
Contact
The TIFF Mailing List Homepage
This list is run by Frank Warmerdam
Archive maintained by AWare Systems
Thread
2006.08.25 13:30 "Re: IPTC tag", by Joris Van Damme
Phil, Phil Harvey wrote: > I have found at least one utility (Nikon Capture) that won't read IPTC > information unless the format LONG, and as you mentioned this is the > way it is written by Adobe. So I don't think any IPTC support would > be very useful unless it reads and writes as LONG too. I just don't buy that sort of reasoning anymore... In my other post, I think I've clearly shown that the scheme Chris Cox mentions is the reading strategy implemented in Photoshop, misses out on some files very likely otherwise suspected to contain valid IPTC data (all files written by LibTiff, with byte order different from machine byte order and IPTC data written as longs, plus all files with IPTC data written as longs by any library at all, and subsequently byte order reversed by tiff handling software (like simple page concat!) an oneven number of times, plus...). I think it's likely nobody ever even gave much thought to this, perhaps people didn't even notice that frequently. In any case, likely no artificial breathing apparatus whose operation was a matter of life and death stopped functioning because of this. We certainly didn't hear of it. So, we're still supposed to try and write exactly same as old bugs, whilst we've just concluded we likely weren't and didn't even notice, the decade that we were propagating this mess? I say we should device a reading scheme that reads all that is out there, which thus outperformes likely most currently implemented reading scheme, which surely outperformes the reading strategy Chris mentioned is currently implemented in Photoshop, which is thus actually more (!) reading compatible with current bug then what you guys are proposing. I say let's pseudo-standardize such a recommended reading strategy and publish it for the common good. But I also suggest we start writing valid data. In conclusion, though, I'm left with the question if my recomendation sucks or does not, and should be abandoned or published on the IPTC tag page. You guys seem to suggest it sucks, but I'm not quite sure you do, and the recommendations you seem to come up with I've shown to contain certain problems that possibly you weren't aware of, and/or that turn out differently when applied to LibTiff instead of Adobe TIFF library, and/or that didn't take byte reversal operations by TIFF page concat software into account, and that just simply don't match with my thinking that we ought to always try and increase the validity in the 'files out there' in the long run. So I'm left with my question... what recommendation to publish? Joris Van Damme info@awaresystems.be http://www.awaresystems.be/ Download your free TIFF tag viewer for windows here: http://www.awaresystems.be/imaging/tiff/astifftagviewer.html