I can see why upgrading to 1.6.* might cause problems for some people due to the lack of gif file type support in the library, but is there a reason why gd isn't upgraded to version 1.5 (which is surely available from somewhere, even though the author no longer gives it to people)?
The legal issues surrounding the Unisys GIF patent make upgrading tto gd 1.anything complicated. We'll leave this one alone, thank you.
That seems opposite then, since 1.6 no longer supports GIF files in order to totally eradicate the problem until the issue is resolved, yet the version of gd that is in RawHide DOES support GIF files, and therefor WOULD have "complications"... 1.3 (what RawHide has) through 1.5 didn't even use LZW compression, and used run length encoding instead, in an effort to get around the problem anyway (which would therefor make moving from 1.3 to 1.5 or anything else a moot issue since all versions have the same problems, the same number of complications, just newer features). If you really wanted to avoid this issue entirely, you would upgrade to 1.6. This is my last argument, simply to clarify issues that I think were misunderstood, I will not carry this issue further if you still don't wish to upgrade.
the new gd 1.5 has a significantly improved interface that will break badly the backwards compatibility we are shooting for between our next release and Red Hat Linux 6.0. Also this type of changes from its changelog are making me king of nervous: "" New GD2 format An improvement over the GD format, the GD2 format uses the zlib compression library to compress the image in chunks. This results in file sizes comparable to GIFs, with the ability to access parts of large images without having to read the entire image into memory. "" I understand fighting the compression algorithms by prividing none, not by implementing "comparable" ones that can be much easily challenged.