<br><br><div><span class="gmail_quote">On 03 Nov 2006 21:24:14 +0200, <b class="gmail_sendername">Jari Aalto</b> <<a href="mailto:jari.aalto@cante.net">jari.aalto@cante.net</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Nicholas Allen <<a href="mailto:allen@ableton.com">allen@ableton.com</a>> writes:<br><br>> The way CVS does it is really bad and it often<br>> makes mistakes by assuming that all files are text files unless the
<br>> user specifies that they are binary (and often users forget this). So<br>> CVSs policy is one of data destruction by default and I do not think<br>> this would be a good idea for bzr!<br><br>I understood that a VCS is primary for text files and only secondary
<br>used for binary files<br><br>CVS way of assuming text by default is for the typical situation and<br>explicitly marking other types of files as binary is logical.</blockquote><div><br></div></div>In more than 15 years of corporate software development i have not
<br>encountered one project without binary files in the repository.<br>Most of the time these files are Word documents (specifications) or Excel documents<br>(example data from customers, formulas, ...), not to forget all the GIF, PNGs and
<br>Photoshop files you need for web projects. And if you ever messed up a complete<br>history of files, because you forgot to set Photoshop to binary, you would wish for<br>binary being the default :-)<br><br>The point i totally agree with you is not to hard wire any 'clever' defaults. Have just
<br>one default - binary or source - and let the rest up to the project using bzr. I do remember<br>one project where they had '.asm' files. I thought they are ASCII files containing assembler<br>code, but they were binary files containing information about assemblies. Much room for nasty
<br>surprises.<br><br><br>Ciao,<br> Steffen<br><br><br>