So we started using ClearCase at work, but I have to say I’m not really impressed so far… The GUI and the command line tool are both driving me nuts. Doing seemingly trivial tasks requires digging through pages of the manual, downloading some 3rd party plugin, or doing some tedious thing by hand. And oh yeah, did I mention it’s slow as paint drying, and just as fun to watch? I’ve written this entire blog post while checking in a chunk of code and files less than 50MB.
A simple example of the little things I keep running into is adding a folder of subfolders and files out on a remote file share to the version control. Should be easy, right? Just look for a recursive option in the ClearCase Explorer? Nope, there isn’t one. Well then, to the command line cleartool. Wait… it also doesn’t have a recursive option? Seriously?
After giving up on the official documentation and going to Google I found there’s another command line tool, clearfsimport, that’s supposed to let me convert a file system object and all it’s children into ClearCase elements. I excitedly go to the command line,
C:\>clearfsimport -preview -rec -nset \\sipsrvr\C\siplib\ Z:\racecode
clearfsimport: Error: Invalid pathname “\\sipsrvr\C\siplib”
UNC pathnames not supported.
Meh, fine, I’ll go map drives and do it that way. But really, for software that costs $4,000 for a floating license, you’d think that it would be a little less rough around the edges. It isn’t helping my experience that if I try to check a file in using my main machine, it BSODs Windows. We believe it’s some sort of issue with the version of virus software I’m using, IT’s going to look into it more tomorrow.
Looking around the Internet, I’ve found that I’m not alone in my opinion of ClearCase,
I haven’t used it more than a day, so I won’t give up on it yet, but I don’t think I’m going to be convinced anytime soon that it’s worth the high license cost IBM charges. Maybe I just overestimate Enterprise software. Hey look my files finally checked in. ::goes back to work::