Main Area and Open Discussion > Living Room

Versioning of files

<< < (9/12) > >>

Ralf Maximus:
What is the problem with how I do things?  Honestly... I can see the allure of a pretty interface, but what functionality am I lacking?  I really want to know.

What will I gain by using a proprietary versioning system as opposed to collections of .zip files?  What functionality do you perceive I am lacking?

It takes me literally seconds to archive my project, and only additional seconds to roll it back.  Or forward.  To any point by version number or date.  All or part.   

And I can throw the whole thing on a DVD and carry it to any computer on the planet and access my source.

Help me understand why I am not as smart as you guys.  Seriously, lay it out.

f0dder:
Ralf Maximus: subversion (and other versioning systems) obviously won't shine until you're dealing with multiple developers per project, but after "getting into the mindset", even as as single developer I wouldn't do without subversion.

You might have some of the same functionality with your scheme, but with subversion everything is pretty easy. I can (quickly) check out any revision I want. I can "freeze" stable releases, or branch off into development releases. There are graphical tools for showing revision tree. (When using a repository server on the internet) I can check out from any machine I want to. File storage is done smartly to minimize disk usage (only differences are stored) - this is used when checking out stuff, too. I can use either commandline tools, or explorer integration.

Most of the stuff is probably doable with your system as well, you'd have to give subversion a chance and actually use it for a while before you learn to appreciate it... took me a while, even though the learning curve isn't steep.

Ralf Maximus:
Thanks, f0dder.  The two things you've listed that I can't do with my system are:

1. Graphical view of forked projects.  That would be nice to have, but thankfully, I don't do a lot of forking.

2. "Differential" storage of files.  However this conflicts with my requirement of self-containment; I must be able to carry a complete revision package from PC to PC if necessary, and requiring the whole archive to access the latest changes won't fly.

Now I want to hear from tinjaw.  Tell me why I am an idiot for not doing version control the way you do it.

MikeMcLoughlin:
I'm currently testing FileHamster at the moment and agree with previous posters that, for a single developer, it is a good solution.

Just thought I'd pass on details though of another product which I use called Developer's Backup http://www.developersbackup.com. It is basically a zipping utility but the way it allows you to set up backup plans and exclude file patterns makes it very convenient to use.

Give it a try, it's free with a condition that you register every 90 days, or shareware at a reasonable price. I paid for my version and use it all the time.

Carol Haynes:
Just tried FileHamster - I love it!

I have been using Adobe VersionCue but I like FH better. It is simple to use, easy to find things, ZIPs stuff up (though presumably you could equally well apply NTFS compression to the backup folder and get the same effect transparently).

WRT proprietary software - it isn't an issue all the previous versions are simply available in a simple directory structure and each version is clearly dated/timed in the revision filename.

WRT Zip plugin - just right click on the library at the top level and there is a link to the Zip Wizard and also details are listed in options.

How have I not seen this before?

I have one wish though ... rather than making an initial snapshot of all watched files/folders it would be really good if FH made a baseline backup of the file the first time it is opened - that way it would avoid duplicating hundreds of files needlessly. I know you can select what to watch and how to create the initial snapshot - but an automated 'as you go' method would be excellent.

Navigation

[0] Message Index

[#] Next page

[*] Previous page