Newbie question - multiple head prevention

[Tony Zakula]

On Tue, Jun 8, 2010 at 3:05 PM, David Dyer-Bennet <dd-b at dd-b.net> wrote:
> (Top-posting fixed.)
>
> On Tue, June 8, 2010 14:03, Tony Zakula wrote:
>>
>> On Tue, Jun 8, 2010 at 1:58 PM, David Dyer-Bennet <dd-b at dd-b.net> wrote:
>>>
>>> On Tue, June 8, 2010 12:31, Michael Diamond wrote:
>>>> Tony,
>>>>
>>>> I would be inclined to make these two separate repositories.  You're
>>>> looking
>>>> at two almost completely unrelated workflows, and I would be inclined
>>>> to
>>>> either:
>>>> 1. Just .hgignore the image directory - this may not be acceptable for
>>>> your
>>>> use case, but I find just keeping backups of the site makes more sense
>>>> for
>>>> binary files like images than checking them into a repository.
>
>>> Depends on what you're doing, but in my experience working on web sites,
>>> matching the image versions to the code versions is absolutely
>>> necessary.
>>> The details and how they're broken up and sizes will change in the
>>> course
>>> of development, and getting the wrong images for the code will look
>>> HORRIBLE, clearly broken.
>
>>> (A scheme where you changed the file name, maybe by appending a version,
>>> and kept that set of file names in a fixed location accessible to all,
>>> and
>>> didn't put them in source control, could be made to work; the references
>>> from the code would then always point to the right names, whichever
>>> version you updated to.)
>
>> I am kind of slow, so excuse me.  Are you saying that you would do it
>> the way I have it setup?  Or just have the paths in the code point to
>> the production server? (Which I personally do not like, but)
>>
>> Another route I considered was storing the images in the database, but
>> that has its own set of issues, and I prefer not to go there.
>>
>> I could make the designers use source control, but that could be a
>> bigger headache.
>
> Possibly because I was reading too fast, I'm not actually sure how you
> have it set up now, so I can't give a simple yes/no answer to that part.
>
> I've found that, since the images ARE an integral part of many projects
> (particularly web sites), it's important to have them under source
> control, and for the graphics people to have access to the repository.  We
> actually had to find a Visual Source-Safe client for Macintosh computers
> at one job to achieve this, but it was worth it.
>
> There can certainly be other ways to achieve it, but making sure that the
> images and the code for the the web site are in sync for each release, and
> if you have to fall back to a previous release, is vital.

Hey,

Thank you to both of you for taking the time to lay all of this out.
I have been struggling with the whole workflow thing since I started
to use HG. (it is a good thing) Many of these projects were not in
source control at all.  I have worked some with Subversion on other
projects with developers so decided to make it mandatory.  I never
realized what a mess things were until I began trying to gather
everything into one project/build!  I have spent quite a few hours
putting just this one project altogether, but I know it will save
hours in the future.

It sounds like I should get the designers to use hg, leave the images
in source, even if it is just the image directory as an hg project.  I
hesitate to give them access to code classes and what not.

I know these answers take time, and I really appreciate the advice.

Tony Zakula