[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: One-Stop Debian Box Config Tool: Call for Collaborators!



On Fri, 2013-09-13 at 01:36 -0700, Jarrod O'Flaherty wrote: 
> 
> 
> Greetings All!
> 
> This is a call to Debian programmers who would be interested in
> spending a couple hours 
> a month working with me on developing a "One-Stop Debian Box Config
> Tool" -- a tool intended to
> become the central and all-encompassing place to go to configure any
> and every aspect of your 
> Debian system. 
> 
> 
> 
> CONCEPT OUTLINE
> =================
> 
> The tool will (subject to the approval of the collaborators ;) :
> 
> 1. Drastically reduce the need to:
>    a) Google every time you want to tweak feature X of package Y.
> 
>    b) Post to message boards when Googling fails to deliver the goods.
> 
> 
> 
> 
> 2. Provide users with an (ever-growing!) common repository of
> step-by-step "recipes" by which 
>     they can tweak / fix / customize / build / repair / upgrade their
> systems.
> 
> 
> 
> 
> 3. Present each step of a recipe in the form of a regular shell
> command, so it can be
>     easily checked, easily modified, and -- most importantly! --
> easily applied.
> 
> 
> 
> 
> 
> 4. Eliminate the need to copy and paste said shell commands by
> providing a special terminal
>     window as part of the interface.
> 
> 
> 5. Reduce or altogether eliminate the need to edit the shell commands
> by intelligently substituting 
> 
>     installation-specific pathnames, module names, version numbers,
> etc. into the commands
>     as appropriate.
> 
> 
> 6. Allow you to search the recipes using a goal-based syntax similar
> to the following:
>    PATTERN) "I want to: VERB + OBJECT [ + to + VALUE ]" 
>    EXAMPLE) "I want to: change the default GTK font size to 18pt"
> 
> 
> 7. Facilitate the sending of feedback to report successes and failures
> using a given
> recipe, automatically collecting and attaching to it relevant
> information on the system setup
> as well as any (error) messages that were output during the process.
> 
> 
> 8. Play The Imperial March every time you report using a recipe
> successfully. 
> (Hmmmm. Then again, there could be some licensing problems there.)
> 
> 
> All frivolity aside, let's start talking about how to automate the
> system configuration and 
> administration process the same way the rest of the *NIX world is
> automated!
> 
> 
> 
> HOW TO GET INVOLVED
> 
> ====================
> 
> Those interested should email me ( jofsama@yahoo.com ) with their:
> 
> * Name
> * Languages Spoken/Written
> * Timezone of Residence
> 
> * Linux Background and Proficiency
> * Linux Flavors Used
> * Programming Experience
> 
> * Ways You Would Like to Help
> 
> Anyone and everyone who enjoys using Linux is welcome to join. 
> 
> And if you would like to participate but are unsure as to how to do
> so, let me suggest
> that collaborators can, initially at least, be of greatest assistance
> in:
> 
> * Setting up a project homepage.
> * Setting up a mailing list or equivalent by which collaborators can
> communicate.
> * Helping to flesh out the project scope and requirements.
> * Drafting up a design document and work plan.
> * Creating a document & code repository on Github or similar.
> 
> Come and join me in collaborating on a tool that's going to be the
> biggest revolution in 
> Linux-box interaction since .inputrc got "history-search-backward"!
> 
> Look forward to hearing from you!
> 
> Yours Sincerely,
> Jarrod O'Flaherty.
> 
> 

---------------------
I'm not a programmer but I do have a opinion. Webmin is already my 'go
to' tool for a lot of those chores. The problem is it is too generic and
not specific to Debian. A good place to start would be to get involved
as a module writer for Webmin that is "SPECIFIC" to to Debian or
generate an installable "current" Debianized fork with modules that are
relative and preset for Debian. If not that then devise a system of
similar capability. As far as I know the only real issue with Webmin is
its just not broad based enough, and with mostly one developer, Jaime
has done a remarkable job. My rarely expressed 2 cents worth.
I would be willing to offer critiques from a user point and ideas if you
get this rolling. I think its a good idea.


Reply to: