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

Re: Where we are going with 1.3



On 3/25/07, Charley Tiggs <..hidden..> wrote:
Chris Travers wrote:
> Hi all;
>
> I thought it might be a good time to update everyone with the progress
> in svn /trunk, which will eventually be released 1.3.  There are a lot
> of exciting developments already in the codebase and no doubt many
> more will be added before feature freeze.
>
<snip>

This is some cool and exciting stuff!  I look forward to contributing in
some way, either as a tester or submitting code that's within my
capabilities to create.



> Also there are a few areas I wanted to get some addtional feedback on:
>
> 1)  There is some debate (so far, over IRC) over whether to move
> templates into the database,
> PRO:
>   a)  makes for easier security because we can have full ACL-based support

A couple of pros to add here:

1) backup.  Backup your database back up your templates.
2) ability to use RESTful or other such interfaces to create your own
means of editing templates.  This enables developers to create
simplified interfaces that allow their customers to edit/create
templates as needed.

I am not entirely sure why database storage is necessary for RESTful
access.  I.e. RESTful defines how the client and web server
communicate, and that has nothing to do with how the web server
handles the request or where the data is stored.

3) portability.  Want to move from one host to another, dump your db and
import on new host.  No need to worry about moving template files.

Ok, I see that there are places where this might be an advantage.  But
portability here does not equal cross-platform (i.e. the big issues
with EOLs etc still remain).

4) one less resource on the file system where less technically savvy
folks installing Ledger will need to know how to set file permissions.

> CON:
>   a)  makes it harder to edit using a basic text editor.

I'm not sure this is a con.  With all the interface changes you're
making to take advantage of more modern tech, perhaps consider including
one of the browser based WYSIWYG editors for this?  I'm betting that
most users will only need to change basic stuff to meet their individual
needs so the need to use a basic text editor outside of a web browser
may not be all that important, depending on the target market.  Just a
thought.  For the more advanced users, it shouldn't be too hard to copy
over to an open window and copy back.  No matter what you do, there's
going to be some copying involved if there's a remote server involved.

How do you propose to recommend editing LaTeX templates?

>   b)  adds one more area to the code that will require more extensive testing

If storing the templates in the db is the best thing to do, then this
doesn't matter, excepting that it requires investing more time.  If the
goal is a better product, not how quickly you can turn around product,
more time spent testing is time well invested.

As is more time spent thinking through the design :-)

Thanks to the core team for all of your hard work and leadership!  It's
becoming quite exciting around here!

Thanks :-)

Best Wishes,
Chris Travers