Re: F-Spot renaissance

[Etienne lepercq <e guepe gmail com>]

2011/2/23 Tim Howard <timothy howard gmail com>

On Mon, Feb 21, 2011 at 8:16 AM, Paul Lange <palango gmx de> wrote:

However, while looking at the code I got the feeling that we may be
wrong in trying to update the old F-Spot codebase and should think about
starting from scratch.

I'm drafting a more complete response to your email that I may or may not end up sending. :) I think the allure of starting from scratch is always strong. However, I certainly don't have the time for an endeavor. Although I'm approaching F-Spot development like any part of it can be from "scratch".

Ruben insists tripod does nothing interesting and is not worth the effort. I can't even get it to build so that may be an indication. It is certainly abandoned at this point.


tim

I am an f-spot user and also developper (but not an f-spot developper). But as a developper, there are _rare_ cases where starting from scratch is a good idea. Most of the time, developpers want to start from scratch because they think they will "get it right this time". And more importantly usually because they think it is easier for them instead of learning how the actual code works.

Most of the time, there are some parts in a software that is "hacky" : but such sections of code _can_ be rewritten, without sending to scrap the whole code. Need an example ? Have a look at the linux kernel !

However, if one think that parts of the code are obscure, hard to understand : just ask here, and let's document/think of a partial rewrite/explain !

This is more productive than starting from scratch. As an f-spot user, recent advances puts it on the right track (faster import, video support, ...). Let's leverage these advancements and recent work to "create" a new f-spot, the easy way : by hacking it !

my2cents

-- 
Sincerily