[sldev] Source reorganization needed

[Michael Miller]

This thread has gone very far off topic. I think, to summarize it,
there are the following issues:
1. Developers(outside of LL) feel neglected because their patches are
not being accepted or incorporated.
2. Developers(outside of LL) feel that the source is outdated and
needs to be updated to use the latest libraries.
3. Developers(outside of LL) feel that the source is very poorly
documented and hard to navigate through.

I think that by fixing (2) and (3), (1) will fix itself. If the source
is up to date, using the latest libraries, and well documented, it
will be much easier to write patches for, and verify patches for
Second Life.

I do realize that (2) and (3) are significant financial commitments.
It's not cheap to largely rewrite a program of Second Life's size.
However, my argument is that this will bring new developers to the
platform and will result in much larger financial gains than if LL
decided not to rewrite / document the source.

As an example to illustrate this, I'd look at Windows. Windows started
out in the 80s(1985 to be exact), and has had to support legacy
software so that older versions of Windows could run today's apps(at
least in theory). This has resulted in bloated code which has had to
support libraries which are decades old. To put this into perspective,
Microsoft only recently (with Vista) dropped support for 16-bit
software(something no one uses anymore).

The comparisons aren't exactly equal, but they show what happens to
code when it relies on old, stale libraries. The code becomes bloated,
and very hard to work with.

It would be much better to do a gradual transition to standardized
libraries now than to wait 5 or 10 years until LL is forced to migrate
because the code is falling apart.

On 8/29/07, Dzonatas <dzonatas at dzonux.net> wrote:
>
>  Dale Glass wrote:
>
>  That 'waste of time' is pretty clear. LL had the notion to relate me as
> a 'waste of time' under a similar condition. I think we find proof here
> with Torvald's Git work that I'm not the 'waste of time' for my attempts
> to send code upstream, however. (I didn't start the fire on that bridge.)
>
>  What waste of time? I grepped the archives and couldn't come up with
> anything that would seem like the explanation.
>
>
>  To make it brief, there is the notion found here and elsewhere (like on
> office hour logs) about pains and wastes of time. I was related to one in a
> particular job at a time I had to deal with some court deadlines. Some
> employees at LL seemed to need to blame someone over wastes of time.
>
>  Obviously, I'm not posting all message sitting in my personal inbox, but
> I've noted the mode of discussion for what discussion existed.
>
>  Probably what gets confused is that some think I'm attacking LL, but I'm
> really focused on the need to solve a silent war on families here in
> America. I'm against those who walk all over other families and steal
> happiness.
>
>  There is a deep niche made when someone says 'waste of time' towards
> someone that has lost a lot of time with their family. It may be easy to
> assume that the reason for the loss is due to one's own created-problem, but
> that is the wrong assumption that has only made kept the war on families
> silent. Once you look at the child support system, you'll see how cruel the
> war on families can be.
>
>  Considering LL's new mission statement, I doubt my efforts are really a
> waste of time. I feel it is the Open Source community and Linden Labs that
> need to be more aware about these issues. There is a mission statement that
> sooner or later will have to deal with it directly. I'm just one to deal
> with for now.
>
>
> --
>  Power to Change the Void
> _______________________________________________
> Click here to unsubscribe or manage your list subscription:
> /index.html
>
>