Hi all, and especially Martin,
first of all let me say that I completely respect your decision. I'm sure,
knowing you're way of decision making, that you've
been thinking it over for a while.
Without trying to change your opinion, I do like to give you my view on the
scenery issue. You know that I support a centralized
scenery database, but through the extensive contact with "the community" (if I
may say so) I did gain some respect for their
reasoning. More on that below.
Now that automated scenery downloading (via TerraSync) is integrated into the
sim there are more scenery-database users
than ever. So it's importance is definitely not decreasing, even the contrary!
Talking to many (new) users and (scenery) developers over the past years I've
found that there are reasons (you can argue
whether they are valid reasons, but they are reasons) for releasing scenery
independant from the scenery database:
Landclassing: submitting landclassing data to the scenery database is rather
easy. I did some parts myself
(West Frisian islands, Dubai, Manhattan). Two years after I submitted Dubai,
it's still not included in the
scenery. If I'm not mistaken, the last scenery build dates back to 2008... I
don't blame Martin for this (absolutely
not), it's just an observation.
Airport layouts: following up on the previous point, updates for airport
layouts can be submitted. Another thing
I've done for quite some airports. But, since the terrain isn't rebuilt after
such changes, airport layouts that were
updated years ago are still nothing more than a few lines in a .dat file.Object
placement: various airports are/were so incorrect in the scenery (there's an
airport in Belgium that's 300m
shifted IIRC) that it's impossible (and a waste of time) to place objects in
such a way that they don't interfer with the
(bad) layout.Annoying the scenery maintainers: when you add a lot of objects,
and are still updating them frequently, you
don't want to annoy the scenery maintainers with dozens of requests/updates.
Oliver is tackling this part with the
automation.
Licensing: we don't have much scenery that's non-gpl, but there are a few
areas. This is obviously a good reason
that we cannot do much about.
To give one recent example that applies to most of these reasons: LOWI. It has
an updated airport layout, custom landclassing
and is under heavy development. The author told me that he'll put all of it in
the database once it is finished. Altough I don't
know his exact reasoning, I imagine the above mentioned reasons are certainly
part of it.
In order to tackle these reasons, we need to have more frequent scenery builds.
If I'm not mistaken that is/was on Martin's
todo-list. After submitting an airport layout or shapefile update, scenery
should get rebuilt automatically. I am aware of the
challenges with that (eg. gaps between tiles). I'd like to learn how I can
help. Yves, I'll try to come on IRC the next week for
sure, excellent idea!
I'll stop writing now, already typed to many words for the current University
project this morning. Getting dizzy from all the
words that passed my screen the last 6 hours or so...
Last but not least: Martin, thanks for all you've done! I'm sure there will be
one day that the majority will be thankfull for
everything you did! In fact all TerraSync users already are (without knowing I
assume).
Cheers,
Gijs
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
