Why is Japan NOT on terrasync?

Support, HOWTOs, information, development, and everything related to obtain scenery via Github with terraGIT
User avatar
IAHM-COL
Posts: 6455
Joined: Sat Sep 12, 2015 3:43 pm
Location: Homey, NV (KXTA) - U.S.A
Contact:

Why is Japan NOT on terrasync?

Postby IAHM-COL » Mon May 09, 2016 7:23 pm

KL-666 wrote:That brings me to a question. Would it not be possible to also push these airports to terrasync? Because i do like the download on the go aspect of terrasync, so that i do not need to download a whole country at once.

Kind regards, Vincent


If you are talking about "technical" possibility is way too easy to do. If you are talking about executive possibility the answer is no. Not possible. Terrasync repository is closed and someone lost the key.
The only thing possible to add up to terrasync is new models, but even this case scenery is much in the grey area.

Models need to have a very specific, and frequently inconvenient format, and the smart usage of tiling and texture sharing used, per example, in chubu is not acceptable. I did a good amount of work making Chubu compliant with terrasync forms and submitted Chubu to terrasync. Due to being me who submitted it, Chubu was vetoed. The argument of "why not" was fabricated, and left a clear point that Chubu was not going to be uploaded to terrasync any time soon.

Other situations, include the smart use of OSM buildings and Roads, constructed in places like Narita (RJAA) and Chitose(RJCC), and even Kansai Cargo area (RJBB) where they produced, with the proper methodology a radical change of appearance and rapidly improved scenery without the need of making too many custom 3d buildings. In addition to be more accurate depiction of reality, than merely filling blanks on shared models (like your dreary tower). But still, what's interesting here is that OSM buildings are not acceptable submissions for terrasync.

Continuing with the Project 3000 shared models, whose many locations enriched Japan, and these are now available to be installed worldwide in terraGIT, but these will just output Model not found errors if pushed to Terrasync. Same situation with the shared models by J. Maverick, which are useable in terraGIT but rather of no use to be pushed to terrasync.

Furthermore, Jetways and Groundnet specification files are not acceptable to terrasync.

Finally, terrain layouts: This is also not acceptable in general form, since to update, terrain needs to be patched with partially built texture, which breaks the Core developers policy that world needs to be generated in a single run. [url]Our method of tile patching does produce visible tile boundaries that look sometimes like a thin breakage in the terrain. In the core developers view this is a totally unacceptable situation[/url], and thus KHAF and TNCM will remain underwater, SGAS will remain under Grass, and OMDB will remain under sand until a new scenery WS3.0 comes around. Many of these fixes had been pushed to terraGIT, as we deal with a world that takes small patches with ground corrections, on the expense of having sometimes some stitching evidence.
For this last reason, the layouts in Japan were updated in many --many airports, but they remained being either 810 specification, underwater in the island, or single default runway layouts, for terrasync users.

As it stands, the current situation is that the exquisite control over scenery development that is needed to finely control scenery development is lacking on the terrasync business model, and thus, making necessary the creation of an FGMEMBERS affiliate scenery repository for both: the development and distribution of new scenarios.

On the last point here: FG is moving terrasync from its historical protocol (subversion) towards a more classical transfer protocol (the good old http). Reality is that we will eventually be able to offer an http mirror for terraGIT. This would allow users to modify the configuration files in FGDATA to fetch terraGIT in a purely automated (on the fly) way: Just redirecting the http server from the official servers to the TerraGIT mirrors, thus, providing a final solution to the problem that concerns you right now.


IH-COL
https://raw.githubusercontent.com/IAHM-COL/gpg-pubkey/master/pubkey.asc

R.M.S.
If we gave everybody in the World free software today, but we failed to teach them about the four freedoms, five years from now, would they still have it?

User avatar
IAHM-COL
Posts: 6455
Joined: Sat Sep 12, 2015 3:43 pm
Location: Homey, NV (KXTA) - U.S.A
Contact:

Re: Why is Japan NOT on terrasync?

Postby IAHM-COL » Mon May 09, 2016 7:31 pm

IAHM-COL wrote: Chubu was vetoed.


https://sourceforge.net/p/flightgear/ma ... /34921577/
Catalanoic and Thorsten on Devel.List wrote:> Isn't better to move the scenery devel folder to Github? As for another
> custom scenery related works are done.
>

That'd be a complete change of how we handle scenery.
Currently almost all of our scenery is maintained in a database and the
resulting files are automatically generated.
We try hard to avoid incompatible chunks of scenery that don't match at
their corners.

Martin Spott has alway tried to encourage people to maintain data upstream
and I think this is the right way to go.

Torsten


https://sourceforge.net/p/flightgear/ma ... /34934069/
Torsten on DeveL.List wrote:Hi,

we are currently holding back a couple of contributions for static objects
in our scenery database.
Some people think that these objects have issues that should stop them from
being included to our default scenery.
The issues are mostly related to texture size

RJGG_Terminal - two 2048x1024 textures approx. 50% unused space each
RJGG_Hangar1 up to RGJJ_Hangar11- two 1024x1024 textures (looks oversized
to me for mostly simple hangars)
RJGG_Name - this is just the word "CHUBU CENTRAIR" and is made of two
1024x1024 textures
RJGG_Terminal_jetwaysN, two 2048x1024 textures with mostly repetitive
pattern
RJGG_Terminal_jetwaysW, two 2048x1024 textures with mostly repetitive
pattern
RJGG_Terminal_jetwaysS, two 2048x1024 textures with mostly repetitive
pattern

And there is also
RJGG_tower_red-light.xml - just a simple red light, could be reused from
Shared Models.

The texture sizes are uncommen for our scenery models, we have many
examples to achieve a high level of detail with smaller images.
A private mail request to the contributor suggesting to resubmit with
smaller textures was rejected

Question is how to deal with these contributions. In some private e-mails I
already collected responses from "so what, include 'em" to "way too large,
reject them".
I don't think we have explicit rules how big textures may be and I think
it's good not to have them and keep the liberty of having detailed textures
where required.
>From looking at the textures with my limited but existing experience with
textures I dare to say that with a bit of work and texture magic it is
possible to significantly reduce the size of the textures without loosing
any level of detail.

I see three options how to proceed:
a) Accept the models as is and don't care about possible frame rate impact
b) Reject the models, requesting to resubmit with smaller textures
c) Accept the models and resize the textures by 50% each side (2048x1024
becomes 1024x512)


My suggestion would be c) but I'd like to get some response from this list
before I proceed.

Thanks,
Torsten


https://sourceforge.net/p/flightgear/ma ... /34936995/
Torsten on Devel.List wrote:Thanks for all your responses,

after a some privat emails we decided to reject the submissions. The
reasons were:
- The Models use high resolution textures not adequate for our database
- The Models were copies or derivates of work found here:
http://www.grafikavirtual.com/fgfs/?sec=escenarios.php
- The author refused to cooperate
Torsten


Noteworthy the last decision of Reject, and final period, was not even a) b) or c), initially considered (they did not accept my offer of resize texture with a cost on quality, but instead Torsten Suggested I was unwilling to cooperate). And certainly not any of the answers and comments listed by people responding on the devel list.
https://raw.githubusercontent.com/IAHM-COL/gpg-pubkey/master/pubkey.asc

R.M.S.
If we gave everybody in the World free software today, but we failed to teach them about the four freedoms, five years from now, would they still have it?


Return to “terraGIT”

Who is online

Users browsing this forum: No registered users and 4 guests