World Scenery with Submodules

[SHM]

True. Everything was fine until you gave a proper workaround to the gaps as suggested by Thorsten. Then the post disappears. Hopeless!

SHM

[IAHM-COL]

try making it dissapear from here.
You can't kill an idea. Ideas are immortal.

A samurai gives up when he realizes that books are not burnt, posts are not deleted, that ideas last forever.

Translation from japanese: "I will let this thread run its end"

[legoboyvdlp]

ONLY 600MB...




Hardly only, don't you think?
So that's for objects, airports, and that, riht?
Or 600 MB worth of 'three lines of text'

[legoboyvdlp]

You are welcome to post links to other information sources, but posting banned member messages verbatim violates the spirit of a ban. No worries, but I am locking the thread and moving it out of the main forum area. You can see that SHM posted a verbatim reply and we are already in a flame war ... <sigh>.

[IAHM-COL]

ONLY 600MB...




Hardly only, don't you think?
So that's for objects, airports, and that, riht?
Or 600 MB worth of 'three lines of text'

That's total
6 lines of texts for 600 subrepos cant barely make a few kbs in the repo size. The 600 is mainly out of Airport config files and shared models; which are global.

[IAHM-COL]

You are welcome to post links to other information sources, but posting banned member messages verbatim violates the spirit of a ban. No worries, but I am locking the thread and moving it out of the main forum area. You can see that SHM posted a verbatim reply and we are already in a flame war ... <sigh>.

What could SHM do? Thorsten is wrong. Life continues. He also... being a shame!
But we grew used to his unskilled set of thought. Shameful for a PhD in physics, until you know is hopeless, and shames not more, but sadness, until sorrow spents out and you just wish he thinks then he exists, instead of the "just exist".

Off course the emperor was to pad his back and clean the tears.

[SHM]

Wow I finally get mentioned by a mod haha.

Looks like Hooray had written a post and it is also gone, since he is the only one who I think makes sense there.

And we made a record today for the most people online

SHM

[SHM]

In case curt decides to remove this

The 2nd copied post was actually a reply to Thorsten when he mentioned about having gaps in the scenery. Agreed that the last line may've caused some flame war, but the right thing to do as a mod of the forum would be to remove that sentence not delete the whole topic.
For anyone looking what was mentioned kindly refer to the jabberwocky.net

SHM

[SHM]

To be fair, while you and I may disagree with the decentralized/distributed approach for the time being, Torsten himself suggested to look into that (see the devel list for details, e.g. his postings related to "BOINC" and distributed scenery creation).

Unrelated to fgmembers and their attacks, it is obvious that FlightGear as a project faced a number of incidents where we had a single point of failure, no matter if that means infrastructure downtime (forum, wiki), or actual data loss.
Equally, the transition from gitorious to sourceforge also was rather tedious, if not even extremely painful.

The current mapsever situation is rather unfortunate, too - and others, including senior contributors, have shared this sentiment in private, and some senior core developers have said so in public, to sum things up: it is not a good idea for an open source project like FlightGear to let a single person, piece of equipment/infrastructure handle critical details (or data ...).

So I think it is only fair to point out that, had a different (distributed/decentralized) approach been used (like the one suggested by fgmembers, or a completely different one), the FlightGear project would not be facing the mapserver dilemma at all.

It is far too easy to look at someone's posting history and disregard even valid arguments on those grounds - but everybody can check within seconds that Torsten and others have basically made the same suggestion - with the main difference being that they were looking at distributed computing not necessarily a process where distribution/development would be distributed and hat everything would still be in the hands of the project, i.e. not used in a distributed fashion, but only set up accordingly.

To be perfectly honest, it is a little contradictory to encourage contributors to participate using their real name, while people who are contributing using their real names have no way to get out of the "purgatory", and others can easily get a fresh forum/wiki account whenever needed.

I'd suggest we give them the benefit of the doubt, it's not exactly like terragear/terrasync and mapserver development is exploding due to the sheer number of developers ...

the mapserver/scenery incident is basically the worst case scenario, i.e. a long-term contributor "pulling" his contributions/work from an otherwise GPL'ed project, fully aware of the implications of his actions.

It would definitely help us improve our "bus factor", and if you don't like what fgmembers (or Thorsten) has to say, just look at what the people having to deal with all the mess are saying:

Is the Scenery Workflow broken?

Even I am unsure what the future scenery will be built from. Right now, Torsten is heading up the scenemodels part of the scenery. I am unsure if he has plans on taking on the shapefiles and terrain generation side of scenery as well.

I created an account on AWS, but I am unsure if I will pursue this much further. Truth of it is, we relied heavily on Martin for the scenery data, and now we'll need to find an alternative

Subject: Is the Scenery Workflow broken?

the only thing missing is the huge dataset that Martin has somewhere on his private hardrive.

Hi all,

when our main scenery resource will be gone in a couple of weeks, we will
have a hard time to create a new world scenery. Generating the entire world
was a job that ran continuously for days - if not weeks, even on a powerful
machin with almost unlimited resources.

As we have to re-think the entire process anyway, I just had the idea to
use our users resources for that, probably by using something like
BOINC: https://boinc.berkeley.edu/trac/wiki/AppIntro
We sure need to do many tweaks within the tool-chain, but I can't think of
any reason why this should not work.

Does anybody have experience with BOINC or alike?

Torsten

[SHM]

To be fair, while you and I may disagree with the decentralized/distributed approach for the time being, Torsten himself suggested to look into that (see the devel list for details, e.g. his postings related to "BOINC" and distributed scenery creation).
(...)
It is far too easy to look at someone's posting history and disregard even valid arguments on those grounds

FYI, Torsten has made it rather clear what is acceptable and what not in that context (which is strangely absent from your quotes). I'm surprised you haven't understood Israel's particular vision of 'decentralized' yet, but I assure you it does not meet the criteria.

So unlike you suggest, there is no valid argument here.

I'd suggest we give them the benefit of the doubt,

I repeat, that has been done for the better part of a year. I don't think anyone wants yet more of the same. I for once do not. Good luck in finding any developer who does.