artificial satellites appears before 1957

[pablo petrowitsch]

If I search the sky aspect in dates before 1957, it also appears artificial satelliites

[Bogdan Marinov (daggerstab)]

So? The Satellites plug-in's database is a snapshot of (some of the) currently Earth-orbiting satellites and it needs to be constantly updated to show their positions with any accuracy. It's not intended to be a complete and/or historical database of every satellite ever launched.

[Alexander Wolf (alexwolf)]

I think we need hide all artifical satellites for time before 1957

[Matthew Gates (matthew-porpoisehead)]

Maybe we could have a new data entry LaunchDate in the JSON file for
each satellite, before which the object is always turned off. This
value could default to 1957 if not specific.

However, I feel like this is somewhat insufficient. Better would be
to have some sort of period of accuracy (a window a few weeks either
side of the last TLE update), and hide satellites outside this period
(TLE predictions get inaccurate really quickly!). Such a time window
should be enabled by default, with an option to ignore it...

On 26 October 2011 16:54, Alexander Wolf <email address hidden> wrote:
> I think we need hide all artifical satellites for time before 1957
>
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/831702
>
> Title:
>  artificial satellites appears before 1957
>
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/stellarium/+bug/831702/+subscriptions
>

[Alexander Wolf (alexwolf)]

A fix has been committed as revision 5086 of the trunk branch in Stellarium's Bazaar repository at Launchpad:
http://bazaar.launchpad.net/~stellarium/stellarium/trunk/revision/5086

[Bogdan Marinov (daggerstab)]

Matthew, the launch date is a part of the International Designator which is encoded in the TLE set. So no additional field in the JSON file is necessary if you go that way.

[Bogdan Marinov (daggerstab)]

I don't think that this is a good fix and the author of the plug-in seemed to agree when I told him about it. It removes all satellites if the date falls before 1957, but it doesn't put them back if the date returns to the present.

So I've removed the code. As this "problem" is related to the problem of satellite orbits quickly becoming outdated, I suggest leaving the solution to the author of the Satellites plug-in, Matthew Gates.

[Alexander Wolf (alexwolf)]

OK. I'm propose also hide artificial satellites before date it's launch.

[Matthew Gates (matthew-porpoisehead)]

Hiding before date of launch is good, although I still feel that we should hide all satellites when the simulation time is more than some period from the last TLE update... e.g. 2 months (because TLEs are not valid for very long)

[Bogdan Marinov (daggerstab)]

I had an idea yesterday - add an option to automatically remove those satellites that are not in the source lists when updating. This may help a bit.

Another thing: you can add the orbit epoch date in the upper-left-corner info field, and display a big red warning in the same place when the data starts to become stale. I don't know how often the epochs are changed.

[gzotti (georg-zotti)]

I just commited another quick fix (r5553): just don't draw anything before Sputnik1 launch date. However, I agree that the tracks should only be drawn from reasonably fresh elements, or at least satellites should be hidden on a one-by-one basis before their launch date.

[gzotti (georg-zotti)]

OK, following the previous hint from Bogdan, I allowed myself to commit (r5554) a simple solution: Hide satellites prior to their launch years. They are just not drawn, the data list is not affected by this. It's nice to scroll through the decades and see the population of satellites grow. Matthew, I hope you agree with this for now.