The Astropy project supports and fosters the development of open-source and openly-developed Python packages that provide commonly-needed functionality to the astronomical community. A key element of the Astropy project is the core package Astropy, which serves as the foundation for more specialized projects and packages. In this article, we provide an overview of the organization of the Astropy project and summarize key features in the core package as of the recent major release, version 2.0. We then describe the project infrastructure designed to facilitate and support development for a broader ecosystem of inter-operable packages. We conclude with a future outlook of planned new features and directions for the broader Astropy project.

We present the discovery and spectroscopic confirmation with the ESO NTT and Gemini South telescopes of eight new 6.0 < z < 6.5 quasars with z$_{AB}$ < 21.0. These quasars were photometrically selected without any star-galaxy morphological criteria from 1533 deg$^{2}$ using SED model fitting to photometric data from the Dark Energy Survey (g, r, i, z, Y), the VISTA Hemisphere Survey (J, H, K) and the Wide-Field Infrared Survey Explorer (W1, W2). The photometric data was fitted with a grid of quasar model SEDs with redshift dependent Lyman-{\alpha} forest absorption and a range of intrinsic reddening as well as a series of low mass cool star models. Candidates were ranked using on a SED-model based $\chi^{2}$-statistic, which is extendable to other future imaging surveys (e.g. LSST, Euclid). Our spectral confirmation success rate is 100% without the need for follow-up photometric observations as used in other studies of this type. Combined with automatic removal of the main types of non-astrophysical contaminants the method allows large data sets to be processed without human intervention and without being over run by spurious false candidates. We also present a robust parametric redshift estimating technique that gives comparable accuracy to MgII and CO based redshift estimators. We find two z $\sim$ 6.2 quasars with HII near zone sizes < 3 proper Mpc which could indicate that these quasars may be young with ages < 10$^6$ - 10$^7$ years or lie in over dense regions of the IGM. The z = 6.5 quasar VDESJ0224-4711 has J$_{AB}$ = 19.75 is the second most luminous quasar known with z > 6.5.

We use the SDSS/SEGUE spectroscopic sample of stars in the leading and trailing streams of the Sagittarius (Sgr) to demonstrate the existence of two sub-populations with distinct chemistry and kinematics. The metallicity distribution function (MDF) of the trailing stream is decomposed into two Gaussians describing a metal-rich sub-population with means and dispersions (-0.74, 0.18) dex and a metal-poor with (-1.33, 0.27) dex. The metal-rich sub-population has a velocity dispersion ~ 8 km/s, whilst the metal-poor is nearly twice as hot ~13 km/s. For the leading stream, the MDF is again well-described by a superposition of two Gaussians, though somewhat shifted as compared to the trailing stream. The metal-rich has mean and dispersion (-1.00, 0.34) dex, the metal-poor (-1.39, 0.22) dex. The velocity dispersions are inflated by projection effects to give 15 - 30 km/s for the metal-poor, and 6 - 20 km/s for the metal-rich, depending on longitude. We infer that, like many dwarf spheroidals, the Sgr progenitor possessed a more extended, metal-poor stellar component and less extended, metal-rich one. We study the implications of this result for the progenitor mass by simulating the disruption of the Sgr, represented as King light profiles in dark halos of masses between $10^{10}$ and $10^{11} M_\odot$, in a three-component Milky Way whose halo is a live Truncated Flat potential in the first phase of accretion and a triaxial Law \& Majewski model in the second phase. We show that that the dark halo of the Sgr must have been $\gtrsim 6 \times 10^{10} M_\odot$ to reproduce the run of velocity dispersion with longitude for the metal-rich and metal-poor sub-populations in the tails.

The shape of the Galactic dark halo can, in principle, be inferred through modelling of stellar tidal streams in the Milky Way halo. The brightest and the longest of these, the Sagittarius stream, reaches out to large Galactocentric distances and hence can deliver the tightest constraints on the Galaxy's potential. In this contribution, we revisit the idea that the Sagittarius Stream was formed from a rotating progenitor. We demonstrate that the angle between the disk's angular momentum and the progenitor's orbital angular momentum does not remain constant throughout the disruption. Instead, it undergoes a dramatic evolution caused, in part, by the changes in the progenitor's moment of inertia tensor. We show that, even in a spherical potential, the streams produced as a result of a disky dwarf disruption appear to be "precessing". Yet, this debris plane evolution is illusory as it is solely caused by the swaying and wobbling of the progenitor's disk. Stream plane precession is therefore not an unambiguous indicator of asphericity of the dark halo.

Motivated by recent observations of the Sagittarius stream, we devise a rapid algorithm to generate faithful representations of the centroids of stellar tidal streams formed in a disruption of a progenitor of an arbitrary mass in an arbitrary potential. Our method works by releasing swarms of test particles at the Lagrange points around the satellite and subsequently evolving them in a combined potential of the host and the progenitor. We stress that the action of the progenitor's gravity is crucial to making streams that look almost indistinguishable from the N-body realizations, as indeed ours do. The method is tested on mock stream data in three different Milky Way potentials with increasing complexity, and is shown to deliver unbiased inference on the Galactic mass distribution out to large radii. When applied to the observations of the Sagittarius stream, our model gives a natural explanation of the stream's apocentric distances and the differential orbital precession. We, therefore, provide a new independent measurement of the Galactic mass distribution beyond 50 kpc. The Sagittarius stream model favours a light Milky Way with the mass 4.1 +/- 0.4 x 10^11 M_sun at 100 kpc, which can be extrapolated to 5.6 +/- 1.2 x 10^11 M_sun at 200 kpc. Such a low mass for the Milky Way Galaxy is in good agreement with estimates from the kinematics of halo stars and from the satellite galaxies (once Leo I is removed from the sample). It entirely removes the "Too Big To Fail Problem".