r2 - 27 Apr 2007 - 15:53:46 - BobMannYou are here: TWiki >  VOTech Web  >  StageFive > DS6Plan05

DS6: Stage Five Plan

Introduction

The DS6 workplan for Stage05 focuses on filling in the gaps in the DS6StudyReport, continuing the development of specific software (VisIVO, VO-neural) and standards (SIA, SSA, Characterisation Data Model) and undertaking a testbed science scenario, which pulls together many of the aspects of DS6 work to date. This work is divided into a series of (purposefully unequal sized) tasks listed and described below.

Stage05 Task List

Number Title Who
T01 Testbed science scenario: Kernel Density Estimation AstroGrid
T02 Advance Characterisation Data Model to Proposed Recommendation CDS
T03 Characterisation Data Model: follow-up of implementation and uses CDS
T04 Release of DAL-extension based services for CFHTLS and CGPS CDS
T05 Standards development on SIA and SSA CDS
T06 VisIVO dissemination Catania
T07 New features in VisIVO: unified view, structure data on meshes, volume rendering Catania
T08 Porting VisIVO for Mac Catania
T09 VisIVO database interoperability Catania
T10 Implement SNAP protocol in VisIVO Catania
T11 New VisIVO website Catania
T12 Completion of release of VO-neural v1.0 Napoli
T13 Discussion of application messaging protocol All
T14 PLASTIC: further developments CDS

Stage 05 Task Descriptions

T01: Testbed science scenario

The goal of this task is to assess whether it is possible to undertake a realistic data exploration analysis using the VO as currently constituted: this follows a previous exercise on Doing science with prototype Virtual Observatory tools? conducted several years ago. The testbed scenario chosen is efficient selection of quasars from photometric data, following the example of Richards et al. (2004), who developed a Bayesian classifier on top of a Kernel Density Estimator running in colour space. Richards et al (2004) performed their analysis on the SDSS DR1 dataset, while we propose to analyse SDSS DR5 and/or the current overlap between that and the UKIDSS Large Area Survey. This work is intended to address two of the main goals of DS6, namely running data exploration operations as distributed services (using the VOTECHBroker) and applying them to very large datasets through the use of sophisticated data structures, like k-d trees.

T02: Advance Characterisation Data Model to Proposed Recommendation

DS6 staff at CDS are very active in the development of the Characterisation Data Model, and it is intended that this should be advanced to Proposed Recommendation level by the time of the Beijing interop meeting.

T03: Characterisation Data Model: follow-up of implementation and uses

This work parallels T02, and concerns the consideration of experience with the draft Characterisation Data Model. In particular, this will include assessment of the experience of ElizabethAuden in applying Characterisation terms in VOEvent work: this task is in collaboration with Ds4.

T04: Release of DAL-extension based services for CFHTLS and CGPS

The CFHTLS and CGPS datasets have been used as testbeds for extensions to DAL, and these services will be released during Stage05.

T05: Standards development on SIA and SSA

Standards development continuing during Stage05 will include the inclusion of a draft annex in the current draft of the SSA standard, and the initial work on SIA 2.0

T06: VisIVO dissemination

Dissemination activities planned by the VisIVO team for Stage05 include: participation in the IVOA Interop. Meeting in Beijing; participation in the 6th International Workshop Data Analysis in Astronomy (15 - 22 April 2007, Erice, Italy); running the Second VisIVO day in Catania (May or June); and further seminars demonstrating VisIVO to potential users in their institutions.

T07: New features in VisIVO

New features to be incorporated into VisIVo? during Stage05 include a unified view (so that points, volumes and vectors are displayed in the same window), a new way of handling arrays (to facilitate the manipulation of structured data on meshes), the offering of variable cell size in each dimension, and the optimization of volume rendering.

T08: Porting VisIVO for Mac

A port of VisIVO for the Mac has been requested by many users, and should be addressed during Stage05, although its completion will be dependent on support from the developers of the MAF.

T09: VisIVO database interoperability

Database interoperability is being added to VisIVO, primarily so that users can readily extract data from archives of numerical simulation datasets.

T10: Implement SNAP protocol in VisIVO

Making VisIVO understand the SNAP protocol will enable users to download a subset of a simulation, defined by graphical selection.

T11: New VisIVO website

A new VisIVO website will provide users with a ready means of downloading VisIVO installation packages, documentation and related papers, as well as sending messages to the VisIVO mailing list. It will also include a FAQ section and an image gallery, plus, in future, an online tutorial.

T12: Completion of release of VO-neural v1.0

The release of v1.0 of VO-neural (the VO-enabled reworking of Astroneural) will proceed through the completion of work on a series of modules (for Multi-Level Perceptron; Self-Organising Maps; Probabilistic Principal Surfaces; NEC; NeXT?), together with associated visualization tools

T13: Discussion of application messaging protocol

The IVOA Applications Working Group is currently debating the development of a standard application messaging protocol, and the experience of DS6 members in developing, and using, the PLASTIC protocol will provide an invaluable input into that process.

T14: PLASTIC: further developments

CDS continue to enhance Aladin with further PLASTIC-enabled features.

-- BobMann - 27 Apr 2007

Edit | Attach | Printable | Raw View | Backlinks: Web, All Webs | History: r2 < r1 | More topic actions
 
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback