Version 0.2 Implementation of an XML Schema describing a Simple Image Access Service. Based on "The Simple Image Access Specification" by Doug Tody et al. Diff 0.01-0.02: The Topologi tool apparently doesn't like annotations of global elements referenced locally. Although, this does not appear to be in violation of the XML Schema standard, they have been commented out for the time being. Diff 0.02-0.1: Changed target namespace to http://www.ivoa.net/xml/prop/VOStdService. Diff 0.1-0.2: Forgot to put in all-important BaseURL element in ParamHTTPGet. A service invoked via an HTTP Get with a set of arguments consisting of keyword name-value pairs. Note that the URL for help with this service can be put into the Service/ReferenceURL element. Editor's Note: specification for inputs needs work. The capabilities of an SIA implementation. It includes the listing of the columns that appear in image query output VOTable and SIA-specific metadata. Editor's Notes: * This is a prototype definition to illustrate how to extend the schema to a speicific standard interface. * Staging information is not included yet. The columns returned in a VOTable. The largest number of records that the Image Query web method will return. The maximum image file size in bytes. The maximum image query region size, expressed in decimal degrees. A value of 360 degrees indicates that there is no limit and the entire data collection (entire sky) can be queried. The maximum image query region size, expressed in decimal degrees. A value of 360 degrees indicates that there is no limit and the entire data collection (entire sky) can be queried. The largest image that can be requested. The class of image service: Cutout, Mosaic, Atlas, Pointed This is a service which extracts or "cuts out" rectangular regions of some larger image, returning an image of the requested size to the client. Such images are usually drawn from a database or a collection of survey images that cover some large portion of the sky. To be considered a cutout service, the returned image should closely approximate (or at least not exceed) the size of the requested region; however, a cutout service will not normally resample (rescale or reproject) the pixel data. A cutout service may mosaic image segments to cover a large region but is still considered a cutout service if it does not resample the data. Image cutout services are fast and avoid image degredation due to resampling. This service is similar to the image cutout service but adds the capability to compute an image of the size, scale, and projection specified by the client. Mosaic services include services which resample and reproject existing image data, as well as services which generate pixels from some more fundamental dataset, e.g., a high energy event list or a radio astronomy measurement set. Image mosaics can be expensive to generate for large regions but they make it easier for the client to overlay image data from different sources. Image mosaicing services which resample already pixelated data will degrade the data slightly, unlike the simpler cutout service which returns the data unchanged. This category of service provides access to pre-computed images that make up a survey of some large portion of the sky. The service, however, is not capable of dynamically cutting out requested regions, and the size of atlas images is predetermined by the survey. Atlas images may range in size from small cutouts of extended objects to large calibrated survey data frames. This category of service provides access to collections of images of many small, "pointed" regions of the sky. "Pointed" images normally focus on specific sources in the sky as opposed to being part of a sky survey. This type of service usually applies to instrumental archives from observatories with guest observer programs (e.g., the HST archive) and other general purpose image archives (e.g., the ADIL). If a service provides access to both survey and pointed images, then it should be considered a Pointed Image Archive for the purposes of this specification; if a differentiation between the types of data is desired the pointed and survey data collections should be registered as separate image services.