http://guinan.gsfc.nasa.gov/AIS.html (World Wide Web Directory, 06/1995)Nick White, NASA/Goddard Space Flight Center-HEASARC.
The High Energy Astrophysics Science Archive Research Center, HEASARC, at NASA/GSFC can be accessed through the WWW. The HEASARC server allows access to a vast selection of data from X-ray and Gamma-ray astronomy missions including ROSAT, ASCA, Compton GRO, Einstein and EXOSAT. There is a beta release astronomical forms interface to the HEASARC database management system, which allows browsing of various astronomical catalogs and archival datafiles. This report describes our aims & objectives, experiences, & conclusions in updating access methods for HEASARC. It also outlines the technical solution we adopted.
There was a queer momentary sensation of being turned inside out. It lasted an instant and Baley knew it was a jump, that oddly incomprehensible, almost mystical, momentary transition through hyperspace that transferred a ship and all it contained from one point in space to another, light years away. Another lapse of time and another Jump, still another lapse, still another Jump.Science fiction writers solved the problem of cosmic distance, by utilizing the mathematical property of hyperspace that all points are interconnected. Theodore Nelson applied this concept to text, coining the term hypertext. Tim Berners-Lee of CERN implemented a variant of that concept on the Internet, as the World-Wide Web. And in 1993, Marc Andreessen and his colleagues at the NCSA astonished the Internet community with a GUI browser (user interface) called Mosaic, that finally let the full potential of hypertext blaze forth, like a supernova.Isaac Asimov, The Naked Sun, Ballantine, 1957.
As we have learnt from previous revolutions in technology, especially in the fields of transportation and communication, there will be profound challenges to our abilities to harness & benefit from the powers opening up before us. The first & most obvious of those is to do with navigation. How do you locate the information you want ? A second has to do with quality, or pollution. Cyberspace is a new frontier (Richmond 1993). The lawyers haven't got here yet. There are many links (e.g. from What's New) that look interesting, but end up either not living up to snuff, or in some cases, just not living at all..
We've taken great pains to ensure that when you find our WWW servers at HEASARC, then, if you're looking for information from, or about, HEA missions, such as ROSAT, ASCA, etc.., you'll not be let down. We hope you'll find it a rewarding experience, and the rest of this paper addresses the technical issues underlying that goal.
At the heart of the system is the BROWSE program, a command-driven environment that allows a user to search one or more database tables by coordinates, name, object class, or any other valid parameter combination. The user then can display the selected data, or run analysis software on it.
Some time ago, it was decided to replace the old BROWSE CLI with a GUI. Our first prototype used a commercial portability toolkit; but then, in 1993, we observed the massive trend in the astronomical community towards the use of the World-Wide Web (WWW), & tools such as NCSA's Mosaic, Gopher, & WAIS.
We determined that our objectives could be met by using the facilities of the WWW, instead of the commercial product that we had been using. We initiated StarTrax-NGB as a WWW node to provide uniform access to multiple services of the HEASARC, e.g. bulletins, catalogs, & proposal & analysis tools of BROWSE & ROSAT MIPS.
Hypermedia (or, more loosely, hypertext) documents contain hyperlinks to other documents, anywhere on the Web. A hyperlink is a segment of text, or an inline image that refers to another document (text, sound, image, movie) elsewhere on the Web. When a hyperlink is selected, the referenced document is fetched from the Internet, & is displayed appropriately.
A WWW browser allows navigation through hypertext links defined as Uniform Resource Locators (URLs). The user simply specifies the link to be followed & a new document is retrieved & presented. WWW provides gateway access to several of the other systems, including Z39.50 (Fullton 1993).
NCSA Mosaic is a multi-platform GUI hypermedia browser that helps you to find, fetch, & display documents & data from the Internet. Although there have been browsers for the Web since 1991, Mosaic shot rapidly to prominence because it seamlessly integrates a great deal of useful functionality in a very pleasant & easy-to-use user interface. Incorporating Web, gopher, WAIS, NNTP, & FTP protocols, NCSA Mosaic talks natively to these servers, and can gateway to others (Hardin 1993).
In the `classical' mode -- which we started this project in -- the developer(s) spend a great deal of time wallowing in relatively low level code. We haven't yet achieved the goal of `reusability' in spite of the promises of modern software engineering methodologies.
In WWW mode, using HTML (HyperText Markup Language) (& ideally, Perl), since you build on that infrastructure, you can deliver functionality very quickly. This not only delivers the promises of rapid prototyping (throw one away), but also of rapid development.
The way we did the development was, one of us is a scientist, with a positive interest in this development. The other is a software developer with technical competence in the WWW. The work was very much a synergy: the scientist would propose a design, and the software developer would explain to him, why it was impossible to implement. A little while later (depending on how impossible) the developer would invite the scientist to try out his design; the scientist would then propose impossible improvements...
The iteration cycle time was often of the order of minutes, rather than hours or days -- as classically. Not only was this due to the pre-existing WWW functionality, but also because we chose to use Perl, instead of C. Perl is ideally suited to this kind of work, because it too provides a great deal of ready-made high-level functionality. For example, we converted some 30 lines of C code, for decoding URLs, into 3 lines of Perl. This is certainly an order of magnitude improvement...
The home page has 2 main features: a clickable image map, and a list of the HEASARC services available from the StarTrax-NGB WWW servers. This list is made up of hypertext links, or hyperlinks, which means that clicking the mouse on one of the underlined phrases, activates a new page associated with the corresponding HEASARC service.
xpaint was used to create the composite image that is now
most of the home page.
That composite image was ISMAP'ed so that the grid areas could
link to other parts of the server.
The image map displays a similar list of HEASARC services,
spread out on a pictorial background. Clicking on or near a phrase
activates the associated page. Some of the main features are:
We have implemented Help as hypertext. For example, see the Introduction page giving basic instruction, overview, & installation instructions.
The user can enter comments about StarTrax-NGB, which will be e-mailed to the HEASARC.
A form (html) for the client collects user information. A program on the server side processes the data, & answers the client by echoing what was entered. This is similar to the Feedback process.
Searches the HEASARC catalogs for named objects. The user is able to specify the equinox. It will use a Name Resolver, which is primarily a function to translate the name of an astronomical object to its coordinates.
This allows the user to input RA, & Dec in either degrees, or hms, plus the outer cone radius in arc min, arc sec, or degrees. The user is able to specify the equinox.
The daemon chosen determines the functionality of the server. It basically controls the flow of information between itself and the client program accessing information on the server. There is a variety of ways to structure & present information resources, & the selection of one or more of these depends on several factors; e.g. natural structure of the resource; user's expectations; available technology. Until quite recently it was usually easiest to impose hierarchical structure. Selecting one or more of these is fundamental to establishing a reliable information service. None of them yet fulfills all requirements -- they are all still evolving -- so a solution was found to minimise the effect of this by using a daemon combination. There are significant differences between them that have to be well understood to succesfully maintain a robust & useful information daemon. Consideration also needs to be given to anticipated longer-term maintenance & support from the developers of these daemons.
After installing & testing 3 of the major HTTP daemons
(NCSA's httpd, Plexus, GN)
we settled on NCSA's httpd, supplemented with some perl
programs. The information supplier can structure resources into any
arbitrary `network topology'.
The httpd daemon will also be supplemented with
Gopher & WAIS daemons.
This will greatly extend the search capabilities of the daemon while
providing additional network communications back ends.
Gateways can be used for a variety of purposes, the most common being the handling of ISINDEX and FORM requests for HTTP. Gateway programs, or scripts, are executable programs which can be run by themselves. They have been made external programs in order to allow them to run under various (possibly very different) information servers interchangably.
This sends a URL to the HEASARC WWW server, which dispatches
it to a perl program: Squery.pl.
Replies to browsers are made as MIME types (Multimedia Internet Mail
Extension), typically text/html, which is hypertext markup
language.
The user can click on one of these listed data products, to retrieve it. This invokes ftp directly on the machine holding the data products. The client interacts directly with the server at the data publishing site to retrieve images and relevant textual data. This need not be the same site as the server.
The client-side has mechanism for specifying which `viewer' to
use for a given data-type, e.g. xv for gif's.
These have to be MIME-types (Multimedia Internet Mail Extensions)
or sub-types (e.g. x-fits).
The browser consults the mime.types configuration file to
determine what action to take with the data.
For example, it may invoke a `viewer' such as saoimage.