Click for next page ( 26


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 25
4 Technical Path to an Open System The advantages of a digital production environment for data-intensive prod- ucts such as nautical charts and chart information were discussed in the previous chapters. The questions that remain concern the details of how to achieve the transformation most efficiently and rapidly. This chapter reviews NOAA's digital production activities and outlines the steps necessary to move the digital system forward in the short and long term in the most cost-effective and technologically advanced way. Detailed descriptions of NOAA's current nautical information systems are found in the technical re- view and evaluation of ANCS II (Goodchild et al., 1995~. CURRENT SOFTWARE TOOLS ANCS II is a digital chart production system developed for NOAA through a private vendor. ANCS II comprises six major custom software modules in addi- tion to commercial software (approximate number of lines of custom code for each module is shown in parentheses): task management (40k lines) source logging (20k lines) source evaluation (20k lines) nautical information database software (160k lines) chart graphics database software (240k lines) chart finishing (20k lines) Task management, source logging, and source evaluation are functions for which ANCS II provides integrated software tools. Outside ANCS II, NOAA has 25

OCR for page 25
26 NAUTICAL CHART PROGRAM a stand-alone database system for source logging. Task management and source evaluation currently are carried out manually. The nautical information database (NIDB) software manages the application of new source data to the NIDB, the repository of scale-independent vector data. The chart graphics database (CGDB) software manages the maintenance of vec- tor data describing scale-specific depiction of NIDB data in specific chart prod- ucts. This information on how to build each specific chart product from the NIDB is stored in the CGDB. Both NIDB and CGDB are approximately 3 percent loaded at this time, although the data are not yet up to date. Chart-finishing software produces a raster chart image from the CGDB vector data. Mariner, a commercial software product, is used in conjunction with NIDB and CGDB software to cull survey source data from soundings to be retained in the NIDB and to select from NIDB soundings that will be shown in a particular chart product. Mariner has a small installed base (fewer than 100 licenses), and unlike other products that are part of ANCS II, the vendor has no plans at present to link Mariner to a routinely supported platform; consequently, it will be expen- sive to maintain. The Super Computer-Assisted Revision System/Computer-Assisted Compi- lation (SCARS/CAC) is a software tool developed from commercial off-the-shelf (COTS) products and chart-finishing software used in ANCS II to compile and edit raster chart images. The data repository associated with SCARS/CAC con- sists of raster images representing each chart product. This repository is fully loaded, but is not up to date for all charts. SCARS/CAC allows cartographers to edit raster images of paper chart products, and it eliminates the expense and delay associated with manual engraving. Current capabilities require scanning markups and applying changes to the raster chart by superimposing the markup. Planned refinements will increase the ability of SCARS/CAC to accept vector source data as well as scanned (raster) images. Most of the workstation hardware on which ANCS II components were de- veloped is obsolete by present-day standards. Furthermore, the components of the ANCS II operating system are no longer routinely supported by the vendor. This will make future maintenance expensive and complicate the replacement of obsolete hardware. To move digital production system tools forward in a direc- tion consistent with the themes and considerations described in chapter 3, NOAA would need to move to a hardware platform and operating system that could be updated and expanded with COTS software and hardware (hereafter referred to as "sustainable operating environment"-for example, Windows NT or UNIX/ open systems). This move would also give NOAA greater flexibility in managing the total life-cycle costs of operations and maintenance refurbishment, replace- ment, and system evaluation. To make the transition to an open environment, decisions must be made on how best to implement each functional component of the data management and production system in software.

OCR for page 25
TECHNICAL PATH TO AN OPEN SYSTEM 27 PATH FORWARD To move forward, choices will have to be made in selecting in-house tools, contractor-developed tools, and COTS software to create the most economical means of providing required functions. The evaluation of each component should take into account the performance and productivity, as well as the total life-cycle cost, of each alternative. Some limits on performance may be acceptable (e.g., with COTS products) if they can be acquired and maintained at a much lower cost. In general, these trade-off analyses need to employ a cost-recovery or cost- benefit analysis that takes into account total life-cycle costs. In the following sections, the elements that should to be taken into account in conducting the evaluation of ANCS II are discussed. In each case, comparison should be made between the migration and maintenance cost of the existing ANCS II or SCARS/ CAC component and the cost of a COTS alternative. This comparative informa- tion could be used to support a decision to either migrate the ANCS II component to a sustainable operating environment or abandon the ANCS II component and use COTS (and SCARS/CAC) software for this function. The following discus- sion examines alternatives for each function. Task Management Automated task management is important for the efficient maintenance of nautical chart data and products. Migration and maintenance costs of the ANCS II task management components should be compared to the costs of COTS prod- ucts that provide similar functions. In conducting the evaluation, it should be noted that the nature of tasks to be managed in the evolving data management and production environment will change over time and may differ (even in the near term) from the ANCS II tasks if ANCS II is not deployed as designed. Source Logging Source logging refers to tracking source material disposition from the time it arrives at NOAA, through its use in updating charts and other information products, to its storage in an archive. Source logging is necessary to the efficient manage- ment of source material for use in updating NOAA databases and products. Source Evaluation Source evaluation is the analysis of source materials for reliability and rel- evance to the updating of databases and products. Source evaluation is important to the efficient and accurate application of new source data to databases and products.

OCR for page 25
28 NAUTICAL CHART PROGRAM Vector Database Management The master vector database (the NIDB in ANCS II) is at the heart of NOAA's function as data manager. Broadly, this database is the repository of the largest- scale (most accurate) representation of nautical chart information for U.S. waters. New source information updates this database, and the database, in turn, updates data sets and products. As with the other elements of the ANCS II system, migration and mainte- nance costs of the NIDB software component could be compared to the cost of a COTS solution to the vector database requirement and serve as the basis for a decision to either (1) migrate all or part of the NIDB software to a sustainable operating environment or (2) abandon the NIDB software and use COTS products for this function. In conducting this comparison, it should be noted that not all functions of the NIDB software may be needed, depending on the choice of output management tools (see below). If the NIDB software is retained, the Mariner com- ponent might still be replaced with COTS products if the existing Mariner com- ponent cannot be migrated economically to a sustainable operating environment. Three main options for loading the vector database (the NIDB in ANCS II) are (1) loading specified themes for critical regions from existing source data, (2) importing Digital Nautical Chart/Vector Product Format (DNC/VPF) data, and (3) digitizing NOAA paper charts. These options are not mutually exclusive. It appears that loading data from an appropriate combination of the sources above is a reasonable route to acquiring sufficient vector data for creating prod- ucts in the short term. If SCARS/CAC continue to be used (see below), vector data used to update raster images in SCARS/CAC could also be brought back to the vector database with relatively modest effort and cost. New source data could routinely be brought into the vector database once it is operational. On-Line Data Warehouse To meet evolving customer needs in the long term, it is likely that NOAA will need to develop the capability of making digital data available to users and vendors directly through an on-line data warehouse. Such a service would pose substantial challenges to NOAA technical, financial, and organizational-as well as create the need for additional expertise to develop appropriate network- ing, distribution, and interactive capabilities. To enable vendors to provide prod- ucts from NOAA data with NOAA's official sanction, NOAA might consider developing standards and procedures to ensure that the data reach users without degradation. Output Management Output management includes the production of raster, paper, and vector prod- ucts, as well as the generation of data sets to be made available through a central

OCR for page 25
TECHNICAL PATH TO AN OPEN SYSTEM 29 data clearinghouse or warehouse that would be easily accessible to users. The two main alternatives are: (1) Vector editing of output. Migrate ANCS II CGDB and chart-finishing software to a sustainable operating environment and develop software to convert vector data from the CGDB to a format suitable for export to electronic chart vendors and users (this requires loading the CGDB). (2) Raster editing of output. Use SCARS/CAC to produce raster and paper products and develop software to convert vector data from the master vector data- base to a format suitable for export to electronic chart vendors and users. This may require more extensive software development, since the master vector data is not compiled to specific scales. Alternative (2) is the (:C)TS alternative, since SCARS/CAC is built largely on COTS software. Note that with either alternative it will be necessary to de- velop or acquire software to extract vector data for export to electronic chart vendors and users. The alternatives must be compared in terms of total life-cycle costs. It is likely that database loading costs will be greater for the first alterna- tive, while (new) software costs will be greater with the second alternative. EXAMPLES OF ALTERNATIVE SYSTEMS The following are three examples of possible alternatives for the data man- agement and production system: 1. ANCS II as envisioned in the Nautical Charting Services Production Modernization Plan 2. partial ANCS II with SCARS/CAC 3. SCARS/CAC plus COTS, no ANCS II These examples illustrate the range of possible configurations and are illustrated in figures 4-1, 4-2, and 4-3. Alternative 1 (figure 4-1~. ANCS II is implemented as envisioned. NOAA proceeds with full deployment of ANCS II as described in the Nautical Charting Services Production Modernization Plan (NOS, 1995) and discontinues further development and use of SCARS/CAC. NIDB and CGDB are fully loaded. A migration strategy is developed and pursued to move ANCS II to a combined hardware and software environment that makes it economically sustainable. Alternative 2 (figure 4-2~. Partial ANCS II with SCARS/CAC. NOAA retains the NIDB software and database components of ANCS II, as well as task management, source logging, and source evaluation, and links the retained ANCS II software with SCARS/CAC. The retained parts of ANCS II are migrated to a sustainable platform; Mariner is replaced by COTS as appropriate. Source data are managed and maintained in the NIDB and routed from there to SCARS/CAC.

OCR for page 25
30 Source P ..-.. NIDE ; ~ Vector : Themes: . . . , _ ., ~ . NAUTICAL CHART PROGRAM I Alternative 1 | ~- ANCS II ANCS II _ SourceP.' _ Source Logging ~Ed aluation Pit' ~ ANCS II NIDB Software I , ANCS II EC (to be den eloped) 1 , ANCS II ANCS II CGDB ' _ Chart Sore are , Finish JO \ SCARS ~CAC ~\ _ _ _ _ _ _ \ ,%~ . . .. Raster : Separates: . . , . .,~ . - _ Vector Data Products r Rasterlpaper ~Product EC = electronic chart p = paper v = sector r= raster FIGURE 4-1 ANCS II as envisioned. Note: Components not used in this configuration are indicated win broken lines. SCARS/CAC remains the production system for raster and paper chart products. Vector products are provided from the NIDB with software to be developed to the extent that it is loaded. Alternative 3 (figure 4-3~. SCARS/CAC plus COTS, no ANCS II. ANCS II is abandoned entirely. SCARS/CAC is developed further and integrated with a vector database and COTS software for management, logging, evaluation, and other functions. Vector products are provided from the vector database, with soft- ware to be developed, to the extent that it is loaded. The alternatives described above require different database preparation methodologies. Alternative 1 (ANCS II as envisioned) requires population of both the NIDB and the CGDB. In this alternative, the CGDB must be loaded through the automated and interactive processing of data from a fully populated

OCR for page 25
TECHNICAL PATH TO AN OPEN SYSTEM 31 NIDB using ANCS II. Alternatives 2 and 3 do not make use of the ANCS CGDB and, therefore, do not require it to be loaded. However, they each require the development (or procurement via COTS) of new software to compile vector prod- ucts from the main vector database (NIDB or others). NOAA has several choices for achieving the population of the NIDB: using original hydrographic source data and compiling it into vector data using the digitizing capabilities of ANCS II to populate the NIDB from paper-source photographic separates using commercial contractors to digitize paper-source photographic separates using digital data automated by the Defense Mapping Agency in the digi- tal nautical chart format ANCS II Source p.v / _' Vector Themes Alternative 2 ANCS II Source . Logging p.\ ~ p.r NOB Source Evaluation ANCS II P,~ NIDB Soft vare ANCS II CGDB Soft vare SCARS / CAC _ ~ ~J C ~SePar ~ ~CGDB ' -"' ANCS II EC (to be den eloped) . ........ : ANCS II Chart Finish Vector Data Products rRaster/Paper ' Product EC = electronic chart p = paper = sector r= raster FIGURE 4-2 Partial ANCS II with SCARS/CAC. Note: Components not used in this configuration are indicated with broken lines.

OCR for page 25
32 NAUTICAL CHART PROGRAM | Alternative3 l , . : ANCS II : Source Logging - Management p.` Logging Source ~Es aluation (sit` or manual) ' in\' .1 I/ ,' Vector Themes NIDB . ...... : ANCS II ANCS II EC (to be developed) , .............. : ANCS II : NIDB Software , .............. ANCS II , ............. : ANCS II . . CGDB Chart : Softu are : : Finish . , . \ps.r \9 Of / 1 ~SCARS / ~^ , Vector -> EC Software /- ~(to be / / des eloped) '/r C ~ ~ -'), Raster Separates C r Raster/Papcr Prr`A''rt Vector v ~ Data Products EC = electronic chart CGDB v - ~ ector r= raster FIGURE 4-3 SCARS/CAC Plus COTS, no ANCS II. Note: Components not used in this configuration are indicated with broken lines. These choices are not mutually exclusive. A discussion of costs associated with the generation of vector data for the NIDB is provided in appendix E. Under alternative 2 (partial ANCS II with SCARS/CAC), the CGDB does not exist. Options for the population of the NIDB are the same as for alternative 1. Under alternative 3 (SCARS/CAC plus COTS, no ANCS II), the CGDB also does not exist. The requirement to move toward production of ECDIS (Electronic Chart Display and Information System) vector data would imply that a vector database of some type would still exist with content similar to the proposed ANCS II NIDB. Again, the options for populating it would be the same as alternative 1, except that the option to use ANCS II digitizing capabilities for population would not exist.

OCR for page 25
TECHNICAL PATH TO AN OPEN SYSTEM COMMENTS ON THE NOAA MODERNIZATION PLAN 33 The Nautical Charting Services Production Modernization Plan (NOS, 1995) (see summary in appendix B) recognizes that decisions must be made in the near future about the best options for fulfilling the functions of the digital nautical data management and production system. In light of the choices avail- able to NOAA and discussed in this report, the plan would benefit from being modified to reflect more explicit decisions and to ensure that the necessary in- formation to support these decisions is generated in a timely fashion. Such a modification would minimize the likelihood of using scarce resources on projects that will not be sustained.