This is a common challenge for many Cadastre or Cartography projects, which in the 2000- 2010 period integrated Microstation Geographics as a spatial data driver, considering reasons like the following:
- Arch-node management was and continues being highly practical, for cadastral
- The DGN is an attractive alternative, whereas its version in the same file, which has not changed in 15 years, contrary to other formats in which we have seen many incompatible versions every three years.
- In 2002 free software was a distant dream of what we have today.
- OGC standards were not even important on proprietary software.
- The shp files were limited to high depth projects and spatial bases were still very closed to nonstandard schemes that commited servers performance… and money.
- Remote connectivity was incipient compared to what we have now.
So, to implement a GIS based on a “Linked CAD” scheme was a viable solution, despite usability was sacrificed for attractive presentation purposes. The VBA API was abundant for programmed transactional management routines connected to ProjectWise for physical files control and the possibility to use GeoWeb Publisher for Spatial Server Analysis, although the publication was limited to ActiveX in Internet Explorer (which in that year was the only browser).
The problem is not having gradually evolved and rather than passing to Geospatial Server, or more ProjectWise robust versions, what we wanr is to survive a GIS from physical files, taking full license Oracle Spatial potential and having ability to develop. So that was our challenge.
1. The database: Postgres, SQL Server or Oracle?
In particular, I would have preferred the first. But when you’re facing a not service-oriented transactional system but that is working well, in which part of the logic and integrity is as PL in the database, the change to an OpenSource base is not an emergency. No, unless you have in goal to develop a new system version that is not in the immediate term.
Nor it is making a taliban action belittling everything that smells as proprietary. So stay with Oracle is a wise decision, if it is running fine, if the size and demand is large, if it is well designed, protected and if it is taking advantage of the support. Topic for another time.
So that what remained was to develop features so that the data is migrated to this basis, publishing services and vector data transactional management tools.
To control the roles and users, who were formerly managed from ProjectWise, it was created a modular tool allowing:
- Manage users and roles from the VBA and Bentley
- Assign, from the user with administration rights, the right to departments and municipalities.
- Assign right to cadastral tab by project.
- Have right to available tools in construction, edition, publication, consultation and administration modules. In this way, new applications are only created and appeared to users according to their role or specific assignment.
- This login panel also simplifies the BentleyMap common complexity projects, such that with only entering the categories and attributes tree defined in the Geospatial Administrator.
A panel like this solves little understanding hassles and new users risks to features such as Data Interoperability. That is another topic, as Bentley edits natively in Oracle Spatial, which is wonderful but also risky If you don’t have the transactional control.
Thus, for example, the construction module had the following tools:
- Assign Features
- Geographic link Wizard
- Batch Spatial Migration
- Delete objects
- Edit polygons
- Shp/CAD Export
- Shp/CAD Import
- Geoline Migration
- Geodot Migration
- Georegion Migration
- Map Register
- Geo-line Link
- Geo-point Link
- Region Link Geo-
The complementary tools were gradually added, including some to directly edit Geospatial Administrator.
- Administrator to view features
- Topological analysis
- SAFT Consultation
- Check Feature
- Convert curve to LineString
- Create Features
- Create properties
- DBConnect configuration
- DBConnect inquiry
- Edit Xfm feature
- Edit Xfm project
- Eliminate Xfm Features
- Land identification
- Modify symbology
- On-write features
- Theming by classes
- Theming
- Theme by unfolding list
- Xfm Utilities
2. The data: DGN to space-based migration: Oracle Buider or Bentley Map?
The most interesting challenge in this was, that it required a controlled migration and taking note that DGN files after receiving update for more than 10 years might have topology problems a really sheer madness.
Indeed it was. The main maps problems are here:
- Plot Modification at the file border (sector or area) implies that there must be a modification of both, including the matching nodes in cases such as when a sector is a single line but in the neighbouring that line is segmented.
- There are files that after 300 maintenance transactions stored in the DGN history can corrupt.
- There are more complex problems non-controllable in Cabinet, as when an area overlaps on another neighbor one in another file, by amounts that cannot be resolved on the map because it would imply field inspection to avoid affecting a third party.
- Bad practices, such as the maps inclusion in different projections, in this case there were sectors in NAD27, though the standard was WGS84. In extreme cases adjustments between data of different projections were made, directly ato the wicked mode.
The solution was a Wizzard type tool for a mass migration, which can migrate a map individually, several or even all of a municipality (Town Hall) or Department.
Basically what the tool does is take the Geographics project data and promote them to Benltey Map features, then make validations series, such as:
- One to one relation between geometry and data base,
- Non-duplicates validation,
- Centroid-area consistency validation,
- Objects validation with respect to inactive objects in the database,
- Topology Validation with respect to existing space-based topologies
After validations, panel allows you to add information on a massive scale, such as measurement method and standard quality control of such data.
Finally, it posts to the database, generating a report eventually. From said to the fact there is a tremendous extend, but it was finally adjusted to the Oracle Spatial vagaries that does not cease to be so crazy as Bentleys and its way of seeing the complex plots or plots with many vertices.
3. The publication: Geoserver or MapServer? OpenLayers or Leaflet?
It was built a viewer using OpenLayers and some plugins. For the first time after 10 years of the space part development neglect, a new Viewer that replaced the GeoWeb Publisher ActiveX was visible. MapFish code was used for the printing, geojson to control side tree, and from Geoserver OracleSpatial layers were served.
Finally the technologies replacement was made according to the following chart. As you can see a free code combination, keeping the database and land management using proprietary software.
4. Construction and editing, directly to Oracle Spatial. Bentley Map or QGIS?
This is another story. Bentley Map edit the native at the spatial base, which generates conflicts if it does not, will work with a Web Feature Service Transactional (WFS). The conflict is:
How to solve a rule that does not allow overlap topology, if you are editing and when posting reports that the object affects itself?
This is resolved versioning it before, editing directly and validating that when posting, if something fails, versioning recovers but leaves the transaction in failed state.
Another problem that had to be overcome is the massive data entry, considering that users should stop using Geographics and had several projects raising massive cadastre.
This was easy since it was only done with a similar tool similar to the existing one to integrate data into Microstation Geographics, facilitating the task with the BentleyMap potential and one more controlled assistant.
The image shows how this tool was developed, with some particularities, such as the vertices creation and registration and the Dotplot inclusion, as a ready functionality in case that the measurement method of some vertices do not meet some quality standard.
Definitely this flow was very well, because the users knew what tools were most frequently used. It was necessary to make them a mentality change between the passage from multiple features to management levels, promoting new benefits so that they forget the archaic Microstation V8 2004, such as the WMS service, transparencies and DWG files native recognition of recent versions; what not be said about interoperability with shp, kml, gml for the most astral.
Similarly were done tools for cadastral maintenance, having the option to directly edit in shapes or lowering them to arch-node for complex cases.
5. Client to municipalities via GML. QGIS or gvSIG?
QGIS. But, thats another story for telling later.