Based on initial investigations of storing source code in the database, I’m now focusing on storing assemblies in the database, serialized as blobs. Doesn’t make sense to be recompiling all the time. Instead, I just serialize the dll file as a byte array, then store that via memoryblobstream to a blob field. Deserialize using System.Reflection.Assembly.Load(Byte[]). In effect, this allows using a DBMS in lieu of the GAC (Global Assembly Cache).

The deployment issues this addresses are not peculiar to GIS, yet I can’t find much discussion of this. I’d be interested in hearing from anyone who has tried this. Treating code as data is receiving attention these days, so why not store assemblies in the database?

ESRI’s discussion of Enterprise DBMS seems to avoid discussing the inability to centrally manage business rules using triggers in a versioned geodatabase. I’m working on an Editor Extension that manages triggers instantiated from a database assembly cache to compensate for this.

Another common use of editor extensions is to trigger some code in response to editor events. (from “About Editor Extensions” in EDN)

It seems like a domain specific language could be derived that focuses specifically on managing triggers, similar to the approach described here by the Mechanical Bride, except the DSL code would be compiled into an assembly that would then be stored in the database.

The Atlantic Monthly has an interesting article suggesting that Section8 vouchers are correlated with crime. I find it odd that no maps are provided (haven’t seen the print edition yet).

On the merged map, dense violent-crime areas are shaded dark blue, and Section8 addresses are represented by little red dots. All of the dark-blue areas are covered in little red dots, like bursts of gunfire. The rest of the city has almost no dots.

A description of Memphis from American Murder Mystery

Somebody has some splaining to do.

I like to think of myself as a patient person, not in a hurry. However sometimes when I use the Field Calculator in Arcmap my patience is challenged.

I have a featureclass of sidewalks related to a street centerline network. 284000 sidewalks and 54000 street centerlines. When I calculate a field on the sidewalks, setting it equal to a field from the street network, it takes several hours. That’s in a file gdb with the key field indexed. Somebody has some explaining to do.

Last time I read ESRI’s EULA, I recall it prohibiting publication of benchmark statistics, so I have not included precise numbers. Looks like they are following Oracle’s policy.

Unlike Oracle, ArcGIS does not allow me to determine what execution plan it is using (see EXPLAIN PLAN). I’m quite certain it always uses the same plan, which in this case is a very poor one.

To make it faster, I cache the key/value pairs from the street attribute table into memory using a System.Collections.Hashtable. This takes about 1.4 seconds. I then open a cursor on the sidewalks table and loop through it, looking up the value from the hashtable using the key field value from the sidewalk featureclass. This takes less than a minute.

Interestingly an update cursor is slower than using a cursor created via IFeatureclass.Search. I think this is just with a file geodatabase though, on ArcSDE I believe an update cursor is generally faster, assuming proper rollback segment sizes are configured.

Maybe what ESRI should do is to beef up the MemoryRelationShipClass so that it allows the user to examine and/or specify an execution plan. That way I could tell it to use a hashtable when it does the join, alleviating me from having to roll my own.

For a good comparison of hashtables, sortedlists and dictionaries, see this post.

The shortcoming of this approach is that I can’t do field calculations, just simple assignment. Some day I’d like to try using CodeDOM to generate code from an expression the user has provided. It would need to substitute in the field values for field names. Since square brackets are used in C# I’d need different field name delimiters.

Enough room for plywood. Plus its got a microwave.

A benchmark for minivans is whether or not they can carry a 4′x8′ piece of plywood. Presumably minivan drivers often need to do this.

What if there were web site where that allowed a neighborhood of homeowners to collectively bargain for products like lumber, gardening supplies etc., reducing delivery costs?

Users would put together a wish list of materials and open it up for bids. Sellers like Home Depot, Lowes, etc. would put together a bid, including the delivery charge. Since houses in the same subdivision often have the same construction materials, all of similar vintage, things like fences, faucets and shingles often need to be replaced at about the same time throughout an entire neighborhood.

The MyHouse concept I’ve blogged about previously could be leveraged so that neighbors could connect with other homeowners whose houses have the same material. By coordinating their purchases, they could get more competitive bids. Think of it as BIM for the masses.

Our old house had a concrete tile roof, along with all the other houses in the neighborhood. When we had a leak we found a tile had cracked and needed to be replaced. The repair man had great difficulty locating a seller that had any in stock. Likely others in the neighborhood also need tiles too.

Sure seems like this sort of neighborhood based social networking could be monetized.

More on this later.

Been looking through Digipede and wondering if it would be possible to take a grid based approach to map tile generation. Unlike most other grid solutions, this one is based on .NET, which means grid based ArcObjects apps should be possible. Since ArcSDE would likely become the bottleneck, seems like each node could cache a geographic area into an InMemory workspace, and generate tiles from that.

While I never bought into the view that the military operates at the direction of the oil industry, I began to have second thoughts this morning. I was looking for a particular software vendor with whom I had an appointment in one of the exhibit halls of the convention center. Booth after booth showed all sorts of high tech military surveillance systems. ESRI had a booth, so I figured I must be in the right place. Wrong. Turns out it was a conference for the Armed Forces Communications and Electronics Association. The registration desk staff in military uniforms had given me an exhibit hall pass, no questions asked, after I handed them my AAPG invitation. I’m not sure what sort of registration system they were using. It was electronic though.

After realizing I was in the wrong hall, I headed next door, where the American Association of Petroleum Geologists were holding their meeting. I missed hearing Ray Hunt deliver the keynote speech the previous evening. You may have heard of him. He’s an oil man from Dallas who’s on the board of directors at Halliburton. Also coincidentally, he was appointed by Bush in 2001 to the President’s Foreign Intelligence Advisory Board. I’m not sure if he was also speaking at the military conference.

What is striking about the AAPG exhibit hall, is that among the booths of high tech hardware and software vendors are many booths selling beautiful stone jewelry. The ammonite fossil jewelry seemed to present extinction as an art form. Ammonites look a lot like the Chambered Nautilus, in Oliver Wendel Holmes’ poem. While the poem celebrates unlimited growth, the ammonite fossils suggest limits.

From Jess Strykers’ Landscape Sprinkler Design Tutorial

MyHouse is a concept I’ve been thinking through for a while. It is an idea for an internet site that allows a person to spatially organize their life, using their house as a starting point. This article examines a use case for water sprinkler system design.

Suppose the user has bought a house from a seller (builder) that was smart enough to include georeferenced parcel boundary, building footprint and underground utilities for the house on the MyHouse website (a big selling point!). Pipes have been tagged with diameters. 3D trees have been placed around the yard. The user would now like to install a sprinkler system.

Landscape Design (Bring Me a Shrubbery!)
The user logs into his MyHouse account and draws plants onto a map of his yard. MyHouse has links to nurseries where the user can browse, drag and drop things like shrubbery. Each plant placed on the map can be tagged with its water requirements and height. The user marks off barriers (areas/lines) where laying pipe is not allowed.

Design the Pipes
The user submits a job, specifying whether the design should minimize water usage or material cost. The job runs on a server in the background. The user does not care if it uses a monte carlo method, a genetic algorithm, or cheap overseas labor. The job may request one or more designs to be generated. Each design includes a bill of materials. The design takes into account the barriers for underground pipes, as well as potential spray blockage by trees/shrubs.

The user can check on the status of his job, along with an option to cancel the job.

Design Review
When the job is completed, the user receives an email with a link to a 3D design showing spray patterns and how the shrubs/trees block the spray pattern. An operation plan is provided that times each section of the system to assure different plants receive proper water amounts, including maps of “rainfall” levels.

The user then looks over the alternative designs and selects his favorite. He then clicks on a button that says “request quotes”. Competing sprinkler system installers are sent an invitation to bid on the project.

Construction
As the system is built, the design is updated to reflect actual materials used.

Calibration
The user can choose to calibrate his sprinklers. The system instructs the user locations where rain gauges are to be placed around the yard, along with an operation schedule. The user places the gauges, and records water levels collected at each gauge. The system then creates an observed rainfall map, and generates a refined operational plan intended to meet stated watering requirements. The user may re-calibrate the system as many times as he wishes.

The service is free to the user. Advertisers pay to place ads for sprinkler heads, plants, fertilizer, etc. Installers pay to receive invitations to bid.

See also Sunshine maps.

While they do have slick maps, I don’t think HotPads should refer to them as heat maps. (via The Map Room).

To me, heat maps are rasters created using an inverse-distance-weighting method, or something similar. Geochalkboard has an example of a true heat map.

There’s an interesting article in the Proceedings of the National Academy of Sciences published a few weeks ago. The researchers found a strong correlation to ducks and rice fields for the spread of avian flu. I wonder if this suggests closer monitoring of large poultry farms located near rice fields here in Texas would be useful.

Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia

I see news that Microsoft is rolling out ClearFlow. When I search for INRIX and ClearFlow I don’t get any hits. Microsoft spun off INRIX a few years ago to do realtime traffic modeling. Washington Post has a write-up on INRIX here.

Note that INRIX is essentially a crowdsourcing (or VGI) scheme.

So is Clearflow competing with INRIX, or is it using it somewhere behind the scenes?