After finding out that Google Maps has a hard limit of 2000 points per layer I had to figure a way around that to have all the points I've done so far. I was able to split the layers I've done in to three layers, a west, north and south. There are a couple of points that are on the wrong layer, but that doesn't affect anything. Those came from when I was reviewing my work and noticed I had missed them, the database puts things in order of entry.
The breakdown of the different districts is as follows:
West:
Jefferson, Petersville, Middletown, Jackson, Catoctin and Hauvers.
North:
Emmittsburg, Mechanicstown, Creagerstown, Woodsborough and Johnsville.
South:
Buckeystown and Urbana.
There are some 3600 names now on the map. By my estimate I'm more then half way done, there are still five more districts to do, but there are also all the smaller cities to finish up.
Showing posts with label GIS. Show all posts
Showing posts with label GIS. Show all posts
26 January 2015
07 January 2015
Google Earth issues
I'm on a steady pace on entering all of the Titus atlas into Google Maps, however there seems to be an upper limit to the number of items that can be entered. Currently I'm around 2700 names, I believe there are around 6500 names total in the atlas. Google Maps seems to have an upper limit of 2000 items per layer and will not let me import my data. Currently I'm trying to find a work around, possibly splitting the layers in to 3-4 different ones.
So far I have finished the following districts:
Buckeystown, Middletown, Creagerstown, Emmitsburg, Catoctin, Hauvers, Petersville, Jefferson, Mechanicstown and Jackson.
Next I will start working on Urbana and I will also try and figure out a way to put all the information I have on Google Maps.
So far I have finished the following districts:
Buckeystown, Middletown, Creagerstown, Emmitsburg, Catoctin, Hauvers, Petersville, Jefferson, Mechanicstown and Jackson.
Next I will start working on Urbana and I will also try and figure out a way to put all the information I have on Google Maps.
19 December 2014
More on the online Titus atlas
If you've seen the online atlas you've most likely noticed that each point is only given with a last name. The reason for that is because of the way I've set up the database.
This is what the columns of the database look like. The first three are for the GIS, the fourth is what is generally the first initial of the property owner. Sometimes it will say "Heir to" and occasionally even give the full first name. Then is a column for the surname, and finally one I've labeled "Source", which I'm thinking about changing to "District", which lets me know which map it came off of making it easier to find.
The reason I separated the first name and surname was to make it easier to search, and at a later date I hope to change all the known abbreviations to full names. However, this causes problems that I never would have thought of when trying to convert the file to something Google Maps can use. For whatever reason, even though when the file is converted to a .KML file all the columns remain, going to Google Maps the columns are not kept unless you specify one column. I have been looking for a way to combine the two, but so far have not had much luck.
If you look at number 1661, it says "Null" for the first name and is blank for the last name. What this is is a place holder. At times the maps will place a square but not have a name or leader attached to it making it hard or impossible to know who owns it. I've tried not to guess at who owns the property, but I wanted as many of the properties listed as possible; on Google Maps they're listed as "Untitled."
Below I'll point out some of the difficulties in reading the maps and how I've dealt with those difficulties.
Looking at 1 you'll notice there are a couple of "{" type leaders, this shows that person, in this case Dr. W. S. Pherson. However, the left symbol also connects it to one lower and closer to the road, but the "{" points to the Iron Ore as well. At first I thought that maybe Dr. Pherson owned three properties, two of which were Iron Ore deposits. However, the Iron Ore symbol is just above the I & O.
Another problem in this area is the "Aburn Farm", something I have no idea what it really means. Obviously there is a farm there, but the exact location is not given. In this case I have placed it on a different layer that has not been added to the Google Map yet.
Near 2 you'll notice the Catoctin Furnace, a grist mill and a saw mill. I used the symbol below the C for the location of the furnace. For the two mills however I didn't know which symbol was being referred to so I used the next closest ones. In the same general area there are a number of unlabeled symbols, all of which are labeled as "Null".
The 3 on the right side shows another problem in the area. Notice how the B.S + Rectory has a leader showing which building they are in? Directly below that it says "Epis. Ch." but there is no leader. Which building is it? Is it in the same as the B.S. and Rectory or one just to the south? What I've done is placed the church on its own layer in the same area. Zooming in on the area in Google Maps shows that Harriet Chapel is right there and has been in existence since 1833 and is most likely the church on the map.
You might also notice in the same area the name J. B. Kunkel, he seems to own a lot of property in that general area.
Luckily these problems do not come up often, generally it's pretty easy to see who owns what property. The other times an educated guess is all I have to go on.
This is what the columns of the database look like. The first three are for the GIS, the fourth is what is generally the first initial of the property owner. Sometimes it will say "Heir to" and occasionally even give the full first name. Then is a column for the surname, and finally one I've labeled "Source", which I'm thinking about changing to "District", which lets me know which map it came off of making it easier to find.
The reason I separated the first name and surname was to make it easier to search, and at a later date I hope to change all the known abbreviations to full names. However, this causes problems that I never would have thought of when trying to convert the file to something Google Maps can use. For whatever reason, even though when the file is converted to a .KML file all the columns remain, going to Google Maps the columns are not kept unless you specify one column. I have been looking for a way to combine the two, but so far have not had much luck.
If you look at number 1661, it says "Null" for the first name and is blank for the last name. What this is is a place holder. At times the maps will place a square but not have a name or leader attached to it making it hard or impossible to know who owns it. I've tried not to guess at who owns the property, but I wanted as many of the properties listed as possible; on Google Maps they're listed as "Untitled."
Below I'll point out some of the difficulties in reading the maps and how I've dealt with those difficulties.
 |
| Mechanicstown, District 15, page 37. Southern part. |
Looking at 1 you'll notice there are a couple of "{" type leaders, this shows that person, in this case Dr. W. S. Pherson. However, the left symbol also connects it to one lower and closer to the road, but the "{" points to the Iron Ore as well. At first I thought that maybe Dr. Pherson owned three properties, two of which were Iron Ore deposits. However, the Iron Ore symbol is just above the I & O.
Another problem in this area is the "Aburn Farm", something I have no idea what it really means. Obviously there is a farm there, but the exact location is not given. In this case I have placed it on a different layer that has not been added to the Google Map yet.
Near 2 you'll notice the Catoctin Furnace, a grist mill and a saw mill. I used the symbol below the C for the location of the furnace. For the two mills however I didn't know which symbol was being referred to so I used the next closest ones. In the same general area there are a number of unlabeled symbols, all of which are labeled as "Null".
The 3 on the right side shows another problem in the area. Notice how the B.S + Rectory has a leader showing which building they are in? Directly below that it says "Epis. Ch." but there is no leader. Which building is it? Is it in the same as the B.S. and Rectory or one just to the south? What I've done is placed the church on its own layer in the same area. Zooming in on the area in Google Maps shows that Harriet Chapel is right there and has been in existence since 1833 and is most likely the church on the map.
You might also notice in the same area the name J. B. Kunkel, he seems to own a lot of property in that general area.
Luckily these problems do not come up often, generally it's pretty easy to see who owns what property. The other times an educated guess is all I have to go on.
18 December 2014
Online version of 1873 Titus Atlas of Frederick County
For those that don't know I am a cartographer by training and trade. Currently I make aeronautical charts for the FAA. I've always had an interest in maps and have recently begun collecting maps of Frederick county and Maryland.
For those that know the maps of Frederick county the Isaac Bond map is probably the most well known. Another is the 1873 Titus atlas of Frederick county.
Both of these maps gives the location of many of the land owners in the county making it a valuable resource for genealogists. There are problems though, especially if one doesn't know about where their ancestor owned land, or if they were even listed on the map at all. The link above for the Bond map does allow for searching for names, but even with that I've had a hard time.
About a year ago I created an index to the Titus atlas to help locate land owners. Even with an index it can be difficult to locate exactly where the property is located. After finding the location of the property it can be difficult to then know where the property would be today.
In order to help I have begun creating a digital version of the Titus atlas. To do this I have been using a Geographic Information System, (GIS), to plot the approximate locations of properties in the atlas. There are of course problems that arise with such a project, some of which have workarounds, while others cause headaches.
First a bit of background about these maps. Generally these maps and atlases were not surveyed by actual surveyors or cartographers, they were done by men trying to make money. What happened was something like this: a company would decide to make a map of an area, they would then hire people to canvas the area with a machine that looked a bit like a wheelbarrow that calculated distance. Along the way they would tell people they were making a map and would include the property owner on the map, for a price. Basically it was sort of a scam, you got to be on the map, but you paid for the privilege, sort of like the "Who's Who" books. [see footnote]
Because the maps were not surveyed by professionals, the maps are more distorted then they otherwise would be. This makes the task of overlaying an older map on a modern map difficult. However, using a GIS can help by warping the map to get a best fit model. Depending on how well the map was made the warps can come out to be quite good, or really bad. Because many roads have not changed much over the years I used intersections as points of reference. After using 4-5 points the map is warped, adding more points can help warp the map further, though it can make the distortion worse.
Below are some examples from the Titus atlas of warped pages. The red lines are modern roads, blue streams, the black lines are roads on the map.
As you can see, here in Wolfsville, the warp came out quite well, many of the roads are not far off. One can also see where roads used to be and where newer roads have been added.
This warp didn't come out as well. In the upper left corner the roads match well, but down the middle the roads are starting to drift further away. While on the right side the roads are well off. When this happens I simply shift the map a bit.
When I have gotten a warp as close as I feel I can I stop adding points. Giving too many points can create strange warps that are unusable. What I do is shift the map so the roads line up as best as possible and place my locations that way. When I go to a new area I shift the map as needed. What I try to do is keep the property locations on the proper side of the road and near any intersections. Because the creeks and such were not done very well I have tended to ignore placements based upon them.
While this is not going to give one an exact location of where the original property was, it should give a good starting point. After putting the map in to Google Earth I have found that the original placement of the Washington Monument is about 300 feet off from where it should be.
Currently I have only been able to upload the points by last name, I am looking in to being able to label the locations by first and last name. I have also uploaded layers for businesses, mills and schools. I have completed the Mechanicstown, Hauvers, Catoctin, Jackson, Middletown, Petersville and Jefferson districts.
The online map can be found on Google Maps.
[I've been informed that one does not pay to be in the Who's Who books. One didn't actually pay to be in the atlas either, one just paid for the atlas and when you paid your name was placed in to the atlas.]
For those that know the maps of Frederick county the Isaac Bond map is probably the most well known. Another is the 1873 Titus atlas of Frederick county.
 | |
| Outline of Frederick county and election districts from Titus Atlas |
Both of these maps gives the location of many of the land owners in the county making it a valuable resource for genealogists. There are problems though, especially if one doesn't know about where their ancestor owned land, or if they were even listed on the map at all. The link above for the Bond map does allow for searching for names, but even with that I've had a hard time.
About a year ago I created an index to the Titus atlas to help locate land owners. Even with an index it can be difficult to locate exactly where the property is located. After finding the location of the property it can be difficult to then know where the property would be today.
In order to help I have begun creating a digital version of the Titus atlas. To do this I have been using a Geographic Information System, (GIS), to plot the approximate locations of properties in the atlas. There are of course problems that arise with such a project, some of which have workarounds, while others cause headaches.
First a bit of background about these maps. Generally these maps and atlases were not surveyed by actual surveyors or cartographers, they were done by men trying to make money. What happened was something like this: a company would decide to make a map of an area, they would then hire people to canvas the area with a machine that looked a bit like a wheelbarrow that calculated distance. Along the way they would tell people they were making a map and would include the property owner on the map, for a price. Basically it was sort of a scam, you got to be on the map, but you paid for the privilege, sort of like the "Who's Who" books. [see footnote]
Because the maps were not surveyed by professionals, the maps are more distorted then they otherwise would be. This makes the task of overlaying an older map on a modern map difficult. However, using a GIS can help by warping the map to get a best fit model. Depending on how well the map was made the warps can come out to be quite good, or really bad. Because many roads have not changed much over the years I used intersections as points of reference. After using 4-5 points the map is warped, adding more points can help warp the map further, though it can make the distortion worse.
Below are some examples from the Titus atlas of warped pages. The red lines are modern roads, blue streams, the black lines are roads on the map.
As you can see, here in Wolfsville, the warp came out quite well, many of the roads are not far off. One can also see where roads used to be and where newer roads have been added.
This warp didn't come out as well. In the upper left corner the roads match well, but down the middle the roads are starting to drift further away. While on the right side the roads are well off. When this happens I simply shift the map a bit.
When I have gotten a warp as close as I feel I can I stop adding points. Giving too many points can create strange warps that are unusable. What I do is shift the map so the roads line up as best as possible and place my locations that way. When I go to a new area I shift the map as needed. What I try to do is keep the property locations on the proper side of the road and near any intersections. Because the creeks and such were not done very well I have tended to ignore placements based upon them.
While this is not going to give one an exact location of where the original property was, it should give a good starting point. After putting the map in to Google Earth I have found that the original placement of the Washington Monument is about 300 feet off from where it should be.
 |
| Difference in placement of Washington Monument |
Currently I have only been able to upload the points by last name, I am looking in to being able to label the locations by first and last name. I have also uploaded layers for businesses, mills and schools. I have completed the Mechanicstown, Hauvers, Catoctin, Jackson, Middletown, Petersville and Jefferson districts.
The online map can be found on Google Maps.
[I've been informed that one does not pay to be in the Who's Who books. One didn't actually pay to be in the atlas either, one just paid for the atlas and when you paid your name was placed in to the atlas.]
Subscribe to:
Posts (Atom)