The May 2009 release introduces a new set of features that improve usability, enhance filtering, improve data import and allow you to easily share screen shots of your reports. Since the initial beta release of SpatialKey, we have listened closely to our users feedback and focused on taking steps to enhance and improve SpatialKey. We will continue to improve and provide monthly releases with additional bug fixes and new features. Your feedback is important to this process so please let us know what features would make the biggest difference for you.
In addition to the new features, we also continue to strive towards an error/bug free environment. These fixes are ongoing and many are in direct action from bugs submitted from our beta testing community. Thank you for helping us to make SpatialKey the best product it can be! Keep the bug reports coming.
What follows next is a high level overview of the features introduced in this release:
- Zoomable filtering components
- Unique value list with numeric fields
- Range filters within reports
- Import improvements: Updating datasets with override or append
- Sharing a screenshot of a report
Zooming within the timeline has been a much sought after and asked for feature from our customers. Previously, if your data spanned a large range of time and you wanted to view a smaller subset, you had to filter this data out through the Data Set Manager or open the same date/time column mutliple times. Zooming provides a better approach that is more intuitive and user friendly.
In the example below we start with a dataset that spans from 2002 to 2009 ( figure 1.1 ). In figure 1.2 we have selected a date range from 2006 to 2007 which filters the data in the other pods within the report exactly as SpatialKey did before. Next comes the great new feature of SpatialKey! You can now see the “zoom” button at the bottom of the timeline after we have selected a range. By clicking “zoom” we zoom into the selected area and the timeline is redrawn with the new range. At the top of the timeline in figure 1.3 you will notice a bread crumb trail that shows the zoomed range and by clicking a range within the bread crumb trail we can go back to that specific range. In figure 1.4 we have zoomed down to 2007.
The timeline can continue to zoom to the lowest granularity of any date column you have. For example, if you have a date column that has a granularity of seconds you can zoom down the seconds level.
Figure 1.1 – full dataset range from 2002-2008
Figure 1.2 – select 2006-2007 to filter then click zoom
Figure 1.3 – zoomed range from figure 1.2 is drawn
Figure 1.4 – zooming down to 2007
It is common for numeric data to have a concentration of values in a small range and a small number of values that are outliers well outside of that range. When visualized with a histogram this poses several problems; First, it is difficult to visualize the trend of the data since you end up with one large bar (see figure 2.1 below) and second it is impossible to filter a specific range since only one bar is selectable.
The zooming histogram works much like the zoomable timeline in that you first select a range to filter, then click the “zoom” button to zoom into the selected range. Figure 2.1 through 2.5 show an example that would have otherwise been very difficult to visualize and filter without the new zooming capabilities of the histogram.
|Figure 2.1||Figure 2.2|
|Figure 2.3||Figure 2.4|
The only option for visualization and filtering of numeric data within a report was within a histogram prior to this release. If you wanted to visualize and filter numeric data with a categorical filter you were forced to import the data as a text field. It does not make sense to store numeric data as a text field so we have added the ability to visualize and filter numeric data in a categorical filter pod.
Here is an example with sales data that does not have a signfiicant number of unique values. When we open up the sample sales data with the numeric histogram you can see there were around 800 transaction in the 1000 dollar range but when you open up the same sales data in a “Unique Value List” you see the exact distribution.
|Figure 3.1: Price viewed within a histogram||Figure 3.2: Price within the new “Unique Value List”|
With this release we have made several improvements to our filtering capabilities within reports. Many of the same features that were only available within the data set manager are now exposed within reports.
There are two new type of filtering pods, the filter only pod, and the within last X pod. The filter only pod is provided within text, date and numeric columns and allow you to filter the data based on a specific value or range of your choosing based on the data type. The last X pod is only available within date based columns.
Filter Only Pod
For a text based column you can filter either an exact value, or the first part of a string see figure 4.1.
Figure 4.1: Filter only pod with text field
With date fields you can either use a timeline to filter by a range or you can use the filter only pod to select a date and range of before, after or between. This allows you to be more exact when selecting a date range. You can use these in combination with timelines by limiting the date range first with the filter only pod which will affect which dates are shown in the timeline.
|Figure 5.1||Figure 5.2||Figure 5.3|
With numeric data you can use the histogram to filter by a range or you can use the filter only pod to filter the data in several ways; equals, not equals, greater than, greater than equal to, less than or less than equal to. The filter only pod gives you the flexibility to filter based on a specific value or range instead of just relying on the filtering features of the histogram.
In figure 6.1 I have used two filter only pods to limit the price range for home sales between $50,000 and $230,000 and added the histogram to visualize the range we have selected. The filter only pod allows us to be more precise in selecting our range since we can type in exact values as opposed to just selecting a range in the histogram.
In some cases you may want a report to always show the most recent data. For example you may want to look at the last 7 days of sales data or incident data each time you open a report. Prior to this release there was no way to accomplish this so you were forced to select a date range each time you opened a report. With this release, we have introduced the “Within X” filtering pod for date columns. Using this filter you have the ability to filter by last X (minutes, hours, days, months years). It is important to note that this filter works against the last date in the dataset, not from today’s date.
For example, if the “last” date in the dataset is 4/10/2009 and you apply a within the last X days filter, the filter will look for records X days from the 4/10/2009 date as it is the “last” date in the dataset.
Figure 7.1 below shows an example using this filter pod filtering on the last 7 days of data.
Updating datasets with override or append
If you have reports tied to existing datasets it doesn’t make sense to delete that dataset and start over again with a new one. In doing so all of the linked reports would no longer function and you would have to start from scratch. With this release we introduce the ability to overwrite or append to an existing dataset.
This option is available under the “Actions” button on the data set tab (Figure 8.1). After selecting “Append or Replace Records” for a data set you will be provided the option to “Append more data” or “Replace all data”. Append will add on to the existing data that you have in the data set and replace will delete your existing data with the new data you are uploading. With both of these options the dataset needs to be in the same exact format as the original file (number of columns, specific columns for geocoding if you are geocoding, types of data in each column). All options that were selected during the initial import will be used for the update.
Sharing static screenshots of your reports just got a lot easier. Under the “Options” button within a report you will find the ability to share a screenshot (figure 9.1). This feature provides three different options; Save as file, Email, and Upload to Flickr (figure 9.2). All of the options provide the ability to generate a screenshot as a JPG or PNG as well as the ability to resize the image to a maximum size.
Figure 9.2 – Options when sharing a screenshot of a report
After choosing the options show in figure 9.3 below simply click “Download the Screenshot” and you will be prompted with a “Save As” dialog within your browser to download the file.
When you choose to email a screenshot you can add multiple email addresses a custom message and choose the attachment format.
In order to upload to Flickr you will need to authorize the SpatialKey account to upload to your account. After SpatialKey is authorized you will not be required to authorize you account again as long as you are on the same machine and web browser.