There was no more waiting this week. I was able to download the data. I have had to put off this part of the project for the last two weeks, so I was excited to finally get this going.  Before starting the downloading process, the first thing I had to do was to make sure that I wrote down every step. This included even the smallest of details, such as checks boxes or OK messages. So I got some paper ready and was on my way. Below is the downloading procedure in detail. I will go over how I transferred the data from the HOBO devices to the computer, and then I will detail how I got that data into the Excel spreadsheets. I will describe most of the steps involved with downloading the data, but I will not be as detailed as the actual documentation.

The first thing I needed to do was to get the HOBO Shuttle. The HOBO Shuttle is a device that allows you to store data from multiple HOBO devices, all in one location. From there, you can just download all of the data by just connecting the HOBO Shuttle to the computer. Once I had the shuttle, I went to the devices and I plugged each device into the shuttle and downloaded all of the data. When you plug a device into the shuttle, the shuttle will flash orange when the data is offloading and then will flash green once it is completed. Now with the all of data from each device on the shuttle, I moved over to the computer with the BoxCar Pro software. Before I could plug the HOBO shuttle in and offload the data to the computer, I had to create a new folder for this month’s data. This was my first opportunity to use the new folder scheme I created a month ago. On the library’s network drive, I went into the 2008 folder inside the HOBO data folder and I created a new folder for this month. I made sure to name the folder in accordance with the naming scheme I put in place (02 - February 2008). Once the folder was created, I was able to move on.

After creating the folder, I connected the HOBO shuttle to the computer and opened up BoxCar Pro. From there, I ran several BoxCar Pro menu functions, such as Logger - Launch, which launches the shuttle and prepares it for offloading the data to the computer. Next, I selected Logger - Readout, which began reading the data off the HOBO shuttle. BoxCar Pro is capable of autosaving the offloaded data. As such, every time you try to download data, a prompt will appear asking you to either say yes or no to it autosaving. You can also have it not display the reminder again. I clicked yes as this allows the data to automatically be saved to the root HOBO data folder. From there, the files can be moved into the corresponding new month folder. The file it saves is BoxCar Pro’s file format, .dtf. These .dtf files can be opened at anytime, allowing you to see the data that was downloaded from the devices. From there, you can export it into a text file, an Excel spreadsheet, or just view it in the BoxCar software.

As I was downloading the data off the shuttle, I realized that I still had four devices that needed new descriptions. If you recall from Chapter 3, when I created the map for the library, I wanted to rename some devices so that they could match the name on the map. However, back then, I was unable to do so because the only way to change a device’s description is to download the data. Now that the devices were empty and the data was on the shuttle, I was able to rename these devices without hindering the newly downloaded data. To change those device’s descriptions, I had to plug the devices in one-by-one. I changed those four devices so that they all followed a similar naming scheme. With the device names in place, I had to go back and tweak the map a little bit so that it matched the new names. I wrote the new names on labels and put them on their respective devices. I then set them all to start recording data at the same time and then I returned all the devices to their locations.

With the device names taken care of, I moved on and completed the process of pulling the data off the HOBO shuttle. When doing this, I found something interesting. The data still remains on the shuttle even after you download it. The shuttle seems to keep a backup of the monthly data. This data stays on the shuttle until new data is downloaded the following month. This is a smart failsafe. This way if something were to happen to the files, you could still just download them again from the shuttle. Next month, when I offload the data from the devices to the shuttle, the previous month’s data will be replaced.

Now the data was downloaded, and auto saved in the .dtf files, it was time to export it to the Excel spreadsheets. While doing this part, I was following along with the procedures written by the last group that attempted this project. They had the user export the file to a text file first and then import that text file into Excel. I have absolutely no idea why they decided to do this. BoxCar Pro allows you to export the data directly to an Excel spreadsheet, thus there should be no need to use a text file. I choose to veer away from their instructions and I decided to export the data directly to Excel spreadsheets, just to see if I could notice any difference. From what I can see, the only thing that was different was that when it exports it to an Excel spreadsheet, it adds in the devices name and some other information into a column. All that you would need to do would be to delete the column and you would have the necessary data. This is much easier than what the last group required the user to do. They had the user export the data to a text file, go into Excel, use the import feature, follow a few steps and then save it as an Excel spreadsheet. With my procedures, you export the data to a spreadsheet, delete a column, name the spreadsheet, click save, and you are done.

With that figured out, I had all the data saved into six spreadsheets, one for each device. These files were saved into the new month folder, 02 - February 2008. From there, I had to go back and move those .dtf files that were auto saved to the main directory. I copied the .dtf files and moved them over to the February folder. With those moved over, there were six spreadsheets and six .dtf files. This completed the downloading procedures. I unhooked the HOBO shuttle and returned it to its location. Finally, after two weeks, the data download was complete.

I am confident that new downloading procedures I created are more efficient, quicker and will be easier to follow than what was in place before. My procedures will also save some space and reduce clutter as they do not require the user to save .txt files as well as .xls files. After I completed the procedures, I began putting the instructions I wrote down into an electronic format. I took eight screen shots during my run through the process. These screen shots will help point out the more complicated steps, thus making it a bit easier on the user. With the procedures documented and in an electronic format, I was completed with this part of the project. The next thing I need to do is have Pam review the documentation to make sure it is sufficient. Once reviewed, I will make any necessary modifications and I will come up with a final draft that will be saved in electronic format and will be printed out. You can view the completed documentation by clicking here.

In the end, this was a great week. I completed a large chuck of the project and I feel proud of the work I have completed. The documentation for the downloading procedures is detailed, but is still easy to follow. With the documentation in place, I am confident that next week I will be able to move on to other parts of the project. At this point, all that remains is for me to create an algorithm that will take out the out of range points. Then I have to document those steps as well. So next week will mostly be a brainstorming post. I will discuss some of the ideas I have so far in regards to the algorithm. Hopefully I will have a good algorithm ready by the end of next week so that the following week, I can take real data and complete the procedure that will successfully pull out the out of range points and prepare them in a printable format. With that said, until next week, thanks for reading.

Since beginning this project, I have been able to hit the ground running. Everything up to this point has pretty much gone according to plan. Well, that was until this week. My original plan for this week was to get the data downloading procedures completed and documented. However, I had assignments due this week in my other classes that were more pressing. These assignments forced me to push my plans back until next week. On top of that, it has not been a full month since previously downloading the data, another issue that delayed my plans. If I wait until next week, it will have been a month since previously downloading the data. As such, much to my chagrin, I decided to hold off on downloading the data for yet another week.

Since I could not complete the downloading procedures this week, I was pretty flustered. I really wanted to complete them this week and move on to something else. However, as they say, patience is a virtue. It just so happens to be that patience is not entirely my best virtue. This is something I am going to have to learn how to handle. There will be many times in the future where other factors will either delay or alter plans for a project. I will have to learn that I cannot complete everything all at once and that projects such as these are dedicated, long-term tasks. I feel this will come with more experience. I simply have not had enough projects under my belt to realize the pitfalls, trials, and tribulations that come with all projects. I have come to learn that there is more to this project than just trying to learn the IT skills. In fact, the IT skills are the least important thing in my mind. The lessons I am learning about project management will be invaluable to me. This includes the skills I am gaining in writing project plans. On top of the project, we have an assignment in the class that requires us to write up an official project plan. So, by the end of this project, I will have not only learned how to work with IT projects, I will have also learned how to write up project plans, a skill that will definitely assist me in any future project.

With that said, the entire week was not a wash. I did have some free time, so I began taking a look at some of the possible options I could use to separate the out of range data points. There are many options to choose from and I as began looking at them, I became slightly overwhelmed. I realized that this could take me a bit longer than I anticipated. There are a lot of possible ways to take the two columns (relative humidity and temperature) and pull out the out of range data points (RH greater than 55 and temperature above 70). As I looked at them, I realized that I am not entirely sure what the best approach is to this problem. However, I am confident that I have the ability to come up with a working procedure.

First, I realized that the downloading procedures that I will create next week will affect how I start the procedure for the out of range data points. When creating the downloading procedures, I have to get the data into a format that will allow me to create the procedure for extracting the out of range data points. Thankfully, I had already come up with a few ideas when I downloaded the data for the first time last month. Last time, I followed the downloading procedures that the previous group had come up with. After completing those downloading procedures you end up with four Excel spreadsheets, one for each HOBO device (this was before I added two more devices). I realized at that time that it was going to be far easier to handle all of the data if those four spreadsheets were combined into one. If you left the data from the devices in four separate spreadsheets, you would have to run the same procedures four times, which, needless to say, is quite inefficient.

So I went ahead and came up with a rough idea of how to get all of the data into one spreadsheet. First, you still need to follow the same downloading procedures that will create the individual spreadsheets for each device. From there, you would copy all of the data from each of the individual spreadsheets and you would paste it into a master spreadsheet. Then you would run the data procedures that would separate the out of range data points for all of the devices at one time, rather than having to run it for each device separately. Once you were done with the data in the master spreadsheet, you could run a macro in Excel that would clear the data, making the master spreadsheet ready for the next month.

I feel I have a pretty good strategy for getting all of the data together into one file. Once the data is in one file, I have to figure out what to do with it. As mentioned, there are several ways to tackle this. You could sort all of the data by temperature first and then just take out the ones that are above 70. Then you could do the same for relative humidity. This would get you the outlying points, but that does not let you really track or compare any data. This would only get you the number of out of range points. So, I began looking at some other ideas. However, I was unable to really come up with any concrete. My plan is to separate the out of range points based on the date and time they were recorded and then map those points to the corresponding device. This would allow you to see which device, on a certain date and time, registered an out of range point. However, I have not been able to determine the best approach for extracting the out of range points. Some of my ideas include using pivot tables, filters, or macros. However I choose to do this, I am confident that I can find a procedure that will pull out the needed data and get it organized into one printable sheet.

Even though this week was quiet and I did not get the chance to work very much on the project, I still had an opportunity to learn. I learned that you must sometimes be patient and that you have to work around problems. There will most certainly be unforeseen factors that will delay my future projects. I must learn to find ways to advance the project while remaining patient for the problem to become resolved. So although it was a somewhat frustrating week, there is no doubt that it was a great learning experience. Next week, I will describe the process of downloading the data from the devices and I will describe the new documentation I will be writing. Until next week, thanks for reading.

This week was relatively light. With the folder scheme in place, and with the two new devices running, it was time to move on to other activities. As I mentioned last week, the next task I wanted to accomplish was to create a map that pinpointed the locations for each device. The map Pam had already created, the one I alluded to last week, served as a good starting point. I opened up the map, printed out a copy, and was ready to jot down the locations of the devices. However, as I began going through the library, I noticed that the map was very rudimentary. It was missing shelves, tables, and it neglected entire sections of the library. On top of that, the map was also missing sections where some of the devices were located. I needed to have these sections on the map so that I could properly mark all of the device locations. The original map was intended to be simple, but due to my detailed nature, I felt the need to correct it.

In my effort to create a more complete map, I went around the library and began jotting down corrections on the printed map. I first added a section for the offices which were not on the original map. I needed these offices because they are where two of the devices are located. I also added in one of the computer catalog rolling carts to the map as one of the devices is located on the cart. From there, I decided I would add the missing shelves and the missing tables. I also corrected the front desk and I added in a few other missing rooms to complete the map. After completing my edits to the printed copy, I had to go and edit the map on the computer to include these changes. The original map I printed was part of a Word document, so I had to copy out the inserted picture and edit that. Using Photoshop, I added in the offices, the catalog cart, and the other missing parts to complete the electronic copy. You can click here to see the version of the map that does not yet have the device locations.

Now that the map was more accurate, I was ready to place the device locations on the map. However, before I could do that, I wanted to make sure the names of the devices matched what I was going to place on the map. Last week, I explained that each device has a description, or name, that is based on its location in the library. Each name is coded into the device using the BoxCar software. So before I put the locations on the map, I wanted to make sure all the descriptions were accurate and that they actually matched the location of the device. I hooked up the devices to the computer and opened them all up in BoxCar. I found that four of the devices had names that needed to be altered so that they would properly match their locations. For example, one of the devices was named Bound Periodicals. This device is actually located on the shelf for Current Periodicals. Now this may seem like a minor detail, but when you download the data, it saves the data file based on the device’s name. In this case, it would save it as Bound Periodicals.dtf (.dtf is BoxCar’s file extension). With the name the way it was, you would have been in a situation where the device was saying one thing and the map was saying something else. Although this would probably not hinder the procedures that much, there is need for this unnecessary confusion.

With that aside, I was prepared to go ahead and change the names on the necessary devices. Before I could do that though, I discovered that I could not edit the names at this time. The only way to change the name of a device is to offload its data. Since it had not been a full month since previously downloading the data, I could not rename the devices at this time. Speaking of downloading the data, I know mentioned in my last post that I might do that this week, but I decided against it. It was not as close to a month as I originally thought. As such, I have to wait until next week to download the data and rewrite the procedures. So next week, when I download the data and rewrite the procedures, I will make sure to change the names on the devices. I was hoping I could change the names this week so that the devices and the map matched immediately, but, as in most projects, sometimes things do not always go your way.

As much as I wanted to change the descriptions, I knew it was not an option. So I went ahead and plotted the device locations on the map and labeled them with the names that I intend to put on the devices next week when I download the data. On the map, I marked the location of each device, placing a marker labeled L1 through L6 at the device’s location. As I alluded to before, L1 simply represents the number of the device. With the locations correctly mapped, I created a legend. In this legend, I labeled the description and the unique ID number of each device. Last week, I discovered that each device has its own unique ID number. These ID numbers are separate from the L1 numbers the library gives each device. So, I went ahead and wrote down the ID number as well, this way if the labels that are on the devices were to fall off, they could still be identified using the legend on the map. With the legend now in place, anyone with access to the map would know that device L2 was located in the Professional section. So even if the person who is downloading the data has no idea where any of the devices are in the library, they should still be able to find them using the map.

With the legend added and the locations marked off, the map was complete. I saved the image that I edited in Photoshop and I replaced the old image in the Word document with the new, corrected image. I saved two versions of the map. I saved one version that just had the corrections. I then saved another version which included the corrections as well as the legend and the locations of each device. With two separate versions, the library would have a copy of the map without device locations, allowing them to use or edit a clear map as they saw fit. With that said, you can click here to view the final map with the HOBO device locations.

That wraps up this week. It was a relatively quiet week and I was hoping to have accomplished more. However, I realized that sometimes certain parts of the project are just out of your control and that you must go with the flow. Late next week, I hope to download the data, rewrite the procedures, and change the necessary device descriptions. Until then, I may begin looking at some of the test data that I have so that I can begin my work on determining the best way to make the data useful. Until next week, thank you for reading.

This week, I hit the ground running. I was excited to finally get the project going and I was eager to begin my work. As I mentioned in my last post, I decided to begin the project by first reconstructing the folder organization. First, let me breakdown the previous folder scheme. In the old scheme, every time a user would download the data from the HOBO devices, they would store the files into a folder named after the month and year. For example, if I were to download new data today using the old scheme, I would save the files into a folder called Feb 08 hobo data inside the HOBO Data main folder. The library has been downloading data for roughly 3 years now and this has led to a lot of folders. With the current system, trying to find Oct 06 data would take some time as you would have to track it down among 30 other folders. This scheme seemed very unorganized and somewhat haphazard to me. Although it does not really hinder the user from using the newest data, it would be very hard to pinpoint and find older data should it be required. Here is a screenshot of the old scheme to give you an idea of how it was organized. Now that I have discussed the old scheme, I can talk about the new folder scheme I developed this week.

When thinking about what to do for a new folder scheme, there was really only one idea that came to mind. The new scheme had to first organize the data by year. The library has collected data from 2005 to 2008. So in the main folder, I had to create four folders, one for each year of collected data. Next, inside those folders, I created twelve folders for each month of the year (I only created two folders for 2008). I first started by creating folders named after each month. For example, in 2005, I created folders called January, February, March, etc. As I finished creating the month folders for the 2005 folder, I realized that this was not going to work very well because Windows would automatically sort the folders alphabetically.

To test my theory, I went back to the main directory and then I went back into the 2005 folder. Upon doing so, I found that my assumption was correct. Windows had sorted them alphabetically, just as I anticipated. This meant that April was the first folder at the top. Having the months sorted alphabetically is completely different than the sequential order that we are used to seeing. Since March is the third month of the year, I naturally found myself looking for it at the top of the list, where alphabetically, it is actually in the middle. Although this problem is trivial, I did not see the point to causing any unnecessary confusion. So to correct this problem, I added the number of the month to each folder. So instead of January, I named it 1 - January and so on. So now every time you go into the folder, the months are in the order that we are more accustomed to seeing.

With the month folders now in their normal order, I decided I would make another change. I was concerned that when downloading the data, you may not be able to see which year you were working with. Without being able to see the year, someone could have trouble trying to figure out which January they were in. Granted, this is easily fixed by just going back one directory, but I felt that it would be better to just to complete each folder name and maximize its accessibility. So, I decided to add the year to the end of each month folder. For example, the names now looked like 1 - January 2006, 2 - February 2006, and so on. With those additions, I was finally satisfied with the naming system for the month folders. To get a better idea of the final naming scheme, here is a screenshot of the new folder scheme.

With the folders created, and the naming scheme in place, I was ready to move the old data into the new folders. The first thing I made sure to do was to backup the data in the older folder scheme. I created a backup folder and then I moved all of the old folders to the backup folder, leaving them exactly as they were. This way, should the need arise, the older data can be accessed using the older scheme. From there, I then began moving the data from the old folders to the new folders, making sure to compare the new folder with the old folder to make sure they matched. So for example, I copied Jan 06 hobo data into 2006/1 - January 2006/. After copying over all of the data, I was pretty satisfied with my new folder scheme. I feel that this new scheme is far more organized and will make the data more accessible and easy to find.

That was not all I did this week. I also began my work on trying to fix the two non working HOBO devices. I met with Pam this week to discuss the folder scheme. I also asked her what was wrong with the two non working devices. She informed me that she had replaced the batteries on the two devices, but doing so did not fix the problem. Every time she tried to connect a single HOBO device to the HOBO Shuttle (a device that allows you to collect all the data from multiple HOBOs into one place) it would give her a communication failure error. So I began looking into this problem. First, I tried connecting a HOBO device to the shuttle myself. The results were identical as I also received the same error. I then went and tried to connect the HOBO device directly to the serial port on the computer. I used the BoxCar software to see if it could pick anything up. This did not work either as the software could not even detect the device. So I began researching some solutions. However, while doing this, Pam decided that I should forgo my research. She told me to just replace the two non working HOBO devices with two working devices that she received from the College’s iTec services.

With these two new devices, the library was back to six total devices. First, before these new devices could be used, I had to clear the existing data off of them. The only way to do this is to download the data off the device. Once you offload the data from a HOBO device, whether it is to a computer or to the HOBO Shuttle, the device is cleared and is ready for new data. So to clear these two devices, I hooked each one up to the computer and downloaded their data. I saved this data to the backup folder just in case anyone ever needed it. Once the data was off the devices, I also needed to change some of their configuration options. First, I had to make sure that they were recording temperature and relative humidity data every 30 minutes. Each HOBO device can record data points at a variety of intervals. The lowest option is every five seconds, but you could also record points once a day, or once a week even. Pam had set the other four devices to record data every 30 minutes, so I set the two new devices to the same interval. Next, I had to also reset the device’s description (this is device’s name in essence) so that it indicated where in the library the device was located. For example, I renamed one of the descriptions to L5 (E) - Circulation. This description explains where the device is located in the library. L5 represents the number of the device. This is the 5th HOBO device. The E represents the east side of the library. Finally, the Circulation part signifies that the device is located in the Circulation section of the library.

When renaming the descriptions for these devices I came up with an idea. Since these devices are spread throughout the library, someone who did not know where they were would have no idea where to look. Most of the devices are mounted underneath shelves and can be hard to find. My idea was to create a map of the library that would pinpoint the locations of the devices. This map would help anyone find the devices, regardless of who was asked to download the data. Thankfully, Pam already has a map of the library so all I need to do is just draw the locations of the HOBO devices on the map.

Now that I have the new folder scheme in place, and I have six devices working, I think I am ready for the next part of the project. I am excited about what I have done so far and I feel have made good progress. I feel confident that what I have done is the first step towards helping the library develop a sound systematic approach to downloading the data. Next week, I will first draw the devices on the library map. Then I will work on recreating the downloading procedures. To do this, I will go through the procedures myself, documenting every step, including all messages, boxes, clicks, and every possible option so that the procedures can be as easy and accurate as possible. This documentation should give anyone, regardless of computer knowledge, the ability to download the data. So, until next time, thanks for reading.