I have great news: the South Carolina Historic Properties Record is now live and can be explored at http://schpr.sc.gov! After much cobbling together of code, I present: SCHPR! (Or Morgan's Monster!!!) Below is a snapshot of the home page featuring tutorials, a carousel of images, and an interactive map of the SC counties. As with any website, I will need to make changes based on user feedback for a better user experience. We have some quirks that need to be resolved (like image captions only sometimes appearing), but overall, I'm happy with the site's performance. If you have a chance to look it over, send me a note on what you think of it!
0 Comments
This is a very short post, but I just wanted to pass along a little bit of knowledge. I'm currently taking an online MySQL/PHP class through Midlands Technical College courtesy of my work. (Thanks SHPO/SCDAH!) It's a bit like going back and learning a foreign language after you've hobbled together some slang and "where's the bathroom, please?" In other words, extremely useful especially if we're trying to be sophisticated. And I'd like to write sophisticated code. Now, I've taken some time off from work to focus solely on getting this class done, and one of the first things I did? Download an AMPPS program for my little macbook pro. What's an AMPPS program you ask? Why only the easiest way to set up everything you need for a homegrown online database. Here's the tagline from ampps.com: AMPPS is a WAMP, MAMP and LAMP stack of Apache, MySQL, MongoDB, PHP, Perl & Python. Right, now I'm not sure that this is necessarily the best way to go for everyday business at the archives, but knowing that this software package was available when I was running the test environment for my database with CollectiveAccess would have have been incredibly helpful. Do you remember those endless days of my banging my head against the desk trying to install all of the necessary applications? I do! But no more! With AMPPS, downloading everything for use in a test environment is a breeze! So, if you are in need of a quick environment for trying out an online database instance or you're playing around with CollectiveAccess, visit http://www.ampps.com. I promise you won't regret it.
By the process of elimination of every other acronym we could fathom (including SARA and HARIS), we at the South Carolina SHPO finally gave the historic properties database its real name. We present to you, the South Carolina Historic Properties Record, or SCHPR (pronounced "Skipper.") Of course, "Skipper" brings to mind several different mental images: Barbie's kid sister, an exuberant Labrador retriever, and of course, the penguin from Madagascar. That last one is my favorite. I am currently in the process with our IT Department of moving our instance of CA and SCHPR from it's current server to two new servers. The first is for the core backend database instance (Providence); the second, the public facing interface (Pawtucket). Finally, media will rest on yet another server that regularly backs up to a safety net. All of this separation is for at least two reasons.
1. An increased performance speed and ability. Currently, our instance of CA is running in a Windows 7 UI. This setup is fine for a test of the system. But in the long run, it's not the most efficient platform for the database. We're moving the backend to live in a Windows Server environment instead. 2. Safety and security. In my mind, I can't begin to understand why anyone would want to try to hack into a historic properties database. But it's not as silly as it might seem. In fact, the point of breaking into a site like this would be to gain a foothold. So if our front end was directly connected to the backend, and the backend to the network, then anyone ramming into our database would simply find themselves with a nice place to rest and think about where to move laterally from there. So really, it's just a best practice to separate out the front and back ends. I'll be working on this new setup in November. At the moment, our plan is to have SCHPR launched in January 2016. At this point, we have 16870 objects with associated metadata in SCHPR; this includes historic properties, historic resource surveys, and national register listings. We have about 50 digitized items associated with historic properties also in SCHPR. Going forward, the goal is to finish uploading all of our born-digital metadata -- historic property information in Access -- and begin data entry from and digitization of legacy data -- historic property site cards in the archives. We'll also work on a batch media upload of digitized national register files. All that said, when SCHPR does go live in January, it will have plenty of material already present. We'll just be working on bulking it up even more. I'll keep you posted! Although the Historic Properties Database was technically ready for beta testers on April 1st, we weren't able to distribute it until this week. I'm glad to have it off my hands for a little while. There are a lot of things that need to be done with the project that don't involve messing with the front end which I'll be working on:
1. Adding metadata for National Register Listings 2. Prepping/standardizing the rest of the data in the Newberry Survey 3. Prepping/standardizing data from another county's survey 4. Adding unique data for Newberry County National Register properties 5. Attend a workshop planning conference call for the NCSHPO tech committee 6. Attend a PALMCOP board meeting to discuss an upcoming event 7. Take care of treasurer duties for PALMCOP 8. Attend and monitor South Carolina's Historic Preservation Conference Although the road to putting out my prototype (powered by CollectiveAccess2015) was bumpy, I think I can say it's been a productive venture. We've learned a lot from the process. Most importantly, though, we're now able to learn from some of our users' feedback. I plan on gathering input and going back to the drawing board on May 1st with whatever contributions I have from the beta-testers. I expect to namely update the look, depth, and usability of the site. I'm excited (nervous!) to hear back from our users! Several weeks ago, I had someone on Twitter ask me what I thought of CollectiveAccess. To be frank, at that point I didn't really know how I felt about CA. Now, however, I can give a detailed description of the what makes CA user friendly and likeable.
1. Let's state the obvious: IT'S FREE. That's a huge deal for a lot of cultural heritage institutions right now. There just isn't always money to dedicate to a new system. And if there is money, there might not be enough. Or the institution really needs to prioritize something else over your project. Either way, a free tool is always a plus. 2. It's open-source. This allows for a much greater level of customization than some out of the box tools. For example, I could create a metadata schema to build the system around and customize user interfaces. Another bonus to this is that users/developers share their code/documents. 3. CA allows for (perhaps even focuses on) item level description! This is a huge deal for me since I'm creating a database that needs representations for each historic property and its corresponding site card. 4. Data entry can be performed through the cataloguing back-end (Providence). This tool is user-friendly for any staff member with a little training. 5. Data is stored in the MySQL database the user sets up on which the Providence tool runs. This means that catalog records exist on your server and would be retrievable if CA failed or you decided to move on to another product. 6. Pawtucket (front-end, web-publishing tool) is extremely visual. 7. Users can register/login. This is important to me since I'll need to create the ability for surveyors to electronically submit their survey results to us in the future. 8. CA support staff/developers are extremely helpful! I've worked with two of them (Seth and Jonathan) and both responded to my requests in a timely manner and made themselves available on a regular basis. On top of all of that, the forum is a productive environment for troubleshooting and problem solving. Once I've presented the website to my colleagues in Historic Preservation, I might be able to tell you a little about the cons from their perspective if there are any. There's only one item on the "In Progress" list! Wow, that looks good. However, there's a lot of "to dos" in that one card:
1. Customize Pawtucket's basic application configuration 2. Customize a theme for Pawtucket 3. Customize Pawtucket pages 4. Customize Plugins I'm looking forward to this part of the project. It's a new challenge, and I'll be able to reacquaint myself with CSS files and PHP. I'll keep you updated on how things go! Just a friendly reminder for the new year. When you feel like you're running up against a brick wall (like I do with some tech), remember that failure is important to ultimate success. As proof, see my finally successful (and correct) installation of Collective Access with the right profile!
An installation profile for CollectiveAccess is an XML document which informs the software how to configure and setup the system. While CollectiveAccess provides several standard installation profiles for various metadata schemas (e.g.: Dublin Core, Darwin Core, CDWA 1.1 Lite, PBCore, EAD, etc.), these profiles are available for editing. This is important to many project developers since the needs of a project may be very specific. Even if a project relied on EAD, for example, the project may require alternative user interfaces to be created that are not inherent to the standard installation profile. The standard xml profiles are available in the installation folder of the CollectiveAccess package.
In the case of the Historic Properties Database, none of the standard installation profiles met our needs specifically. First, there's the metadata schema. I need a metadata schema that addresses each property as two records. Let me describe what I mean. Each site survey card is itself a record. So I need metadata that describes the PHYSICAL survey card. However, those survey cards are themselves metadata for a physical building/site/structure/monument/what have you. So I also need metadata that will describe the architectural structure. You can understand that this would be complicated in terms of location. In quite a few metadata schemas, the term "location" refers to the location of the object/material/record. So, easy enough, the record lives in the stacks of the SC Department and Archives, correct? Correct. On the other hand, I really need that location field to describe the location of the structure, for example: 123 Main Street, Random City, South Carolina. Archival metadata schemas are built around describing a record, not the subject of the record, for good reason. Unfortunately, that just won't work for a Historic Properties Database. So far, Europeana's CARARE Project Metadata is the closest I've seen to a perfect fit. So that's the first and most important reason that I need to create a unique installation package. I originally installed CollectiveAccess with the CDWA Lite Profile. Although I had hoped that works of art would be a close fit for architecture, the profile just didn't match up to my fields. As a result, I am currently in the process of creating an installation profile specific to the needs of the Historic Properties Database. My hope is that the profile could be distributed to other SHPOs that decide to use CollectiveAccess as their tool for an online database. Since the week of Thanksgiving, I have been in the process of installing CollectiveAccess on my sandbox remote environment. Before I could install actual CollectiveAccess though, I needed three other open source tools in order to support it: MySQL, Apache (or any webserver), and PHP. I'd already installed MySQL by the end of October, so there's one requirement done! Onto the next two! However, Apache HTTPD and PHP were not as easily downloaded and installed for me. I'll go ahead and admit I was pretty clueless. I knew that Apache HTTPD was a webserver, and I knew that PHP was a dynamic scripting language. I just wasn't sure to go about installing them. Well after one week of trying to install from the .zip files (which doesn't go well if you don't know what you're doing), I finally discovered the .msi files. .msi files are Microsoft Software Installers. Feeling pretty proud of myself, I launched into installing Collective Access...except nothing happened. Apache was running from what I could tell, but I really had no idea how it worked. As for PHP, I just assumed it was working. I mean, I ran the Microsoft installer, right? WRONG! CollectiveAccess install wouldn't run at least partially because PHP was not actually running. (I tested it by trying to run the PHP info script.) Incredibly frustrated yesterday, I uninstalled EVERYTHING, and then decided to start from scratch...again. (I'd uninstalled and reinstalled apache hundreds of times already in my mind.) This time when I searched online though, I searched by a specific version of PHP, and I stumbled upon ProgramOften's youtube tutorial for installing PHP 5.5. I watched both the tutorials below, and voila! I now have Apache HTTPD 2.4 and PHP 5.5.19 successfully running on my sandbox. Not only that, but I feel like I actually understand both of them a little better. As a result, I was able to successfully install CollectiveAccess. Thank goodness! Of course, I'm still on the uphill slope of a learning curve, but I'm back to enjoying the experience now that I understand the first steps! Believe it or not, I was at the height of my work ethic in my senior year of high school. My zealous intellectual nature continued through the first two years of my undergraduate degree, but somewhere around senior year of college I realized I was coasting. I was still getting excellent grades, but I really just didn't care any more. I think most students go through that stage: senioritis and all that. But to a certain extent I never really recovered and got back to my intense work ethic of high school days gone by. (Or my writing ethics for that matter! Did you see my ending in a preposition? EEK!)
Now, though, I'd have to say that Trello has helped me recover my pride in work ethic. While I could just as easily write due dates on my outlook or google calendars, Trello allows me to visualize the progress and completion of one project in its entirety. Above, you can see my lists for the first prototype of the Historic Properties Database. I can share my self-imposed deadlines with my supervisors (hello, mini-due dates from high school history class). I can attach deadlines, as well as add comments to tasks. For example, I can elaborate on why "Download Collective Access Software" is now a week late in its completion. Being able to visualize project management with Trello has not only allowed me to better manage my time and juggle multiple tasks, but it has also enabled clearer communication between myself and my supervisors about the progress of the database. Both of my supervisors are busy people. By inviting them to view my "Historic Properties Database" Trello board, they can easily keep up with my progress and the general timeline of the project without taking time out of their schedule to talk with me. By being in the loop, they understand the current status of the project and ask appropriate questions when they do have the time and want to meet in person. At the Best Practices Exchange in Montgomery two weeks ago, there was a session about communicating and advocating a project to the top executive in 15 minutes or less. I'd argue that Trello is another great tool for keeping stakeholders and executives in the know about a project's progression AFTER they've gone to bat for you/the project. So, Trello has helped me recover my high school work ethic and communicate my project timeline to my supervisors. One little tool, such potential! |
About Morgan-I graduated in May 2014 with my Masters of Science in Library Science from UNC-Chapel Hill's School of Information and Library Science. I currently work for the South Carolina State Historic Preservation Office. I've been with them since June of 2014. Archives
June 2017
Categories
All
|