APS7: Innovate and Collaborate: Using WordPress 3 to Build a Research Database

Chris Gabel 
Webmaster, University of Kentucky - College of Engineering


The audio for this podcast can be downloaded at http://2011.highedweb.org/presentations/APS7.mp3


This is one in a series of podcasts from the HighEdWeb Conference in Austin 2011.

Chris Gabel: Like he said, my name is Chris Gabel. I'm from the University of Kentucky, specifically the College of Engineering. Just a little bit of a background for you about the project. I came back from paternity leave in February. And they had a meeting that told me they're changing the way that they want to do a research pieces for this year.

A lot of schools put out research pieces. They can range from, you know, like these 20 page overviews in your department to, you know, 300 pages this is everything we're doing in research. Well they decided they did not want to do either one of those options. They want to do a print out that just, it was like tri fold piece and this is actually what it looks like.

This is the printed piece and they want to stress the importance of the collaboration between departments. So that's where the title for my presentation actually comes from. The title of this printed piece was "Innovate and Collaborate".

 

01:00

I guess the title is a little misleading because I feel like a code for this is not really innovative. And, you know, some of the code I used was actually ripped off from somebody else. I give them credit when I get to it but some of you won't call that collaborative or not but that's really what they want to focus on.

My point, I'm hoping to get you guys, this was easy. I came back from February. They already had the print piece together. Alright we're going to mail this in two weeks. We need the website component ready to go. I was like OK. Well, you know, let's get a little background here. What do we need to do?

Before I get too much into that, this is what we're kind of going to go over. I want to go over a little bit of background of national web project. You know, what is it need to do? Who's the audience for it? We evaluated couple solutions before we went with WordPress. A little bit about the development. Some of the lessons I learned and then how are we going to move forward next year because they want to change just about, not everything.

 

02:00

They do want to change the lay out and they want to add additional features this coming year. It's not an HTML5 . We've got a new dean. It's supposed to start in July. They actually have hired him to target date this July first. So they're wanting to relaunch this along with the rest of all our websites, July first. Does that coincide with his hiring?

So the first thing, you know, that everybody ask when you start a project is, what are we going to have and look like? And as developers or web designers, it's probably a bad place to start. They came in. Everybody knows what they want to look like. Nobody really has a goal of who are audiences. What is it going to do?

Luckily, they did already have a design for the print piece. So, it has saved me a lot of time and say I don't have to worry about really doing lay outs or color pallets or a general idea of how the site was going to go. We have a great graphic designer. She loves infographics. That's like her big thing.

 

02:59

If we can give her infographic which is this, she goes crazy. They already had, you know, the basic concept together. So, I don't have to worry about anything like that. So, what I really want from them, you know, who's your audience. And most of the rest of our website, the target audience is predefined.

You know, we prospective undergraduate students, that's where they've been wanting to go with her strategic plan for the last couple of years. This going to change this year. The research piece was specifically for graduate students. Looking our website numbers, we don't get a whole lot of traffic from undergraduate students even though that's our target audience. All the admission stuff is handled by main campus. So that's where the students go.

Financial aid handled through main campus. We have additional scholarships and stuff we offer for undergraduate students but you still have to fill an application for main campus. So when we look at new analytics numbers, all of our traffic is going to the graduate programs. And once they're on the graduate programs, they clicking on specific faculty and we assume it's because they're looking for the type of research they have.

 

04:02

So the audience really is primarily graduate students but also other researchers because they're in this big push right now. We don't have the money like maybe Ohio State or MIT or Stanford in research dollars. So, that's what's they're trying to do to separate us from these other research institutions. You know, you're not going to work with just mechanical engineering You're not going to work with just electrical engineering. You know, you're going to work with psychology, you're going to work with the College of Education and so they're trying to attract researchers from other universities as well.

For some reason, they have a really big partnership approval. You can't say anything good about Google on UK campus unless you're talking about research.

[Cross-talk]

Well, it's the probably true in here too. I don't really know what the view is. You know, they're not in the same conference but I'm not from Kentucky so I'm still trying to learn. And then also corporations. We have faculty that are they're labs based and the research. Although it may not be endowed by corporation, it is actively funded on monthly basis.

 

05:03

So that's who they really want to reach. And my questions were always, you know, what do you want to do? What do I actually need to have it function? And so, what they ask me to do was, you know, we want you to display every active funded project that's listed with the offices' sponsored projects. So that was a kind of a relief to me because that means I'm going to limit it to this year.

And eventually, they said they want to go back five years and put all those in. I've been in really what I was getting into at that point. I pulled the report from our offices' sponsored projects and there were 450 from the last year. They were funded. And they wanted all of that according to the website and then I'm not sure if you guys know what NSF, the National Science Foundation, all engineering faculty should have one of these resumes up. So they wanted this style bio up.

 

06:00

For every researcher, they wanted to have be able to tag those researchers with keywords for the research, the department they're in, the centers they work in. They wanted their education, their current publications and that's what they want to be able to search by all that information which doesn't seem like a big problem.

It kind of freaks me out a little bit because I'll show you what we're working with here. And by the way, I have all these slides already posted at christophergable.com/heweb11. I've got enough SPDF's and I pasted some codes if it's in there in case you wanted any of that and there's some links to some relative solutions.

But this is actually our research site when I stared and what I had to work with when I came back in February. I got hired in 2009. They hadn't have a web developer or webmasters since 2005. And the WordPress or the Dreamweaver templates were built in 2003.

 

06:58

So it's not pulling the main image that was here but everything else is pretty much accurate. So Dreamweaver templates they wanted all the search functionality and they wanted it in two weeks. And I was like, this is not going to happen. You know, we've been test driving WordPress but we didn't have anything up with the kind of capability that they wanted that I could do it quickly. So, we did start to evaluate some of the solutions.

Our first option was, of course, to stick with the Dreamweaver templates. It was not my choice for either our main website or this research website. I was spending a lot trying to fix things. I think the only reason that we hadn't moved to something already was because our IT department is justifiably concerned about security. Universities seem to be a great place for people that just sit around and hacking any network.

 

07:57

And it seemed to happen quite a bit of our homegrown solution. That's actually why we ended up not going with HTML. Despite the fact they has going to pay me forever to get this kind of search functionality built from scratch in HP, our website was hacked over Martin Luther King weekend and we had a content writer who had started a Twitter feed and a Facebook account, quit and did not tell me he had started those and had linked into the RSS feeds.

So when it got hacked, everything went out to Twitter and Facebook as well. So, the IT department was like, okay. You've been pushing for this for a while. We're going to move to a content management system. And the computer science department was using Drupal and the university as a whole had pushed for everybody to use Drupal.

At that time that want to put this up, there are Drupal 6. The university was not offering any support as far as software or hardware set ups for Drupal.

09:03

And it's actually this conference last year in Cincinnati, I've sat down for breakfast, I was talking to people about content management systems and it just so happen they have all switched to Drupal recently and they weren't developers. They were just a content managers and they talk about how they hated the UI. Didn't understand the many structure.

It took them forever to do their jobs. And we actually have a bunch of students that were also one of the university servers using this that said, we just can't use it. We also look at WordPress. We have been using WordPress MU to power our student websites for almost a year. I was pretty comfortable with it. We have a pretty solid plug-in base and a few themes are in together.

So WordPress 3 also came out that gave a lot more CMS functionality that really made it a good option for us and like I said, we're already running it. So we ended up going to WordPress specifically because a lot of the functionality that they wanted has already built in the WordPress.

 

10:00

Categories, the administrative back in that we could customize so people can only have access to certain types of custom fields. There's already built in archiving and searching. And with WordPress 3, we were able to do custom taxonomies and custom post types. And if you're not familiar with, WordPress, it comes with five default post types. But with a little bit of code, you can actually create new post types.

So we decided we needed to have three custom post types. One was just the slider on the homepage and the slider just mimics that infographic. You can hover over the button. It pops up the name of the featured article that they have chosen. This is all just jQuery and its a coda slider with coda bubble box. So we just created a custom content type for the slider. All it has is the title and then a metabox where you link the ID of whatever this project is and it's actually choose what background you want.

11:14

Like you choose between one of the six slides that we have. Click on wherever you want your bubble to be, you hit update and automatically adds that bubble in on the homepage. So hypothetically, they wanted to switch these out occasionally and choose different featured projects.

Another problem we had was they can't decide who needs to be featured because they don't want to hurt people's feelings. So when we got into the development cycle, like I said, HTML was simple. They already had a basic lay out that they wanted to go with. They have a color pallet. I knew right away this is what they want to do. This is what it supposed to look like. So I didn't have to spend a whole lot of time there.

 

11:58

I used the 960 framework which were also probably be changing. I guess they have a fluid version out now but we need something that's a little more responsive. And then we already have snippets from our existing templates that I was able to use for the WordPress specific content. So they came out was pretty quick.

Other custom post types, we had researchers and then the projects. That's the big thing about this database. They wanted to be able to list all these projects. When the project page comes up, they wanted to have tags from what the research areas where they wanted to tag the primary investigator and then the co-primary investigator.

Then they able to link back the researcher bios which again have NSS style bio and they also listed their active or Orcad projects. And a little bit of the code for this. It's short based its 17 lines to register this post type and all.

 

13:00

And there's a lot more features that you can actually do when you register a post type. Like this capability type, I just put page in. All that means is that when the user is added, if they don't have the same privileges as a user, like if I say I have a new author, we'll have the same privileges in the author that has the capability to edit pages or posts.

So we could do something a little more advanced that would tie into plug in like role manager where you can say, okay we can find a custom user and then say, well they can edit their own projects and their own bio but they can't edit somebody else's. We didn't need that functionality because I'm the only person that has to update this. So we didn't really need that advanced functionality. And if you're not interested in writing this kind of code, there is a link and it's themergency.com. It is actually a generator for this.

 

14:00

And I have that link, the hereweb11 page. If you go to this page, it already has the form you fill out the fields. You hit next, it has got four pages of forms and it generates the code for you. One of the drawbacks of WordPress that they brought up in the cage match earlier is that you can't define custom post types in the UI. But if you're not a coder. This is pretty useful. I haven't used it myself but I send it over to a couple of people that were ISP developer and didn't really want to learn the PHP when they started working on WordPress.

And then, there's of course the custom taxonomies. You know, if you don't know what a taxonomy is, it's basically how you can categorize your posts, your pages. In our case, both our researchers and our projects.

 

15:00

It’s even simpler to implement, we’ve got one line of code to register taxonomy. And we basically just say, this is the name that we want to have it when it’s displayed on the back side, you know, we want to have its own tab that says Researchers and if you look at the back, the back end of this, under Projects there is now a text on the list for Research Areas, Departments, Primary Investigators and Co-Primary Prod Investigators, then our Status. 

Originally, they wanted to be able to have statuses, active or complete. They since revised their decision and wanted it to automatically follow out or start based on the start date or end date of the budget cycle, which is actually different than the research cycle, but they wanted these custom-made tabs or boxes for that type of information, so that’s what we added here.

They have Sponsor Name for each project, the Award Amount, the Beginning Date, Ending Date for the funding. And when I said this wasn’t exactly collaborative, the only collaboration I had we have an office-sponsored projects they have a home-grown ASPX system and they can spit out an Excel sheet with everything that I needed.

And so, there’s a plugin for WordPress called CSV Importer and all you have to do is basically format the headers the right way and then it’ll import everything into your database.

16:32

So I had 350-400 projects that I was able to import, in I don’t know, I guess it was 30 seconds or so? The one drawback to that plug-in is that it does not recognize duplicate entries so I had to update it a little bit so before, when entered it into the database will actually search for that title.

And CSV still had some issues because the apostrophes are formatted differently, as far as the character than the phpmyadmin or the mysecretdatabase. And if you want to take a look at this template, this is what a prick basically does, it spits out the title of the project, which is ridiculously long.

And once again, working with faculty members, you don’t want to touch a title, you don’t want to touch an abstract, just put it in as is. If the Associate Dean of the research company says, this is a ridiculously long title, you need to change it then we can change it, but until then we can’t do anything because they’re really touchy about their stuff.

But then it comes, it does spit out all the research areas, the departments they’re in, the primary investigator in this case and then the co-PI. And right now, this project’s actually coin just the content, because this is an older project that was input in to our database. Yes?

17:53

Audience: The primary investigator that goes to...

Chris Gabel: That’s a good question. On this project, it’s going to go, because it’s a tag, when I click on it, it’s going to go to a list of the other projects that he’s associated with. And the reason it does that is he does not have a bio.

Audience: So they are supposed to do their bios.

Chris Gabel: Yes. And that’s what the goal is supposed to be. This guy is a new faculty member in biomedical engineering. They recently moved from a graduate school to the College of Engineering so I did not have access in their NSF files.

So by default, if somebody has a project that’s been pulled from the offices sponsored projects and I import it to our WordPress site, if they don’t have a bio it’ll just list the projects they’re currently working on.

You can still search by that name, and it’ll come up in our search box because it is a tag and we’re using the search everything plugin, so that you can search custom taxonomies. There were some issues with it, with WordPress 3 at first, I believe it has been worked out. I had an update on that plug-in, I guess a couple of months ago.

Audience: But sometimes it's hard to do bios.

Chris Gabel: I did. That’s what I was going to get in to next. We talked about the templates together. Basically, with WordPress, they have something we call hooks that does exactly what it says – it hooks directly in to the Word Press API, and it adjusts whatever you want.

So there’re two basic types of hooks, filters and actions. In this case it’s a filter that we used. So the basic we say is when you click on those tabs, those terms, it looks for a specific type of term. In this case, I limited it to the primary investigator and the co-primary investigators.

So what it’ll do a separate query on that term name and then it’ll look for researchers. It’s based on slugs too, so the slug for the primary investigator and the co-primary investigator has to be the same as the slug for the researchers.

It’s a little walkie but WordPress doesn’t have taxonomy metadata right now. By default, it does not allow you to add specific taxonomies to post, but this is why I said I ripped off somebody’s code.

You may have heard of this guy, or at least his book, “Professional WordPress Plugin Development”. I don’t know this guy but I bought the book and it was fantastic. And he had this post when he talked about linking terms together and once again, it’s a piece of code, it’s a little longer, it’s 35 lines and there are two functions, first the added filter and the term link is the WordPress’ specific filter that is going to tie into and then my term type is the name of the function that it’s going to call instead of the default term link function.

And what it does is it takes that slug and uses the second function, it looks for, based on the converts that slug in to a post ID and if there is no post ID, it will return false and it’ll just go to the regular term list. If it does return true, it’ll adjust the URL for that term. 

21:19

Chris Gabel: And so, Dr. Leo, if you look at his act of projects, when you click on his name, it does go back to his bio. The drawback of this system is that if you have a project with one PI and then six co-PIs which is sometimes common if they include the graduate students on their report, you’re doing separate database calls for every term on that page which is an additional load, because just doing the term list it doesn’t have a database call as a built-in but this is a separate database call for each term in that list.

And then also, another thing about this template, the act of projects, when you tag a project with a primary investigator or a co-primary investigator, those are going to be together as of one. 

If you put in the query, I want to know all the projects that Dr. Leo is part of, it’ll query primary investigators and co-primary investigators together so we ended up half into once again use multiple database calls and we end up having four on each page – because it looks for the projects that he’s a primary investigator, a co-primary investigator, the archive projects where he’s a primary investigator and then the archive projects where he’s a co-primary investigator.

So, the system’s not ideal. It’s one of the drawbacks because there’s a lot of database calls on at least the researcher page. The project’s page might have some additional ones but the researcher’s page has a lot of database calls.

We haven’t seen any real slowdown but our database is relatively new so I just got another round of updated projects and we’re hiring new faculty so as we get more projects and more researchers in there, I’m curious to see if there’s going to be a noticeable slowdown.

23:30

And the idea for this actually came, Justin had a post where he created a movie database and he linked the movie with its description of the movie to a director, or a writer or a producer; and I figured, why can’t we do the same thing which is our research pages instead? 

And there are, I checked this article a couple of days ago, and there are actually some other people that have posted about the same problem. There’s a drawback when you’re doing a multiple database calls to get those terms to redirect to the correct page. 

So they do have a plugin now that you can add metadata to taxonomies and there is somebody else that actually had just a function made and written it was very similar that was supposed to take some of the load off and it is in the track as a future request to add taxonomy metadata.

And some of the plugins that we’re using, C4s is not currently updated on the WordPress repository but it’s actively updated by the UserBase, for some reason it is not added to the repository anymore. We use it primarily because I don’t know when the faculty members are updating their research, when they’re updating their publications.

And so I’ve sent them out, everybody in the faculty will serve as I sent out this link to C4s so they can say this is my name, this is what my title has changed to, these are updated publications and they can request me to update certain parts of their site. 

And I’ve explained to you, we use the CDS Importer and of course, neuball analytics.

The old site was HTML so there was the HTML extension on all the pages so we’re using redirection to take some of the heavily used pages and redirect them to their WordPress equivalent. There wasn’t a whole lot of traffic had to redirect and that’s a good thing about redirection, it actually gives you a list of the pages that are being redirected and account how many times it has been redirected and so we ended up redirecting a few. I mean seriously, look at this, there was nothing there.

The current projects hit link had one project in it, so I wasn’t surprised that their funding was down a little bit. It's just the same thing. “Search everything” does search all the custom taxonomies and the custom post types. And then “Top 10” is just the top 10 post on the page.

We use that on Twitter. Because of that infographic that our graphic designer put together they wanted to have word clouds. So I assume they’re going to be changing in our next provision because not many people are actually using word clouds anymore because they’re kind of hard to read. 

26:28

 So these are the top research areas and then the top popular pages are both using top 10. Like I said, it’s a little sketchy on who you feature when it comes to faculty projects and faculty members, so both the featured researcher and the featured projects are all random.

The WordPress loop that I use just pulls one random researcher every time the page loads five random research projects. So if they complain that they haven’t been featured, it’s because they haven’t submitted anything to be on the sponsored projects website.

Some of the lessons we learned, we haven’t seen any slowdown in this, but I’m concerned about the number of database calls we have going forward. I’d really like to make, especially with mobiIe sites, I’d like to make this as fast as possible.

I like it not break the database and the cage match earlier, the WordPress presenter did say, “Well, if you have more than a hundred users, it’s a pretty good chance it’s going to crash your site”. We may not get more than a hundred users at the time on the research site, but it’s something to think about for our main page.

27:38

Going forward, I really think need to empower the users to update their own content. Another bearer of the whole thing is when they tell you they want to research, the administration tells you they want a research project, it falls to you to kind of update this stuff.

So luckily, the project, the featured projects that we put up are all in a database already in the main campus they can spit out a CSV file, I can import it in the old data. 

We’re requesting with that, the database lady over there, and she sends me report once a month, with updated projects. So that’s no big deal. It’s not time intensive.

Updating faculty profile is extremely time intensive because it’s changing all the time and we’ve got over 200 faculty and as I was talking to one of the ladies here, our school is pretty small, like the mechanical engineering, they got 1,200 people, you know, that’s really big.

And we also wanted to make everything a lot more simple in the future. When we previewed this with our Associate Dean of Research, he has a PhD in chemical engineering, but he did not know what a slide or a score. So knowing that he could click on these buttons he didn’t really understand it.

They also, in the original version, they had a lot more of the infographic and he wanted to get rid of it, because he’s like, you know, we’re engineers, we want graphs, we want bulleted list, don’t give me all the text.

And, actually I kind of like that idea because I can’t stand, I got, maybe it’s because I work in web development, my attention span is now short so if I can’t get even through the text, why is a student going to get through it?

29:21

And some of the things we want to go with in the future, I added a URL rewrite to those custom projects and I don’t know if you looked at this but it pulls the full name of the title in the URL and once again, I can’t adjust that because if I take the wrong word out, because I tried to, I tried to shorten these, some of these manually, and the researchers called me and said that’s missing the word oxidants, your recommendation, you need to put it up, and I’m like your title is 36 words long. I'm like that's a little crazy. So, I may need to go and take the rewrite rule out and just use the built-in WordPress shortner or I'm going to have, once again, this is empowering the users. They can input in their own projects. They can choose the URL for their titles. In standardizing keywords, I pulled in those projects and there were over 800 unique keywords for these projects. And a lot of it is on the same thing.

So, you know, somebody's doing work on solar cells. They might have input it in the database as solar cell, solar array, solar cells (plural); the same with aerospace - aerospace turbines, aerospace technology, aerospace whatever. And so, I met with the librarian and the associate dean, and their librarian for the engineering library supposed to put together standard set of keywords that everybody is supposed to choose from first.

 

31:03

Because when you submit to a magazine or an actual research article or research publication, there is a set keywords for everything you submit. You can't just put whatever. It's very specific for that journal. So we figured that it should be very specific for this website as well. We'll see when that actually happens. And like I said, they do want an automatic end date. It is now in the system that it will automatically pull that in as a custom meta field into our database but it's not there on the custom loop yet.

And the other big thing that is going be coming up this year is, we have Centers and Consortia and they want that as post type. So basically, their research institutions. Everybody in UK that does research can roll out their own corporation basically and so, they get, you know, funding for this projects. And then during the summer, they can sell the research company and they spend all their time traveling instead of teaching.

 

32:05

So they want me to do the same thing with centers that we've done with the researchers. They want a brief description of that research center, the keywords that they are participating research in. Once again, there's like hundreds of those just for the College of Engineering for the undergraduate programs.

And then, I'm hoping that we can test out that plug in for the custom taxonomy meta data or data. So, are there any questions or answer that kind of quick? Yeah.

[Cross-talk]

Chris Gabel: That's the thing with, I guess, has been recorded with slushy. They didn't have an idea of how they want to measure success. They still don't. Success right now is measured by the happiness of the faculty members.

 

32:58

That seem to be the biggest thing right now as they wanted to have somewhere that all the faculty could have something that's uniform with their research projects and their research areas. That's a great thing that we need to, probably, implement of some sort of success. And we've definitely better traffic because the old side didn't have any on. Yes.

[Cross-talk]

Chris Gabel:Yes, I have loop datum. I haven't implemented them. I don't maintain the database. So whenever I installed plug in then it affects either authentication or the database. I forerun it by our database administrator. It recently cut him to 20 hours until he finishes his PhD. It affected his visa status or something so he had to finish school before a certain date.

 

34:00

So, he's kind of hard to get a hold of it at the moment. Yes.

Audience: Imagine users, they've how many users at that time?

Chris Gabel: Right now, it's just me. I have to update everything and that's another, I'm sorry. The question was, you know, how many users do we have. The university just purchased the license for digital measures. It's what they want to use for all of their CD reviews and 10-year promotion, kind of stuff. Apparently, there's an API that I can use to pull the data directly from their database. That's one of the solutions we're going to go forward. So, I don't have to keep updating all the faculty profiles.

If we decided we don't want to use digital measures, we've also looked at NL. plug in so that they can log with their active directory. Authenticate that way and then update their own bio's. Because you actually pull in, you know, their contact information. And everything that's in AV right now, we can pull in the WordPress dynamically when they log in the first time.

 

35:00

And you can set it so every time they log in, it automatically updates from active directory. So, hypothetically, we will end up having a 120 to 200 users.

[Cross-talk]

Chris Gabel: Yes. That's where the caption plug in will come in. And like I said, I don't think that there's going to be a hundred faculty members updating their profiles in one time.

[Pause]

Chris Gabel: Yes.

[Cross-talk]

Chris Gabel: No. The CSV importer by default, as long as you put the headers in properly in the CSV file, it pulls it automatically. The only thing I didn't have to update with the plug in is I didn't check for duplicates. So, it's actually really well documented in the code. I've just had to find the area where they put.

 

36:00

They actually created the entry in the database and I had it searched for the title first before it would import the entry. Is there any code that you guys would like to see posted? Because I got a few of the snippets on the site. It basically, you know, the taxonomies, the custom loop for the researchers and the hook which the hook, like I told you was from Justin Tadlock.

But is there anything else that you'd like to see from, actually, inside the template?

[Pause]

All right.

[Cross-talk]

Chris Gabel: The custom template?

[Pause]

Chris Gabel: Well, at the moment, I don't have it up and we actually. I'm sorry.

[Cross-talk]

Chris Gabel: Okay.

Audience: They're going to see how it pulls their research from it.

36:57

Chris Gabel: Well for the research projects, and this is the not the actual, you know, the WordPress template but this the page that comes up. First, it just checks for the research areas with the term list. You specify what you wanted to say before the terms are stood out, scoops up the term list, the departments, the primary investigator, the co-PI's. And actually, I do have the page where like this specific project that's up right now.

There's the research area categories which just a little off the page. Just the same place where you have categories and tags for posting. It pages if you implement it is the same place where you put the research area, the same place you put the department and the primary investigator. And like I said, the one thing that I wish I could change is the slug for the primary investigator has to be the same as the slug for the researcher for to recognize that there's a profile for that user and re directory URL.

 

37:58

And then, there's actually a lot of custom med-ed information for the professor or the researcher. They wanted the title comes in as first name last name because that's how it's red, you know, by common people. So, I had to put a custom metadata or metafield in for their last names so you could sort by their last name. Or sometimes, they're in multiple departments, so I had to put in their primary department and then the URL for that specific site as one of the features that they wanted afterwards.

There is their contact information, their office location, city state, and this is their primary location. They have multiple offices around campus. One of the biggest complaints I've also had is our directory. It was built in 2003, instead of using a separate field for first name and last name, it's the last name comma first name. And yes, it is awesome but it's used everywhere for the contact form.

 

39:00

So, I haven't had time to change it. So, when you click on contact me, it goes to the contact form. So, we had to put in the URL-encoded username and that's what the form pulls. And then they each have their personal profiles, pages. This guy just started last week. So, he doesn't have this setup and then usually they have additional websites they want added to their profile page whether it's a cluster teaching or their unique research site.

And another feature that we added was a custom-image size for the thumbnail which we actually talked about this morning. It's another piece of code. It's one line. It's a couple of more of that actually because first I checked for that. I checked to see if it allows thumbnails.

 

39:56

And then if the thumbnails exist, we just added a new thumbnail size as 200x240. So, it automatically pulls in and craps the photo to this size and orientation. Any other questions? Yes.

[Cross-talk]

Chris Gabel: Yes.

[Cross-talk]

Chris Gabel: The custom metafields, I believe, were separate table. And in that table has the key that puts some together just post ID.

[Pause]

Chris Gabel: I appreciate it you're asking now and that's the last thing of the day. So, thank you guys.

[Applause]