I’m a bit late to the game to play with ChatGPT, only logging on for the first time this morning. I recently received an email from a faculty member asking me how his students should cite in APA format, so I figured I should take the AI out for a little chat.
The AI and I (that sounds odd) had a polite conversation where I asked it/them to find a list of Ice Cream companies in Columbus, Ohio. The software generated a list of 10 companies in the area. While the AI failed to provide sales information(most are private companies), it was able to provide website links for the companies.
Who am I?
I asked the AI “who is Chad Boeninger?’, but sadly despite my presence on the web, the software did not know who I was. FWIW, there’s not many “Chad Boeningers” or even “Boeningers” on the planet. The AI was not “trained” with data to identify me, and it currently does not have access to the web.
I’m no rock star
I then tried my hand at confusing ChatGPT with the names of musicians who have some fairly common names. Evidently the AI is a fan of DMB and R.E.M. and associated my search with those celebrities, despite the more normal/common names.
Recommendations for more information
ChatGPT does not cite the sources it uses to build an “answer” to your queries. However, if you ask ChatGPT where you can find more information, it does a reasonable job at offering vague suggestions for the types of resources you might consider. The answer below is very similar to what a librarian might teach in an introduction to research class.
A while back I posted about how I was occasionally finding lip sync lag (ala “Kung Fu Theater” when I recorded my on-camera video introductions and conclusions using Screencast-o-matic. No matter what I did, I would still experience some lag when recording my face, and my lips and words did not quite match up in the final video. They appeared to be out of sync by a couple of frames. I’ve since changed my process and have now removed the lag entirely.
I now use the Logitech Webcam Software to record my introductions. I use my webcam to record my video, but still use the Blue Yeti to record my voice, as the Logitech software allows you to choose a microphone source separate from the webcam. This keeps the audio levels consistent between the Logitech software and Screencast-o-matic records, so I don’t have to do much fiddling with audio in my movie editor.
Since I am editing my videos in Windows MovieMaker (yes, it still works!), the additional step in recording in Logitech doesn’t really take any additional time. After I record the Screencast-o-matic demonstration, I download the video file to my computer, and the file, along with the Logitech video file, into MovieMaker to edit.
I made this short video to test the audio and video quality of the Blue Yeti microphone and the Logitech C920 webcam using Screencast-o-matic. Sometimes if you use the audio from another mic like the Blue Yeti, but record your video another source, such as I do with the Logitech C920 Webcam, there can be some voice-to-video lag. When the lag is present, the lips of the speaker will be out of sync with the audio and it can look like a badly dubbed 1970’s Kung-Fu Theater film.
In the video below, I tested using just audio from the webcam, and then audio from the Blue Yeti mic, to see if there was any lag. I had just restarted my computer, so the internal memory and page file was pretty empty, and I had all other apps closed except for Screencast-o-matic. I did not detect any lag in the video from either audio source. The sound quality is also noticeably better using the audio from the Blue Yeti microphone. It appears that if you find lag, restarting your computer and closing all extra applications will help with producing better quality, and in-sync, audio and video.
Last summer I started using some new hardware that has made my video making a lot easier. The new setup helps me make better quality videos while reducing steps and saving a substantial amount of time.
Some time ago I shared how I make my instructional videos. In that post I detailed how I recorded my video introductions with a dedicated camcorder (or using the video mode on a standard camera) and then captured the screencast using Screencast-o-matic. I was recording the camcorder audio with a lavalier mic, but recording the desktop audio with a gaming microphone headset. While the audio was good, the levels from the two different sources never quite matched, despite my best attempts to equalize them in my video editor. My old process also required me to plug the camcorder into my desktop computer, then download the video from the camera. While this did not take a huge amount of time, it was an extra step.
The picture above shows my new and improved setup. On the left is a Logitech HD Webcam that records up to 1080P video. On the right is a Blue Yeti Microphone that records excellent audio. Both are connected to my computer via USB. I’ve stopped using the video camcorder altogether and now just record my introductory video with the webcam, Blue Yeti, and Screencast-o-matic. I then record the desktop demonstration with Screencast-o-matic and the Blue Yeti mic. Because I am using the Blue Yeti for the audio source for both the introduction and the screencast, there isn’t any need to adjust the audio levels. Both sound awesome since they are from the same source!
Since the Webcam is already attached to my computer, I don’t have to combine multiple video files. I will usually record in the introduction (and outro) at the end of my screencast, using the same video file. I then export the file to my desktop and then do my editing in the old faithful Windows Live Moviemaker. Unfortunately, the editing in SOM is still a little slow and clunky on my machine.
In general, the quality of the video is superior to my old method and my new method definitely saves me a lot of time. However, there are times when the video can lag behind the audio in the on-camera personal introduction segments. This is usually caused by having too many applications open on my computer while recording video, so closing unneeded programs helps.
The other drawback of this setup is that since my camera and mic are tied to my desktop, I am by default tied to my desktop. Should I want to take my show on the road and do a video introduction outside my office I’ll need to use a different camera. Other than that it’s a great setup.
Making videos for my library patrons have saved me time, allowed me to better serve my them, and enabled them to quickly find answers to their questions. I first started using web video for library instruction over 6 years ago and have learned a lot along the way. My hope is that the information in this post can help others make web videos and screencasts to reach and teach their patrons as well.
Update: For a video demonstration of how I make videos, please see this post.
Why Web Video?
My job as a librarian means that I help students and faculty find the information that they need. This help is often provided via classroom research sessions, individual in-person consultations, email, chat, phone, and even text. As a single librarian who serves a college of over 2,000 students and 80 faculty, I am constantly looking for ways to provide better outreach and support to my patrons, while also looking to increase efficiency on my end. To this end I strive to reach the most people with methods that can be scaled for the size of the population that I serve. I have found that web video is one of the easiest and best ways to reach as many people as possible with the least amount of my time and effort.
What type of content?
A sample of my video content
The content for the videos that I make generally comes from the questions that I receive. It’s pretty easy to know what kind for videos you need to make based upon the frequency of questions that you get (or may get). In many cases, I make videos for general topics, such as how to find economic information in Passport GMID. In cases like this, my goal is to make a general video that will show users how to navigate a database with a rather complicated interface. The question that this video answers is a question that I am likely to get quite a bit. When I do get the question, I can easily send the user the URL to the video via email or chat. This saves me time from having to type the email explanation of how to do this, and the user gets a visual explanation of how to find the information.
At other times, I make more specific videos that address specific research topics. Currently I have over 200 business students who are doing a feasibility study of placing a popcorn franchise in the local Athens mall. I met with all 200+ students (5 classes of 40 students) last week to teach them about how to research this topic. I also created a specific Business Blog post to address the tools needed to research the popcorn and snack foods industry. Before the class I made three videos, one for finding industry ratios for popcorn stores, one for finding demographics of popcorn consumers, and one for finding popcorn consumption by location. All three videos were demonstrations of what I did in class, and were created so that the students could refer back to them after the class. I even showed the video about popcorn consumption during class, as SimplyMap, the database demonstrated, can often take longer than 5 minutes (the length of the video) to explain during a live demonstration.
Regardless of the context of the video, I try to make all of my videos answer a particular question. That is, I would rather make a video on “How to Find Stock Reports in S&P Net Advantage” than just “An Overview of NetAdvantage”. This helps in a couple of areas. First, I believe that a video that addresses a specific question will help me keep the video focused and shorter. Second, someone searching for “stock reports” on the web via Google or on YouTube is more likely to find the first title, rather than the second.
How much time does it take?
The first video I ever made (before YouTube was a household name) took me over 6 hours to make. At that time (around 2005 or so) you would have to worry about things like video formats, compression rates, streaming, and more. There was very much a trial and error approach and a very large learning curve when putting video on the web. Since then, web video formats and players have become more standardized and web video hosts have made uploading and publishing video a lot simpler. Technology has advanced to the point that people can now focus primarily on the content of their videos, rather than worry about video codecs and compression rates.
When making web videos for library instruction, the time spent is directly related to the amount of practice you have had. If you are just starting out, your first video may take you a while. You may stumble with the screencasting/screencapturing tools. You may fumble with the recording of your content. You may try for perfection, but not achieve it. I’ve had a lot of practice and have established a pretty good system over time (I will try to write up time-saving tips in a separate post soon), so I am usually able to create and publish videos very quickly. With the three examples mentioned above, I was able to record, edit, publish to YouTube, and post all three videos to the Business Blog in just under an hour. My record is seven videos in about 90 minutes. I’m still not as efficient as I would like to be, but I have gotten a lot better over time.
How much time is saved?
One may think that not much time is saved if you have take extra time to learn how to do something new. That may be true early on, but I can guarantee that as you get better making videos, your return on time invested will increase exponentially. At the time of writing this post, those three videos that took me an hour to create from start to finish have been viewed a combined 172 times in the past 10 days. I can tell you for sure that having those videos available has saved me and my students a huge amount of time. They get the answer they need quickly and in the context of their specific research, and I can avoid answering the same question over and over again. That frees up my time so I can answer the more unique (and harder) questions.
How long should my videos be?
I try to keep my videos to less than 5 minutes. Limiting the length to 2-3 minutes is even better, but some concepts or databases can’t be explained in that amount of time. If the video is likely to go over 5 minutes, I try to find ways to divide the video up into several smaller videos. As an example, a single video covering every single feature of Passport GMID (a huge and powerful market research database) would likely take 10-15 minutes minimum. I’ve chosen to divide all the various features of Passport GMID into multiple videos. The important thing to remember is most people are not watching my videos for entertainment. While I try to keep the videos as interesting as possible, the fact of the matter is that few people are going to find “demographics of popcorn consumers” a lot of fun. Therefore, I try to get them through the content to find their answer in the most efficient manner possible.
Script or no script?
Depends. If you are just starting out, I’d suggest going with a script or at least a brief outline. Most of my outlines have been written on a Post-It note, so you don’t necessarily have to have an elaborate script with storyboard. For most of my videos, I will usually run through the search demonstration before recording, then repeat the same basic process for the recording. This practice run helps me to avoid stumbling and having to repeat the recording numerous times. A script is also useful in keeping your thoughts on track and keeping you within your time limits.
Which software for screencasting?
While there are a ton of options for recording your computer screen (screencasting), my favorite is Screencast-O-Mattic. I like it because it is web based, and it has free and paid versions. I’ve been using the free version for quite some time, and I have found that the only drawback is the company watermark in the bottom left of my videos. I recently upgraded to a Pro Account for 12 bucks a year, which removes the watermark and gives me some more advanced editing features. If you are just starting out, stick with the free version for now.
How do I record my voice?
For most of my recordings, I use the microphone from a Logitech Webcam. It’s easy to use and the sound is decent. You can also use the microphone line-in port and a lavalier mic or even a Skype/webchat headset. The hardest part for me (and others in our cubicle farm) is finding a quiet time or place to record the screencast. I’ve found that I can get a lot of recording done before 9 a.m in my office, but you may find that late evening or lunchtime may work better.
Where do you record your videos?
I almost always record my videos in my office in the library, though I have been able to record a few at home. Making videos is a good work-from-home task, especially while a sick child is taking a nap. In a recent video, I actually recorded part of a sports business video in front of our football stadium (in the cold). I’ve got a list of other locations around town that I may use for the introductions in future videos, just to spice things up a bit.
Which resolution?
Screencast-O-Mattic allow you to change the size of your video to meet your needs. With the availability of modern widescreen monitors, I prefer to record with Screencast-O-Mattic’s HD setting of 1280×720. This gives me the largest size file to work with, should I need to resize in my video editor. In general, I upload all my videos in HD 1280×720.
Customize resolution size in SCOM
Why and How do you add the video of yourself in the videos?
In all of my videos, I bookend the screencast with a personal video introduction and conclusion. I do the video introductions to make sure students and faculty place my voice and name with my face, and to personally introduce myself as the expert. Without the video introductions, I feel like I am just a voice on the screen. The video introductions add a little personality to the videos and highlight me as a resource in addition to the databases that I am demonstrating. Plus, as shown in the image below, the video introductions allow me to show off my collection of sweater vests.
A typical video introduction
I use a small inexpensive video camera to record the video clips. I then piece the clips together with the video from Screencast-O-Mattic in Windows Live Movie Maker. The process is a little outside the scope of this particular post, but I will write that up soon in another blog post and link to it from here.
To edit or not?
If you are just starting out, my advice is to try to avoid editing. With editing you add another level of complexity that you may not need or want to mess with. If you totally flub something in your recording, it should be fairly easy to re-do the recording, especially if you are sticking to the 2-5 minute time limit.
How do I know when my video is done?
The easy answer to this is “when you have covered your topic as clearly and as efficiently (short) as possible.” However, the definition of “done” can depend on individual expectations. I almost always record my videos in one take, listen to them once, then publish them to the web, warts and all. If I notice in my listen-through that I provide incorrect information or if I am not very clear, then I may re-record the video. A couple of “umms” and pauses won’t really harm the quality of your video, so don’t be too concerned with perfection. I’m not trying to be Lucas or Spielberg—I’m simply trying to make content that helps students and faculty. If I was super critical of my videos and demanded perfection, my videos would take a lot longer to produce and I might not meet my patrons’ needs in a timely manner. In other words, I don’t think they will care if I mispronounce a word or two as long as they get their question answered. If you think your video is bad, you can search YouTube to find worse ones and feel a lot better about your skills.
Where should I upload?
YouTube. But don’t stop there. You can’t automatically assume that because you upload it to YouTube that your intended audience is going to find it. You will need to cross post and promote your video all over the place. Embed your videos on your libguides/research guides, your blogs, and wherever you think your audience will find it. If you can’t embed the actual video, try to look for creative ways and place to link to the video. We even have an image link in our EBSCO database interface that links to our videos about searching EBSCO.
Our YouTube videos linked in EBSCO
But you don’t use just use YouTube, do you?
That is correct. I actually use a service called Blip.TV as well. I started using them when YouTube was in it’s infancy because Blip had better video quality than YouTube. This may also be covered in another blog post, but for now let’s just say I use both Blip and YouTube. If you are just starting out, just stick to YouTube to keep things simple.
How do I promote my videos?
I have my YouTube account set to automatically Tweet my upload, which occasionally gets picked up by our official @aldenlibrary account, which can then get picked up by students, the official @ohiou account, and others. I’ve often had pretty good success emailing a video to faculty members. I usually tell them that I’ve been helping a lot of students with the same questions, and my video should help others with similar questions. They are usually pretty good about forwarding the email to their students if they know it will help them find better information for their projects. I also usually show at least one video in every class I teach, and show them the link on my blog where they can find all of my videos. I also use a WordPress plugin to show related posts and promote the videos that way (yet another concept that should be its own post when I get around to it).
Twitter mentions in action
What else should I know?
Hopefully by now you have the basic knowledge needed to create your own videos. In future posts I will address related topics such as cameras, editing, WordPress plugins, and additional resources/readings. Stay tuned for more on the topic, but for now have fun making videos. Should you have any questions or comments, please feel free to submit one on this post. Finally if you would like another business librarian’s take on making videos, check out Steve Cramer’s recent post on the topic.
Libraries are often accused of using terms that the general public does not understand (catalog, OPAC, stacks, reserves, reference, periodicals, etc). As this video demonstrates, misunderstanding jargon goes way beyond the walls of the library. In this video, an employee from Google asks random New Yorkers, “What is a browser?”. The answers are both amusing and a little disturbing.
Texting is a great way to answer reference questions, and it doesn't have to cost you a dime.
Many libraries now offer text-a-librarian services to extend reference service to patrons on their mobile phones. If your library has pondered the idea of implementing a text messaging reference service, but couldn’t quite figure out a way to pay for the service, I hope you find this useful. Our library has been running a text messaging reference service since September 1st, and it didn’t cost us a dime. Here’s how we’re doing it, and how you can, too.
We have toyed with the idea of having a text message reference service for some time, but only got around to implementing it this summer. My colleagues and I have attended webinars from companies that that offer a fee-based text-a-librarian service, and have read journal articles about libraries that have set up their own service with just a cell phone. While I understood the desire to offer reference services via text, I never quite saw the need. After all, we already provide services via IM, meebo chat, Skype, email, phone, and in person, and survey results from two years ago suggested that patrons would be unlikely to ask their questions via text message. I also couldn’t really justify the cost of paying a service or a cell phone bill, especially if we built the service an no one used it. However, with the development of our mobile information site (to be discussed at another time) we began to see that a texting service was important. We also found a way to provide the service for absolutely free and we set the service up in a day. If it didn’t cost us anything, why not try it and see what happens? Here’s a step-by-step guide to the tools we used and how we set up the service.
Google Voice
The first thing we did was set up a Google Voice account, which is free. Google Voice allows you to set up a number of your choosing, and then redirect calls to that number to your actual registered phone number. When setting up your phone number, you can either try to get a number in your area code, or choose a vanity-style number. We figured that we’d do something clever, so our number is (352) 354-2733 or (352) 3LI-BREF. Our IM handle is OHIOLIBREF, so we wanted to remain somewhat consistent. Unfortunately, there wasn’t a number in our area code that spelled out our handle, but it doesn’t really matter. Cell phones (and our patrons) have no idea what long-distance is anyway. Regardless, you’ll want to choose wisely, as Google will charge you 10 bucks if you ever want to change your number.
When we set our account up, we just redirected our new Google Voice number to our phone at the reference desk. Note that because our reference desk phone is not a mobile phone, text messages cannot be sent to it. However, as I will demonstrate later, you can receive and send text messages via the Google Voice interface in your web browser.
Gmail
On the Google Voice settings page, you will want to select the option to send incoming text messages to your email. We chose our Google email that we used to set up the service. We use this email address exclusively for our texting service, and it is not made available to the public. The way we have the service set up, we only want text messages coming to the email account, so as not to clutter our inbox with other messages. Since Google Voice and Gmail are both Google services, I would strongly suggest that you use a Gmail address for this. Otherwise, the transfer from Google Voice to another email system may slow the service (and your response) down. The image below demonstrates the settings for the SMS/text to email in Google Voice.
Send-to-email settings in Google Voice
Trillian
Trillian is a multi-protocol instant messaging client that allows you to connect to multiple IM accounts at one time. However, we’re not using Trillian to connect to an IM service. We are using Trillian to monitor our incoming email to the Gmail that are forwarded from Google Voice as incoming text messages are received. Yes, it sounds odd, I know, but bear with me here. It will make more sense in a minute. When you set up Trillian, you will want to use the Gmail account for a Google Talk account. We won’t be chatting with our Google Talk account, but only using it as a way to get our Gmail notifications through Trillian. Once you have the Google Talk account set up in Trillian, look at the Miscellaneous settings and make sure all boxes are checked as shown below.
Make sure you check the boxes on the Miscelaneous screen for your Gmail account
U2
You can set up Trillian to check your email to certain accounts. You can also customize the notification windows and sound alerts in Trillain. Since we are staffing our texting service at the reference desk, we needed an alert to notify us when an incoming text arrived. In the advanced options of Trillian you can customize sound alerts and on-screen notifications. Trillian will play any WAV file as a notification, so we chose the first 5 seconds of U2’s Mysterious Ways (it’s a cool service, so we needed a cool song) . Like I said before, we set the service up in a day; we’ll get more clever with our song choice down the road. One colleague has suggested Help by the Beatles, and others have requested that we do a 12 Days of Christmas theme closer to the holidays. It doesn’t matter what song you choose, as long as it’s loud enough to be heard and recognized above other distractions at the reference desk. Regardless, choose a song that is fun! I love getting text messages while helping other in-person patrons at the desk, as they inevitably ask “What’s That??!??”, and it’s a great opportunity to pitch the service to them. The images below demonstrate how to set this up.
On the Advance Preferences screen, click on "Automation" and look for the New Mail Event
Check the box that says "Local File", then browse to the WAV file on your hard drive.
So how does all this work?
After all that convoluted setup, you’re probably wondering how in the world all this works. Or, you’re likely willing to say “to heck with this” and pay some company $1200 a year for a text messaging reference service. Don’t get your your checkbook yet. Here is how it all works together.
1. Patron sends a text message to our Google Voice number, which we have displayed on our website.
2. Google Voice receives the text message and forwards it on to our Gmail email address.
3. Trillian recognizes that a new email has arrived to our Gmail account, and notifies us with a desktop alert and by playing U2’s Mysterious Ways. (make sure your speakers are on!!!)
4. Librarian goes to the Google Voice inbox via a web browser to retrieve and respond to the text message (sometimes this requires refreshing the page to see the most current message). We have a tab in Firefox open to Google Voice at the reference desk all the time.
5. If the patron responds back, the process goes back to step #1 .
Without Gmail, Trillian, and U2, the librarian would have to check the Google Voice account by refreshing the page to see if new texts had arrived. How many times could we expect a staff member to do this each hour? Likely, it would be forgotten, so instead, Gmail, Trillian, and U2 automatically notify us when a new text message arrives. In other words, Google Voice is our text messaging device (or phone, if you will), Gmail is the service provider (much like At&T) , and Trillian and U2 are the ringer.
So does it work?
You bet it does! In our first month of service (September 2010) we answered 54 questions with our brand new text message reference service. This is compared to 110 email questions and 713 chat/IM questions. Overall, text message questions comprised 1.2% of our total questions received by our Reference Department. That may not seem like a lot, but we are quite pleased with the results. That’s 54 more text messaging questions than we have ever answered before, and we didn’t have to spend any money to do it. I’d say our initial return on investment is pretty good. 😉
See it in acton
If you still have your doubts, the video below demonstrates how the service works and how a librarian answers a question.
I work every Monday night. The business students all know that I work every Monday night, as my hours are posted on my Contact Page, and I tell them in every class I teach when I work. So here is what one Monday night looked like a few weeks ago, between 6:00 and 7:30.
Busy Night on IM
Yep, that’s a lot of IMs. My fingers were going crazy. I was in my office monitoring our general IM/chat reference service, but I also had my own IM open as well. You know what’s funny? Nearly all of those students were in our group study rooms about 150 feet from my office. They could have just come to my office to ask a question, and actually a few other students did. But these students chose to contact me in a way that worked for them. Wonder what would happen if I didn’t make myself available via IM? I would bet that most of those questions would have gone unanswered, and I would have lost a valuable customer. How are you making yourself available to your patrons?
A survey from the researchers, covering the third quarter of 2009, suggested that 52% of smartphone owners use their handsets to check product descriptions, that 36% check rival retailers’ prices when deciding whether or not to buy a product, and that 34% used “m-commerce” channels to make purchases.
An analyst for eMarketer suggests, “A retailer’s best defense for maintaining customer loyalty is to develop a mobile offering that allows in-store shoppers access to customer reviews and other product information on its website.”
Actually, the best way to keep me as a customer is not so show me a flashy mobile website. To keep me in the store, honor the competitor’s price that I find on the web. While Christmas shopping in December, my wife and I went to Border’s to find the Julia Child cookbook for her mother. While shopping in store, I pulled up the book on Amazon, who had the book priced at least 10 dollars cheaper. I showed the price to a clerk, who simply shrugged her shoulders, and said, “Yeah, it’s cheaper there.” We walked out empty handed. Now I know not everyone can honor the deep discount pricing of Amazon, but give me something. Maybe 20 percent off my next purchase, a free cup of coffee, something to entice me to buy your product when I find a better price, something to get me to come back to the store again. My local bike shop is competitive on some things, but generally the bigger online retailers such as Nashbar and Performance beat them on price. However, they make up for the price disparity with the service they provide. They answer my questions, and if I ever have a problem with something I buy there, they take care of me. If I need a product they don’t stock, they’ll generally order it for me. They may not be able to match the prices, but they offer perks. Other retailers should do the same thing, or the next time I find a better price on my phone, I’ll be leaving the store empty handed again.
