The Distant Librarian

I'm no longer concerned about energy use and ChatGPT

Last week, I tripped across a couple of posts from MIT that I thought included the clearest explanations I'd seen of the cost and *reasons* for the cost of training and using Generative AI.

Explained: Generative AI’s environmental impact
and
The multifaceted challenge of powering AI

I added those to my list of bookmarks on the topic, which I've now organized and made available at https://raindrop.io/ppival/energy-and-ai-54867175 - it's currently sitting at 21 bookmarks, but with a few of those being duplicates, or one source pointing at another. (also available as an RSS feed!)

But the point of this post is to highlight Andy Masley's A cheat sheet for why using ChatGPT is not bad for the environment. (ht to Simon Willison)

Andy has done a wonderful job simplifying the issue by comparing the cost of using (including training costs) ChatGPT to other common activities. THIS is the droid I've been looking for! It's a long read, but well worth your time if, like me, you get concerned questions about the insane environmental costs of using AI in every single presentation you make about AI. I thought this was because the youth today really do care about the environment (and they do), but after reading Andy's well-reasoned posts I realize it's because we, collectively, really don't understand how electricity is measured. I know I don't! I find it ironic that towards the end of his post, Andy writes,

It’s 3 Wh!!!!
In a lot of these conversations, I have a very strong urge to grab the other person’s shoulders and say “This is 3 Wh of energy we’re talking about!!!! We agree that’s the number! 3 Wh!!!!!! That’s so small!!!! Don’t you know this?!?!?! What happened to the climate movement????? All my climate friends used to know what 3 Wh meant!!! AAAAAAHHHHH!!!!!”

But I'll readily admit I can't look at a light bulb, or my iPad, or the Keureg in the other room and have even the remotest idea how many Wh of energy it uses!

So here are just a couple of graphics I found particularly meaningful (click to embiggen), and which I'll be including in my future presentations. Do read the post for fuller explanations and additional compelling infographics!

 

The State of Generative AI Use in Canada 2025: Exploring Public Attitudes and Adoption Trends

This is a very current report, with questions being collected between Feb 19-March 1, 2025. From the Social Media Lab at the Ted Rogers School of Management, Toronto Metropolitan University (TMU), The State of Generative AI Use in Canada 2025: Exploring Public Attitudes and Adoption Trends. 1,500 Canadians were asked questions about their use and knowledge of generative AI tools, and they were also asked to back up their statements with a short quiz, upon which they didn't do very well :-/

Regardless of their self-assessed proficiency, on average, respondents answered only ~2.5 out of 7 questions correctly, indicating a relatively low level of knowledge of Generative AI.

I think it's worth spending a little time with the actual report, so don't stop with the executive summary on the linked website above.

I was concerned to see how often respondents think generative AI is used in news media (3.1), and how that appears to be reflected in how little the media and news organizations are trusted by these people (5.4). And 3.1 was a little odd to me because it isn't (and can't be) followed up with any data about how often media and news organizations DO use generative AI.

Things I wish the report also made available:

• The raw survey data, which would likely answer my other wishes
• The list of tools respondents used
• Breakdowns of all the questions by age, gender, geography, political leanings

I hope they add a separate category for Libraries (public and academic) to the list of trusting entities. Maybe Middle and High Schools could be added to the list, too. I wish they hadn't included the answers to the quiz on the same page, which would make it easier for readers of this report to test themselves.

If you want to take the 7-question quiz yourself before seeing the answers on page 25 of the report, you can do that here:

 

Incidentally, ChatGPT got all the answers correct when I gave it the questions to build this.

Gruzd, Anatoliy; Mai, Philip; Clements Haines, Anthony (2025). The State of Generative AI Use in Canada 2025: Exploring Public Attitudes and Adoption Trends. Social Media Lab, Toronto Metropolitan University. DOI: https://doi.org/10.6084/m9.figshare.28664780

Using ChatGPT to keep up on the news is like using a potato to commute to work.

The title for this post came from Stephen Downes as a comment to my Mastodon post last year on the first article in this series: ChatGPT’s search results for news are ‘unpredictable’ and frequently inaccurate.

"Building on our previous research, the Tow Center for Digital Journalism conducted tests on eight generative search tools with live search features to assess their abilities to accurately retrieve and cite news content, as well as how they behave when they cannot." AI Search Has A Citation Problem - We Compared Eight AI Search Engines. They’re All Bad at Citing News.

Fascinating stuff, and some great visualizations.

But Stephen nails the problem. The authors note all the way at the bottom of their page, "While our research design may not reflect typical user behavior, it is intended to assess how generative search tools perform at a task that is easily accomplished via a traditional search engine." (emphasis mine). These chatbots are NOT search engines! They don't store "chunks of text" in any way shape or form, so it shouldn't be a surprise that they can't tell you where any string of text "originally came from". I'm not saying the authors ARE surprised, and I think this is important information for the public to have, so do spread it far and wide, but I hope you, dear reader, are not surprised by these results!

 

Bluesky Mapping Canadian Skeets

Over on Maps Mania, Keir posted a couple of quick maps he built leveraging Bluesky's firehose jetstream, including US State Social, which:

scans Bluesky posts for mentions of the 50 US states. When it detects a mention, it adds a marker to the map and displays the associated message in the map’s sidebar. Once the map has found messages for all 50 states, it removes all the markers and starts the process all over again.

He built it on Glitch, which encourages remixing. After looking at his code, I figured it would be trivial to swap out the States for Canadian Territories and Provinces, so I immediately asked Claude to do that. Got it in one, and now we have: 🇨🇦 Canadian Skeets 🇨🇦.

I don't know enough about the free version of Glitch to know if this thing is always running, so if you follow that link, and it appears to be not working, let me know? I'm on Bluesky as @ppival of course.

I'd love to be able to click the original skeet to be taken to the user's account, but don't (yet) have any idea how to incorporate that :-)

Apparently I should be able to add a "Glitch button' to the project that allows someone to click and remix, but I can't seem to get it to display, so if you want to kick at the code, it's here.

The Sisyphean Cycle of Technology Panics

I just finished Amy Orben's excellent article, The Sisyphean Cycle of Technology Panics, and recommend it.

I thought it was going to be simply an acknowledgement and catalogue of the fact that yes, we have seen this before, with radio, comics, TV, video games, etc., and while it does include that, it does an excellent job of explaining why we keep seeing what appear to be the exact same types of studies, and why politicians keep calling for inquiries into the latest thing that appears to be corrupting our youth.

A quote from the section describing the reinvention of the wheel, "In none of these cases does evidence ameliorate society's concerns; rather, the focus of concerns shifts to a new technology as society simply moves on." :-(

Orben does suggest a couple of solutions to breaking this cycle, but it really doesn't seem to me that either will come to pass, so hop on the train.

The Sisyphean Cycle of Technology Panics

Orben, A. (2020). The Sisyphean Cycle of Technology Panics. Perspectives on Psychological Science, 15(5), 1143-1157. https://doi.org/10.1177/1745691620919372

One of the best uses of GenAI I've seen yet - CanLII the Canadian Legal Information Institute

A colleague alerted me to the existence of this project, which had flown under my radar. The Canadian Legal Information Institute (CanLII) has used "a commercially available Large Language Model (LLM)" to generate summaries of primary case law, and legislative documents for Alberta, Manitoba, Prince Edward Island, and Saskatchewan.

It's the summaries of legislative documents that excites me. Legalese doesn't seem very approachable to the layperson, yet some of this stuff can be pretty important to at least know about. Here's the Alberta Libraries Act: https://www.canlii.org/en/ab/laws/stat/rsa-2000-c-l-11/latest/rsa-2000-c-l-11.html As soon as I start scrolling I want to quit. But now I can click on the AI analysis tab and have a summary that even I can understand! Yeah, if it turns out to be something that should involve a lawyer, that's not going to change, but at least this summary gives me a decent chance of understanding what's going on with any given legislative document.

If it's worth sharing, it's worth saving - Wayback to Facebook

I don't know why this particular solution didn't initially resonate with me, other than my doggedly wanting to link to THE ORIGINAL source, but with the recent demise of my locally-hosted instance of ShareNews, I've been searching for other ways to circumvent Facebook's ban on sharing news sites.

I discovered a couple of solid commercial options in ShareKit and Rebrandly that I'll keep in my back pocket, but I prefer not to pay if I don't have to. (side note: news links shared through Bit.ly and any customized rebrands of it are also blocked by FB) Next up was a locally-hosted version of YOURLs, but just as I was about to deploy that I realized the solution had been there all along, even as already noted by me:

Send the link to the Internet Archive using the Wayback Machine browser extension, then paste the resulting URL into Facebook. This method has the advantage of ensuring your newsworthy article is saved for others to view in perpetuity, should it disappear from the original location for any reason.

The extension is available for FF, Safari, Chrome and Edge, and also as an app for iOS and Android, so this should cover pretty much any platform on which you prefer to work.

Here's a quick screencast showing it in action. It's actually faster than my previous solution. The only potential drawback is that FB bans all links from the Internet Archive, but I'm sure lots of folks have been using this option since the ban began, and they haven't yet, so fingers crossed!

Oh, and other options include posting a news link to a social media site such as Mastodon or Twitter, then posting the link to that post into FB, but then you lose the open graph data that makes the post so much more fulsome. Metadata FTW!

Generative AI courses for your mom

I know it's cliché, but I AM actually trying to help my mom understand Generative AI a little more (hi, mom!).

I just completed this free course, and recommend it as a gentle introduction to generative AI: Destination AI: Introduction to Artificial Intelligence. It will require an investment of time, though maybe not quite the 6 hours it claims. The only big drawback is that if you fail a quiz, you'll have to wait a full 24 hours before you can re-take it. That said, you CAN move on to another module even if you don't pass the quiz, so not that big a deal.

What will you learn?

• What AI really is (and what it’s not!).
• The impact of artificial intelligence on the workplace and on society as a whole.
• Go behind the scenes of an artificial intelligence project and get familiar with concepts like deep learning and machine learning.

What if your mom's a teacher? I haven't actually taken this course, but Google offers Generative AI for Educators, which is shorter, at only 2 suggested hours.

You’ll learn how to use generative AI tools to:

• Save time on everyday tasks like drafting emails and other correspondence
• Personalize instruction for different learning styles and abilities
• Enhance lessons and activities in creative ways

Launching ChatGPT Voice Chat on an iPhone 12 with back-tap

In today's issue of Ethan Mollick's One Useful Thing, I learned that the iPhone 15 has a unique button I hadn't paid any attention to - the action button. He mentioned how the action button can be used to launch a voice chat with ChatGPT, and I found instructions at TechCrunch. But I only have an iPhone 12!

I knew the iPhone has a feature called Shortcuts, but I'd never really found a use for them. I also recalled that there are a couple of accessibility features on the iPhone that allow actions to be performed by, for instance, double-tapping the back of the phone. I asked Perplexity how to recreate the action button thing on my older phone, and got the following excellent instructions:

Based on the search results provided, here's what you can do with the Back Tap feature on your iPhone 12 to launch a voice chat with ChatGPT:

## Using Back Tap to Launch a ChatGPT Voice Shortcut

The search results indicate that the Accessibility > Touch > Back Tap feature on iPhone allows you to assign various actions to a double or triple tap on the back of your device. [3]

You can use this to launch a custom Shortcut that starts a voice conversation with ChatGPT, without having to unlock your iPhone or use Siri. Here's how:

1. Create a new Shortcut in the Shortcuts app called "ChatGPT Voice" that includes the "Start Voice Conversation" action from the ChatGPT section. [1][2][3]

2. Go to Settings > Accessibility > Touch > Back Tap. [3]

3. Select either "Double Tap" or "Triple Tap" and choose the "ChatGPT Voice" shortcut you just created. [3]

Now, whenever you double or triple tap the back of your iPhone 12, it will automatically launch the ChatGPT voice conversation mode, allowing you to speak your questions and get responses from ChatGPT without having to unlock your device or use Siri. [1][3]

This provides a quick and hands-free way to access ChatGPT's voice capabilities directly from your iPhone 12, using the built-in Back Tap accessibility feature. [3] The search results provide detailed steps on how to set this up.

Citations:
[1] https://www.youtube.com/watch?v=h3fOIE_CJXI
[2] https://www.reddit.com/r/shortcuts/comments/181uy7r/chatgpt_voice_shortcut/
[3] https://www.digitaltrends.com/mobile/how-to-replace-siri-with-chatgpt-iphone/
[4] https://www.reddit.com/r/iphone/comments/1702w4p/how_to_replace_siri_with_chatgpt_as_your_voice/

Works like a charm! (though as I read that response, I do have to admit it's not actually a hands-free way to access ChatGPT's voice capabilities, but I'm just fine with that.)

Google Scholar PDF Reader

I use Firefox these days, so this wasn't a priority. I've had this tucked away since it was announced about a month ago, but only now got around to trying it out, and I like it! Last month, Google introduced the Google Scholar PDF Reader. I thought about making a quick screencast, but of course someone else has already done a nice job, so here you go. I had a little trouble getting it to read local PDFs until I realized I could just drag them on to the browser.

Qualitative Data in the News

It seems a rarity, but I have two really interesting sites open in my browser talking about qualitative data in the relatively current environment. The first, Pandemic Journaling Project Archive Opens for Research, is notable because it fills a big gap. As the article notes,

“There are several large quantitative surveys in the social sciences that are broadly used,” says Karcher, “but qualitative datasets that are large enough to be analyzed from so many different angles are very, very rare. The PJP data are going to be a treasure for social scientists for years to come, and we at QDR couldn’t be more excited to be the permanent home for them.”

That does remind me of another tab I have open that I wanted to explore - A listing of qualitative data results from the lunaris data repository here in Canada...

The second is timely in that it talks about the problems of using AI to try to perform qualitative data analysis. How not to: We failed at analyzing public discourse on AI with ChatGPT. A couple of weeks ago, I was able to help a researcher successfully use NVivo's automatic coding using existing coding patterns feature, and wonder if it would've helped in this case.

Related to the above, I know folks are exploring the use of ChatGPT and other AI tools to assist with qualitative data analysis, but I sure haven't yet seen anything that makes me think we're very close yet. And that leaves aside the questions around confidentiality in letting any online tool churn through qualitative data. Though I do still have to delve more deeply into CoLoop and Qeludra.

2023 Access Conference Recordings

There's not (yet?) a link to each recording in the original schedule, but the recordings from the 2023 Access Library Conference, held in Halifax on October 23-25, 2023, are now live on YouTube. Definitely some good stuff worth checking out!

Two simple ways to share news to Facebook in Canada

Update 05/24/24: my hosted version of sharenews was blocked by FB a couple of weeks ago. See option #1 below, or this post for a little more about it.

Update 10/3/23: both of those Sharenews URLs have been blocked by Facebook, but the developer released his code as open source, so you can now run the service from https://sharenews.pival.me/. Bookmarklet links below have been updated to point to that new location.

If you're in Canada and unable to post news articles to Facebook due to Bill C-18, here are two simple options for you.

1. Send the link to the Internet Archive using the Wayback Machine browser extension, then paste the resulting URL into Facebook. Firefox link Chrome link. This method has the advantage of ensuring your newsworthy article is saved for others to view in perpetuity, should it disappear from the original location for any reason.
2. Someone has built a page called ShareNews.ca (now ShareMyLink.ca) specifically to work around this. All it does is ask you to past your URL, then prepends the URL https://www.sharemylink.ca/ to the front of the URL of the page that you're trying to share, and, for now at least, Facebook is allowing that URL to be included in a post. If you'd like to streamline the process a bit, I built a bookmarklet that allows you to perform that same process w/o having to visit the ShareNews.ca website first. Hit the bookmarklet while on the page you want to share, and it'll prepend the sharenews.ca URL to it and place the whole thing in your clipboard, ready to share to FB. Here it is - just drag this to your shortcuts bar: ShareNews

(if that bookmarklet link gets stripped, here's the actual code:

javascript: (function () { var currentUrl = location.href; var newUrl = "https://sharenews.pival.me/" + currentUrl; navigator.clipboard.writeText(newUrl); alert("The new URL is in your clipboard and ready to post to FB"); })();

Thanks for the pointer to ShareNews.ca, Heather!

Update 9/18/23: updated link and bookmarklet to point to https://www.sharemylink.ca

New to bookmarklets? 

Link wherever you want (YMMV)

Quite some time ago, I used a browser extension called TLDRify that allowed me to highlight some text on a web page, and it would create a link I could share that would jump a person directly to the text I had highlighted. It was a way to both create an anchor link when one didn't exist, but with the added benefit of highlighting the block of text you wanted someone to see. At some point I stopped using it because several browsers threw up scary warnings to the clickee that they were being redirected to a different site, which is true, in that TLDRify actually shows your desired content on a different URL (tldrify.com).

The other day, my friend Peter posted about something called URL Scroll-To-Text Fragment, and it looked promising enough to explore. In my explorations, I think I've learned that TLDRify seems to be working well enough again that I'll likely start with that, but you may be interested in figuring out if either of these solutions might be useful to you. It seems that different browsers may play more nicely with one or the other, with Chrome possibly being the best bet, but I have gotten both link types to work across Windows and macOS, on Firefox, Chrome, Opera, Safari and Edge.

The "fix" to TLDRify is that now all the browsers that are warning about the link are doing so by displaying the following at the top of the page, but otherwise the link works as desired.

So here are some links to try - how do they work for you?

• This one, created with TLDRify, should jump you to the final line of text on a page detailing MPOW's Travel Loaner Device Program.
• This one is way uglier as hyperlinks look, but you can't tell since it's still embedded as a link. It was created with the Scroll To Text bookmarklet described at the Supple.com website.
• The Link to Text Fragment add-on for Firefox creates an identical type of link, but does so with a right mouse click after highlighting the text. This solution seemed the most finicky in that it didn't always actually create the link I wanted.

Vaccine Injury Support Program - Statistics - not working in Wayback?

I've been using Visualping.io to watch for changes on the Vaccine Injury Support Program - Statistics page since I became aware of it, and I thought I was sending changes to the Wayback Machine as well, but after today's update I realized that when I visit the snapshot on Wayback, the captured page claims "No statistics available" rather than showing the actual content of the page. Anyone know why that might be? That's the case for all the captures, which makes them not useful at all.

Fortunately, Visualping keeps some captures, so we can still see what changed for the most recent update.

The most interesting figure that jumped out at me is that the average award is up, from $55,585.54 to $65,006.95

I'll submit an ATIP request to see what interesting information can be dug out (how many of these claims are related to COVID vaccines being the one that springs to my mind. Anyone have other specific suggestions?)