Copy Clean Link
Posted on: May 8, 2024
I have been using the Brave browser for a while now. It is built on Chromium but has more privacy protections, e.g., ad blockers and trackers are enabled by default.
One minor but important feature on the privacy front is the ability to Copy Clean Link when right-clicking a URL or hyperlink.
For example, a YouTube video I reflected on yesterday would create a “share in post link” of:
https://youtu.be/u5mNa6KE0lA?si=p17rLDuNq6TExa4AHowever, Brave will take out the unnecessary tracking bits and give you:
https://youtu.be/u5mNa6KE0lAThis might seem like an insignificant feature. It is not. The “?si=…” bit is a tracking parameter. Other media and social sites do this and the URL extensions can be ridiculously long.
Brave’s Copy Clean Link affordance shows respect to the user. I want to share a resource, not necessarily who I am, what I used, or in which context. It is small touches like this that add up.
P&P
Posted on: May 7, 2024
That is my shorthand for preparation and practice.
I watched this five-year-old and hour-long Christmas lecture by Dr Hannah Fry and was impressed by how she captivated her young audience.
Fry is a regular guest speaker at the Royal Institute and her audiences are typically adults who probably have an interest in her topics. But to illustrate the presence and importance of mathematics in daily life to her younger audience, Fry and her collaborators must have prepared and practiced many times to get things right.
I used to regularly orchestrate and facilitate events, courses, and workshops. So I noticed details such as the sequencing of events and tasks, and the positioning and movement of cameras, people, and objects.
The P&P would also have revealed flaws in their initial plans and they would have made changes to fit the space, time, and outcomes.
Fry and her team definitely prepared and practiced more than the generative artificial intelligence (GenAI) demonstrator on the BBC [my APP rant]. If something looks easy, seamless, and effortless, it is probably because of P&P.
Hurry, don’t rush
Posted on: May 6, 2024
I concur with the wisdom of Seth Godin that it is far better to hurry because you know that time is precious than to rush because you are impatient or indifferent.
There is another difference between the two concepts in practice. Rushing tends to happen at deadlines. Hurrying can happen well before that.
The difference each makes to processes and products is down to planning, preparation, and communication. Do these three well and you can hurry to do good work. Fail on any and you will rush for the wrong reasons.
Ping-pong junk mail
Posted on: May 5, 2024
I have noticed my email accounts playing ping-pong with junk mail.
I have several email accounts, but use just two of them regularly. The first is my main one for work and official contact. The second is for signing up for services that I test or am not sure about.
I used to get junk mail in my first account and that is why I created the second. As I removed my main email from mailing lists and spam, the junk mail “pinged” from the first account to the second. My guess is that unscrupulous and leaky services put me on mailing lists or passed my email address to third parties.
Then I created filters, reported junk, and removed myself from as much spam as possible in the second account and it became quite clean. The email provider probably clamped down on junk and spam.
But for reasons I cannot fathom, the junk and spam “ponged” its way back to my first account. This might be due to leaky or hacked databases.
So I persisted with my usual junk and spam extermination. I get very little of that now, but it feels like the little that trickles through seems to ping-pong between my two accounts now. Weird!
GenAI replaces tasks, not jobs
Posted on: May 4, 2024
A common and lazy theme propagated by popular media is that AI, particularly generative artificial intelligence (GenAI) now, will take our jobs. They ignore a recurring lesson from history: Technology will take some jobs, but it will also expand current job scopes and create new ones.
After listening to an archived episode of The Curious Cases of Rutherford and Fry, AI in the Economy, I have one more argument that previously escaped me. Various forms of AI free us from many (often tedious) tasks, but this does not mean that it will take all our jobs.
If one of the tasks in your job is to do any sort of writing, GenAI can suggest an initial draft, provide structure for it, and/or check it for errors. It is unlikely that the task is your job. If the job is that narrow, it might deserve to be taken over by AI. If not, you have other non-AI tasks that justify your work.
There was a saying among some pro-technology educators who countered technology luddites and naysayers — teachers who said they could be replaced by computers should be. Now we might say the same about teachers who worry that they will lose their jobs to tireless GenAI. If these teachers focus on the tasks that GenAI can do better than them, these teachers are not doing their full jobs.
BTS biker vs falcon
Posted on: May 3, 2024
This video and its behind-the-scenes (BTS) partner showcased some of things that appeal to me.
The video was a contest between a downhill bike champion and a peregrine falcon. The biker had to race to the bottom with a lure on his back while the falcon had to retrieve the lure before the finish line.
It was a thrilling chase and the bird won all but one of the races (as the BTS video showed).
The tagging and tracing of the falcon reminded me of a project I worked on when I was a biology undergraduate. I had to take care of and track white-shouldered kites. Raptors are magnificent birds!
The BTS video provided some clues on how much effort it took to make the first video. There was a weather delay, it took several runs, and the falcon even went AWOL once.
Every final product should be accompanied at least one process artefact. You appreciate the product even more and learn from both more deeply.
I have noticed that recent Attenborough docuseries with the BBC end with select looks at the end of each episode. Some series also link to OpenU resources.
The YouTubers and documentarians understand that it is not enough to depend on a well made product to teach. We also need reflective pieces to learn.
GenAI: Cauchy Cauchy Cauchy?
Posted on: May 2, 2024
This image appeared in my Twitter timeline. It was a supposed car seat strap for kids between the ages of 1 to 10.
A reverse image search provided many returns, but I could not confirm if it actually existed, and if it did, when it was available and who offered it.
So I prompted CoPilot and Gemini since both the free versions could take images as inputs. This was my initial prompt:
I am going to upload a photo of a supposed product from the past. Did that the product actually exist? If so, provide evidence that it did. Do not include websites that do not indicate the source of the photo, where it was published, and when it was published. Do you understand?
I then provided the image. I copied and pasted the image in CoPilot, and had to upload the image in Gemini.
This was CoPilot’s response:
Gemini had a confusing “Cauchy” response in its first draft.
Its other two drafts simply summarised the article that accompanied the image and suggested I search “historical Sears catalogs online” or “academic journals or articles on child safety products from the 1960s and 1970s”.
The Gemini response was not helpful. While CoPilot’s seemed better, the only new and useful information was in all of paragraph 2 and one sentence in paragraph 1. The rest of the content was not relevant. The important information was also from just one reference, a vintage website that choose not to reveal who exactly runs it.
That said, CoPilot’s return passed the initial smell test. The chatbot made it easy to share the image via pasting, it provided a believable origin of the product, and it did not cauchy, cauchy, cauchy.
Peripheral and central work
Posted on: May 1, 2024
It is Labour Day here in Singapore. Even though it is a public holiday, I shall be working on a project — home improvement of my dad’s place.
I am resuming the work that I paused because of actual work. Most of the DIY will involve patching and painting walls that have been ignored for about 40 years.
As I have reflected previously, this will involve a fair amount of preparatory and followup work, i.e., moving furniture and objects, cleaning surfaces, sorting, disposing, recycling, restoring, etc.
The peripheral work can take as much time (if not more) than the actual work. But the former also enables the latter. Ignore the first and you cannot do the second. This is a principle that applies to both DIY and professional work.
Overcoming FOMO on GenAI
Posted on: April 30, 2024
After reading this Wired writer’s experiences with generative artificial intelligence (GenAI), I concluded with a basic strategy to overcome the blind desire to try everything.
The writer tested six different types of GenAI largely because she felt the pressure from her community. She gave in to the fear of missing out (FOMO) even though she did not have meaningful use for most of them.
She found some assistive and productive use of Canva and ChatGPT, but she was blindsided by or disappointed with Aragon.AI, AdCreativeAI, Midjourney, and Otter.AI. Her meaningful use was linked to her daily drudge work. The rest were comparatively frivolous or peripheral.
One way to overcome FOMO is to consider your context. If you do not need to create video from text prompts or speak in another language with your voice, you do not need to use those forms of GenAI. Interesting as those tools might be, they are not relevant to your context. Instead of FOMO, consider the consequences of wasting time, effort, and money.
Anticipate, prepare, practice
Posted on: April 29, 2024
I could summarise what this video report on generative artificial intelligence (GenAI) was about. But I will focus on the need to APP — anticipate, prepare, practice — instead.
For some context, the demonstrator of a chatbot that actually chats back could not get it to speak the previous time he was on air. The issue? He forgot to unmute his Mac.
This time he made sure to unmute the machine and raise its volume. But he failed to consider that the interviewer would use a different accent from his and he did not position his computer near to the interviewer.
The result was that the internal microphone of his Mac could not pick up the voice inputs optimally. The chatbot also mistook “avatar” for “guitar”.
This was an ICT lesson I used to teach 20 years ago. Before using any technology in the classroom:
- anticipate issues that might occur
- prepare for them and plan contingencies, and
- practice using the technology in the actual or closely simulated context
This was for basic, teacher-centric use of technology, e.g., demonstration or show-and-tell. If you are planning for student-centric integration of technology, the preparation and processes are even more involved.
The adage is: If you fail to plan, you plan to fail. The chatbot guy failed twice in a high stakes demonstration. Teachers and teachers-to-be should learn from his mistakes.
Another dot in the blogosphere? 