https://twitter.com/AlphinEamonn
I'm going to post all my tech predictions there (as long as they're short of course).
I'm going to post all my tech predictions there (as long as they're short of course).
Tuesday, October 22, 2019
Thursday, January 17, 2019
Future Predictions Jan 17, 2019
Cell Phones:
Cameras will move from the top of the screen to behind the screen itself. First dead center, then partway up the screen, to align with your eyes.
Fingerprint scanners will get embedded in the screen, then moved (when people get sick of having fingerprints on the screen all the time) to the back or side of the phone again.
Cell phone cameras will eventually be as good as any DSLR camera, even with optical zoom.
Cell phone cameras will get night-vision/IR/multi-spectrum sensors. Why? Because.
Cell phones will eventually replace keys and wallets entirely.
Cell phone screens will eventually become stretchable/extendable. Holographics will not catch on because they offer no privacy. The device itself will either be stretchable, like rubber, or extendable, but with as few moving parts as possible.
Mesh networks will become more popular for communication between phones, as a way to decrease data usage. First in small ways, like sending messages and emails, then in bigger ways as security improves: entire networks will be formed device to device.
Some company will experiment with cellphone augmented computers. Using your cellphone as a hard drive and processor in a larger screen and keyboard. It’ll be like a mini-tower, holding your desktop Os and files.
Smart Watches:
They will get additional non-invasive health monitors. Something that gathers data from sweat or breath, or both. Blood sugar monitors will definitely get added.
If smart watches get cameras, they will not point up from the face. They will point out from the side of the watch, down towards the wearers knuckes. Same with a flashlight.
Smart watches will replace keys and wallets in the same way phones will.
Smart watch mesh networks will also become more widely used, in the same way as phones.
Medicine
Computers will eventually replace family physicians and surgeons, but it will occur in stages.
First, an AI will work alongside doctors to provide information and diagnoses. A doctor may consult with the AI for additional information, or to confirm a diagnosis. This is normal, and kind of already used as doctors check resources for information.
Eventually, someone will build a booth at can provide diagnoses. It’ll be placed in a doctors office, or pharmacy. A patient goes up to it, inserts their healthcard, and tells it their symptoms. The system will make a diagnosis based on hundreds of thousands of research papers, and print out the result. The patient will take this result to their family physician, and the family physician can verify it, or provide another recommendation. Later, legislation and confidence in the AI will reach a point where the machine will offer over the counter medications, like Tylenol, if that’s all it determines the patient needs. The machine will be build like an ATM so it cannot be easily robbed.
More legislation and research later, the machine will begin suggesting prescription medication and referrals as needed. A clinic visit and family physician may have to sign off on it though.
MORE legislation and research later and the machine will be able to write and sign it’s own prescriptions and make it’s own referrals without any interference from a family physician. At first the patient will have to take the script to a pharmacy, then later (after more legislation), the machine will begin dispensing it’s own medication.
To reach this point, the machine will have to be able to prove a 99% confidence in diagnosing, in addition to an ability to
A) determine the patient that provided the health card IS the owner of the health card, B) determine that ONLY the patient is present and that two or more people aren’t providing different pieces of the data to build a “composite” patient that tricks the machine into giving them a drug.
C) determine that the patient isn’t lying about their symptoms within at least as much confidence as a physician would have when presented with the same patient.
This machine will probably look like a small bathroom. It will take all kinds of measurements: urine-analysis, stool analysis, blood test, breath test, sweat test, retinal scan, heart rate, pulse, speech analysis, ecg, weight, height, body fat percentage, lung capacity, bone density, age, gender, patient history. It’ll use thousands of records of “tells” as well as the biometric data to determine if a patient is lying about their history or symptoms. If the AI believes the patient is lying, they will refer them to a clinic (which the patient can book right from the machine), and send it’s finding to the clinic ahead of time.
People will agree to use the machine because it’ll be MUCH cheaper than visiting a family physician, and it’ll have a much shorter waiting time, won’t require any embarrassing human-human interaction of a clinic, will be MUCH more accurate than a physician, and have a more complete record of the patient. With multi-patient permission, it will even be able to link relative health records (in its database) and make predictions about patient health. Although these machines will probably start off installed in clinics, eventually they will be found all over the place.
As for surgeons, they will get replaced by robots probably much faster than even family physicians will. First they will require the supervision of a surgeon, then eventually they will be able to work on their own.
EMS and nurses will be the slowest to replace, given the nature of their work. First responders work in a wide variety of environments that cannot be predicted, and nurses handle patients while they are awake and sometimes in unpredictable states. They will get replaced eventually, it will just take a bit longer since robots will be involved, not just sensors and AI.
Law
Lawyers will eventually get replaced with search engines and web-available AI. AI will quickly take over the task of providing forms and information. Their ability to scan and search thousands upon thousands of documents very easily and quickly will make lawyers redundant for MANY situations.
The last to go will be defence lawyers, but even they will not last long: judges and juries will be MUCH more inclined to believe impartial, cold, unfeeling robots over emotionally invested, prone-to-lying, humans. They will also gain popularity as they dispense equal justice without taking race/age/religion/sex,etc. Into account, everyone will get treated the same, for the same cases. But they will also be feared as they will probably not be prone to mercy. Stealing food to feed your family will probably get the same sentence as stealing food because you didn’t want to pay. But the developers will probably find a way to take these factors into consideration, eventually.
First, Law AI will be used to consult on cases, like for physicians: for data lookup and course-of-action suggestions based on the letter of the law and previous cases. Then they will provide sentences for judges to consider. Finally, they will provide sentencing on their own. Judges will be brought on by developers to review cases and adjust the code.
Defence lawyers will be replaced, mostly by public demand, since they can’t be bought, and are truly impartial, and for the reasons mentioned above. Eventually, courts will just become the defendant, the accuser, an AI lawyer for each of them, and an AI judge, and will take about 10 minutes. Of course, all 3 AI’s will have lie-detection algorithms in them.
x
Saturday, December 1, 2018
WeightLifts 2.2.2
****** WeightLifts 2.2 and 2.1 were experiencing a bug that caused the app to crash if you selected a workout. This seems to be a result of the Units update not getting installed properly, resulting in no units being added. Without a unit, the Do Exercise screen encounters an unexpected "null" value, which causes the app to crash.
The quick fix is to go to the Settings, then Edit Units, and add a unit.
This will result in another crash (because it turns out I didn't tell the table how to reload properly), but on reopening the app, the unit will be there and the app should work.
I'm uploading version 2.2.2 as I type this which should resolve all these issues.
********** WeightLifts 2.2 +
In this version, I added two big features: syncing workout data with Apple Health, and pairing with a Heart Rate Monitor to...monitor my heart rate....and calculate calories burned during the workout.
This required a bit of learning on my part (thank you RayWenderlich!) about how to share and access data from Apple Health and how to communicate with bluetooth devices.
How to sync with Apple Health
Upon launching the app after downloading/updating it, you should be presented with the system screen requesting permission to access apple health. In order to calculate your calories burned, the app needs access to your sex, weight, and age, but you can choose whether or not to let it write your workouts, heart rate and calories burned to Apple Health.
How to Pair with a Heart Rate Monitor
**I have only been able to test this with a Polar H7 heart rate strap, which it seemed to work rather well with. I'm not sure if it will work with FitBits, wrist-monitors, or the Apple Watch (but I'm hoping to get one so I CAN make it work with it), but please, let me know if it does or doesn't!
When you start the app, it should already be set to search for heart rate monitors that are already paired to your device. By this, I mean that you have already paired the HR device to your phone BEFORE trying to pair it with the app. Then, make sure the app is set to check for heart rate monitors by going to Settings>HR monitor in the app, and ensuring the MeasureHR item is checked.
Next, put on your HR monitor and go to the summary screen. A box will appear beneath the "Start Workout" bar when the app has successfully begun communicating with your monitor.
If it doesn't show up, you may have to shut down and reopen the app.
Once it's paired and the box appears, you'll see your heart rate in BPM in red text. When you start the workout (by pressing on the green Start button and selecting the workout start time), another box should appear that will show your calories burned.
The calories burned are calculated using the formulas below where:
Heart rate is in BPM (taken as the average between two consecutive readings, which are taken every second or so.)
Weight is in Kg
Age is in Years
Sources:
http://www.triathloninspires.com/ti_fitness_and_health/calories-burned-by-heart-rate.html
http://fitnowtraining.com/2012/01/formula-for-calories-burned/
** if anyone can point me to a more accurate formula from some trusted resource, please do! I tested the formula for males and the results I'm getting at the end of my workouts seem to match what I was getting from the Polar app. Don't take these as 100% accurate though: no formula can be 100% accurate as everyone burns calories differently based on more factors than just heart rate.
So every reading of the monitor the app calculates the calories burned using the average heart rate of the two readings, and adds it to the total.
At the end of your workout, pressing the Stop button will bring up a date picker to let you confirm the time you stopped working, then it will present you with a small summary. At this point you can choose to discard the workout data (which means it won't save it to Apple Health, but you'll still have your WeightLifts record of exercises done), or save it to Apple Health.
WeightLifts does not save your hr or calorie data within the app.
The quick fix is to go to the Settings, then Edit Units, and add a unit.
This will result in another crash (because it turns out I didn't tell the table how to reload properly), but on reopening the app, the unit will be there and the app should work.
I'm uploading version 2.2.2 as I type this which should resolve all these issues.
********** WeightLifts 2.2 +
In this version, I added two big features: syncing workout data with Apple Health, and pairing with a Heart Rate Monitor to...monitor my heart rate....and calculate calories burned during the workout.
This required a bit of learning on my part (thank you RayWenderlich!) about how to share and access data from Apple Health and how to communicate with bluetooth devices.
How to sync with Apple Health
Upon launching the app after downloading/updating it, you should be presented with the system screen requesting permission to access apple health. In order to calculate your calories burned, the app needs access to your sex, weight, and age, but you can choose whether or not to let it write your workouts, heart rate and calories burned to Apple Health.
How to Pair with a Heart Rate Monitor
**I have only been able to test this with a Polar H7 heart rate strap, which it seemed to work rather well with. I'm not sure if it will work with FitBits, wrist-monitors, or the Apple Watch (but I'm hoping to get one so I CAN make it work with it), but please, let me know if it does or doesn't!
When you start the app, it should already be set to search for heart rate monitors that are already paired to your device. By this, I mean that you have already paired the HR device to your phone BEFORE trying to pair it with the app. Then, make sure the app is set to check for heart rate monitors by going to Settings>HR monitor in the app, and ensuring the MeasureHR item is checked.
Next, put on your HR monitor and go to the summary screen. A box will appear beneath the "Start Workout" bar when the app has successfully begun communicating with your monitor.
If it doesn't show up, you may have to shut down and reopen the app.
Once it's paired and the box appears, you'll see your heart rate in BPM in red text. When you start the workout (by pressing on the green Start button and selecting the workout start time), another box should appear that will show your calories burned.
The calories burned are calculated using the formulas below where:
Heart rate is in BPM (taken as the average between two consecutive readings, which are taken every second or so.)
Weight is in Kg
Age is in Years
Males: Calories = ((-55.0969 + (0.6309 * Heart Rate) + (0.1988 * Weight) + (0.2017 * Age)) / 4.184) * minutes
Females: Calories = ((-20.4022 + (0.4472 * Heart Rate) - (0.1263 * Weight) + (0.074 * Age)) / 4.184) * minutes Sources:
http://www.triathloninspires.com/ti_fitness_and_health/calories-burned-by-heart-rate.html
http://fitnowtraining.com/2012/01/formula-for-calories-burned/
** if anyone can point me to a more accurate formula from some trusted resource, please do! I tested the formula for males and the results I'm getting at the end of my workouts seem to match what I was getting from the Polar app. Don't take these as 100% accurate though: no formula can be 100% accurate as everyone burns calories differently based on more factors than just heart rate.
So every reading of the monitor the app calculates the calories burned using the average heart rate of the two readings, and adds it to the total.
At the end of your workout, pressing the Stop button will bring up a date picker to let you confirm the time you stopped working, then it will present you with a small summary. At this point you can choose to discard the workout data (which means it won't save it to Apple Health, but you'll still have your WeightLifts record of exercises done), or save it to Apple Health.
WeightLifts does not save your hr or calorie data within the app.
Tuesday, September 25, 2018
New App: Holter Buddy
A holter monitor is a device that records the wearer's heart activity for the period in which it is worn (usually 24 hours). Someone might be advised to wear one if they have cardiac issues and their doctor wants to get a better idea of the patients heart activity. They're about the size of a match box and have about 6 leads coming from them which are placed on various locations on the wearer's torso. While wearing one, the patient has to record (usually on a form given by the lab providing the holter) their activities and symptoms during the time they have the holter on.
Holter Buddy doesn't make wearing the holter monitor any more comfortable, but it it takes the pain out of the note-taking. The user can make a tracker by tapping the + icon in the top right of the screen. They can give the tracker a name, record it's ID# (if the lab assigned it one), and select the date and time at which the test began and will end. The app will also alert the user when the test is complete. With the tracker created, they simply have to tap the big red button and a screen will pop up allowing the user to record what they were doing, at what time they were doing it, and any symptoms they had.
Tapping on a tracker's row will bring up the history screen, where the user can see all the events they have recorded for a tracker. From the history screen they can also print off (or save to pdf via iOS's built in system) a report for the lab.
Back at the home screen, the user can fill in their profile information so it will show up on the report. I've put 3 extra data fields for anything else the user wants to share, like any medications they're on.
The app is available on the app store for free here: Holter Buddy
Holter Buddy doesn't make wearing the holter monitor any more comfortable, but it it takes the pain out of the note-taking. The user can make a tracker by tapping the + icon in the top right of the screen. They can give the tracker a name, record it's ID# (if the lab assigned it one), and select the date and time at which the test began and will end. The app will also alert the user when the test is complete. With the tracker created, they simply have to tap the big red button and a screen will pop up allowing the user to record what they were doing, at what time they were doing it, and any symptoms they had.
Tapping on a tracker's row will bring up the history screen, where the user can see all the events they have recorded for a tracker. From the history screen they can also print off (or save to pdf via iOS's built in system) a report for the lab.
Back at the home screen, the user can fill in their profile information so it will show up on the report. I've put 3 extra data fields for anything else the user wants to share, like any medications they're on.
The app is available on the app store for free here: Holter Buddy
Monday, January 15, 2018
NwHacks 2018
I had the good fortune of being one of the 850 chosen applicants out of 2150 to attend NwHacks 2018, the largest hackathon in western Canada! The event took place on January 13th and 14th and featured coders from as far south as California and across Canada in attendance. We could work in teams of up to 5 people and we had 24 hours to produce a tech solution. I say tech because you could use raspberry pis, VR headsets, Qualcomm Boards, Arduinos, and whatever else you wanted as long as it was electronic and coding was involved.
There was no overall theme, but the sponsors had themes they were offering prizes for. There were tech prizes, and membership prizes, and cash prizes, Nintendo Switches, gift cards, Raspberry Pi's, and lots more! The sponsors had booths set up, gave out swag, one sponsor had a TV with a N64 and Smash Bros hooked up for attendees to play. Scotiabank had a cup stacking game set up and a giant Jenga set. Attendees stayed at the event overnight, and at 3AM they had a Piñata bash. Food was served, mini-contests were held over slack and it was overall a really good time.
You can read more about the contest here, and you can see the submissions here.
I didn't win anything, unfortunately, but I did get some takeaways, mostly as they pertain to competing in hackathons (in no particular order):
Everybody there used some 3rd party library or interface and it made their projects so much better. Learn to use them and use them.
There were a good number of projects there that I have trouble believing were completely assembled in 24 hours. I'm not saying I should have memorized my entire project beforehand then assembled it on site, but I should have had an idea about what I wanted to build and considered the flow of it, what API's I'd want to use, and so on. I went in with only a minor plan (which I didn't use as I ended up following my teams plan anyway).
My team elected to move to a quieter area in another building, and while that was great for focus, I feel like we missed out on the fun and chaos of the main hall. It's packed, yes, but there's so much else going on you really feel like part of the coder community.
Another mistake we made was spending too much time during the planning phase. Nearly the entire first day was spent just coming up with a problem to solve and defining scope. This ties in with the idea to have a plan BEFORE you come, then fully flesh it out on site. I think coding anything before coming is cheating, but coming up with flowcharts and learning HOW to code a project (as in, learning 3rd party APIs, libraries, and languages) is fine. As long as the actual building occurres on-site.
So, for example, one company was giving away prizes for the best Social Hack. That's a theme. Another one was giving away prizes for the best Accessibility Hack, another for best banking app, and so on. Since the hackathon didn't have an overall theme, picking a sponsors theme will help you narrow down your focus.
Unless you're really intent on ensuring you win a prize, I recommend trying something new. I unfortunately used the same knowledge I used in my previous hackathon and feel I missed out because of it. The sponsors at NwHacks held information sessions about using their products and APIs and I wish I attended them. Don't make my mistake, try a new database system, try a new API. You'll come away with new ideas and feeling better about your knowledge base.
Overall it was a really great experience and probably the first time I felt like I really belonged to part of a group. I loaded up on SWAG, participated in the contests, and although I didn't win anything (seriously, nothing, not even a cinnamon bun!), but I definitely want to try again next year.
There was no overall theme, but the sponsors had themes they were offering prizes for. There were tech prizes, and membership prizes, and cash prizes, Nintendo Switches, gift cards, Raspberry Pi's, and lots more! The sponsors had booths set up, gave out swag, one sponsor had a TV with a N64 and Smash Bros hooked up for attendees to play. Scotiabank had a cup stacking game set up and a giant Jenga set. Attendees stayed at the event overnight, and at 3AM they had a Piñata bash. Food was served, mini-contests were held over slack and it was overall a really good time.
You can read more about the contest here, and you can see the submissions here.
I didn't win anything, unfortunately, but I did get some takeaways, mostly as they pertain to competing in hackathons (in no particular order):
1) Use APIs and 3rd party libraries
Everybody there used some 3rd party library or interface and it made their projects so much better. Learn to use them and use them.
2) Have a plan
There were a good number of projects there that I have trouble believing were completely assembled in 24 hours. I'm not saying I should have memorized my entire project beforehand then assembled it on site, but I should have had an idea about what I wanted to build and considered the flow of it, what API's I'd want to use, and so on. I went in with only a minor plan (which I didn't use as I ended up following my teams plan anyway).
3) Sit in the main hall (unless you absolutely can't focus)
My team elected to move to a quieter area in another building, and while that was great for focus, I feel like we missed out on the fun and chaos of the main hall. It's packed, yes, but there's so much else going on you really feel like part of the coder community.
4) Don't spend so much time planning at the event
Another mistake we made was spending too much time during the planning phase. Nearly the entire first day was spent just coming up with a problem to solve and defining scope. This ties in with the idea to have a plan BEFORE you come, then fully flesh it out on site. I think coding anything before coming is cheating, but coming up with flowcharts and learning HOW to code a project (as in, learning 3rd party APIs, libraries, and languages) is fine. As long as the actual building occurres on-site.
5) Choose the prize you want to win and build a project around the theme of that prize.
So, for example, one company was giving away prizes for the best Social Hack. That's a theme. Another one was giving away prizes for the best Accessibility Hack, another for best banking app, and so on. Since the hackathon didn't have an overall theme, picking a sponsors theme will help you narrow down your focus.
6) Try something new
Unless you're really intent on ensuring you win a prize, I recommend trying something new. I unfortunately used the same knowledge I used in my previous hackathon and feel I missed out because of it. The sponsors at NwHacks held information sessions about using their products and APIs and I wish I attended them. Don't make my mistake, try a new database system, try a new API. You'll come away with new ideas and feeling better about your knowledge base.
Overall it was a really great experience and probably the first time I felt like I really belonged to part of a group. I loaded up on SWAG, participated in the contests, and although I didn't win anything (seriously, nothing, not even a cinnamon bun!), but I definitely want to try again next year.
iTunes Apps Back Up
True to their word, Apple got back to me within 3 days and told me to try submitting the tax form again in 2 business days. I did, and it went through and my apps are back up. They're all free for now too!
Monday, January 8, 2018
iTunes Apps Currently Down
Unfortunately, my apps have been pulled from the app store while I sort out the US tax agreement.
I released the apps while I was living in the states, and since moving back to Canada I've had to move my Apple Developer account back to Canada too. This went smoothly except for the US tax agreement, which iTunes isn't letting me complete due to "An account with that name and social security number" already existing.
I'm working to sort it out, and this couldn't have happened at a worse time, but here we are.
I released the apps while I was living in the states, and since moving back to Canada I've had to move my Apple Developer account back to Canada too. This went smoothly except for the US tax agreement, which iTunes isn't letting me complete due to "An account with that name and social security number" already existing.
I'm working to sort it out, and this couldn't have happened at a worse time, but here we are.
Subscribe to:
Posts (Atom)