Google Calendar and Assistant Reminders Will Migrate to Google Tasks Soon
workspaceupdates.googleblog.com
workspaceupdates.googleblog.com
Calendar is not going away, calendar events are not being touched, this has nothing to do with alerts for calendar events, and this does not require you to use a new/separate app.
Google has simply had different things called both "reminders" and "tasks" across apps, for legacy reasons. For a long time there's been a "reminder" object in Calendar (different from events), and then a "task" object got added as well (that shares data with the Tasks app). There's no good UX reason for these to be separate types of objects, so now "reminders" and "tasks" are being consolidated into just "tasks". But they're staying in Calendar. (And can also be edited in the Tasks app which has been around for years already.)
This is good. It's simplifying.
(As to why they're now splitting off reminders created in Google Keep rather than integrating them too, I don't know for sure. But I've never understood the use case of reminders in Keep at all, so I wouldn't be surprised if they deprecate that feature entirely. Keep reminders came before Tasks existed at all, but now you should just use Tasks.)
Crippling Keep is neither good nor simplifying.
If you were using Keep as a kind of reminders/tasks list, you should switch to Tasks for that, since it's much better designed for that use case.
From the product POV, anyone who has ever used knowledge base will tell you that you should keep everything related to your data/ info in a single app. Spreading them out by any categorization is just inviting recall issue later on.
Tasks in GMail (and IIRC also Calendar) existed long before Google Keep. You are talking about the standalone application of Googles task-system which came with version 3(?). If anything, Google Keep is the problem here. Why was it duplicating features which were already existing in the Google-verse? Instead of integrating with the existing systems and improving them.
There is nothing wrong with spreading out your interfaces and workflows, as long as they all work on the same data. No interface is perfect, so having specialized interfaces for specific jobs is beneficial. But Google is not doing this with Keep, instead maintains another parallel system again.
I have a few recurring Keep reminders, which is a document larger than a couple of sentences that contains line breaks for formatting.
Keep reminders intuitively made sense to me. They're sticky notes that you can label/index/search. Reminders allow you to be reminded of a specific note, which I find tremendously useful. Also location based reminders are great, e.g. grocery list that pops up when you're at the store.
"Tasks" suffered from the obvious- they were integrated in Gmail through labs and then calendar widget, but then keep reminders and assistant reminders appeared on the calendar. It was confusing. Keep makes it relatively simple, though you're right, it can be deprecated soon. Still not sure I'm ready to use "Tasks" yet.
Tasks are things that one should do.
A reminder reminds the user of something at a specific time or perhaps at a heuristically chosen convenient time in a range.
So a satellite flying directly overhead or someone’s birthday or simply wanting to be reminded of a fact on a day is not a task. There’s nothing to do, and checking it off makes no sense. (Although wishing someone happy birthday could be a task.). Simply showing the user the reminder accomplishes the goal. On the other hand, most calendar programs struggle with events with no duration, and overlaying reminders onto a schedule makes the schedule harder to read.
I get that sometimes decisions have to be made about whether to keep or abandon design and UI and nomenclature decisions that are now regretted for whatever reason, but bear in mind that you are creating immense frustration for current users when you do so.
As the designer or manager or developer of an application, you are immersed in it and every change makes perfect sense to you. But your users may use it infrequently, or be invested in the way it works now, and all your changes are just seen as pointless, frustrating churn.
A birthday is a repeating calendar event. You're free to create a calendar event with 0:00 duration in Google Calendar, it works fine. You can also create all-day events that show up at the top and don't make the schedule harder to read.
Such an artifact would be useful to users and Google leadership alike.
Their dysfunction is a form of self-defense (while search + ads keeps printing money).
[0] see e.g. Curtis's Hypernormalisation
We're not taking about your child's lemonade stand. We're talking about a corporation that is by all accounts intentional. Why is this seen as "no way"? Doesn't that make it all the more possible?
The obvious answer is “yes. Because that’s what happens when a company has no external pressure and can throw its monopoly weight around into unrelated markets.
If they've fixed that, then I'll be happy. I never did like my phone suddenly yelling at me in otherwise quiet places.
Well, if you set a destination and start the trip, it shows a big X button at the bottom, to stop the trip.
Guess what loads slowly and goes in the location of the big X button?
You guessed right, the Google voice assistant, which I don't use.
They're really pushing the voice assistant.
Star Trek computer has no idea about advertisements/search optimization, etc, it's more like a helpful shell/magic talking box in most cases. If the Star Trek doesn't understand the request or can't complete it as requested, it prompts the user for more details or parameters and lets you know what you didn't give it. Similarly, the way we see interactions with the computer in TNG is that you define a task for the computer and the computer further defines business rules until it can start a workflow. It's not trying to push you towards a specific source of information or a specific outcome, it's just trying to define the rules to take your input and get the desired output.
AI-based control systems like this need to be established like we'd have a library or framework or protocol first, something that is just built for accomplishing tasks and refining that to a point that it is a workhorse. When it comes to subjective matters, it needs to not try to build a profile on what you want and predict it, it needs to be painfully obtuse in its objectivity.
Even on Star Trek, the computer struggled with matters of personal taste/entertainment even though it could ingest complex technical requirements and write equally or more complex workflows/code on the fly for specific tasks. It was pretty often with things like the holodeck or even just playing music/games to pass the time the computer didn't understand what the requestor meant and even though it produced an output, the output had to be refined a lot in fairly specific terms. Even something like shopping was still pretty manual, just piecing through a catalog of items to replicate was seen many times. The point is that the Star Trek computer was optimized for specific workloads and shined there, and stayed out of the way for absolutely everything else. I'm don't have hope that modern tech companies have the discipline to design something like this.
AI and smart homes I think lack a lot of focus and polish, trying to be way too much at once and also return further revenue for the businesses with ad-tech; I don't have any hope that modern tech companies are going to be the smart-home revolution as much as FOSS will, but that of course has the chicken and egg problem of finding hardware that will actually work with FOSS smart-homes; I'll gladly side with Stallman on this one and position that if the entire stack was truly free, then we would see innovation and closer to Star Trek style smart homes since you could buy any smart-object and just run free software on it that meets your needs. But I don't have hope that any tech company has a desire or interest in this, and personally I cannot get invested in anything from the major tech companies as it's just going to be obsolete in a few years and the services will stop working; Google has demonstrated this with far too many projects, and even Apple struggles to get its smart-object integrations to stick (I think one of the major car manufacturers dropped CarPlay for their new vehicles going forward?) It's hard for me to be excited or interested in smart-anything as I don't have the confidence I can rely on it for more than a year or two.
By doing so, they are putting lots of things at risk. Up until the last year Android Auto has been so good that I think most people would rather use it than the in-built system in their car. The last UI update made it kind of frustrating. So frustrating that car companies will get the misguided notion they can do it better.
The same thing was happening to me for a really long time. I migrated to Tasks and it's been working fine since then (mainly because it's a different app). There are some minor quirks with the new app, nothing that bothers me, it's just different. But overall I'd say it's an improvement.
Hopefully the rollout a d migration works out well.
So they didn't finish the User Story?
It is beyond a cliche at this point that Google unceremoniously dumps products and features (RIP Reader and Bookmarks), but when you're creating "a single experience for managing to-dos across Google," why, oh why, wouldn't you migrate reminders from Keep? Did they forget about it until too late?
Still - renaming stuff can be annoying, they renamed google pay to wallet and I only found out when I went to pay for something and it wasn't there.
My calendar events are not tasks. The reminders for them are not tasks either. So this is worrying for me.
Details can be read and implemented following: https://cinaq.com/blog/2022/11/22/the-missing-link-between-r...
Just yesterday someone out of nowhere contacted me while he was trying out my setup. He loved it. Hope this helps others here on HN.
Love the app, on iOS and Mac. Because it’s got a more sophisticated view on my Reminders. In today‘s view it‘ll group the reminders based on the list or other aspects.
At the end you mention the lack of timer based triggers. You can do this with Shortcut personal automations. For example, I have a WFH focus mode. I run an automation at 8am on weekdays that checks if I’m at home and if I am, it turns on WFH mode until late afternoon.
My problem with your setup is that it doesn’t update the events when I modify or complete the reminders.
Regarding trigger i meant there is no standard scheduled system like a cronjob.
The article is meant to inspire others. It solves a very specific problem i have. It’s far from perfect but does the job.
Will take a look at your WFH mode suggestion. Timer/alarm based triggers require manual run confirmation. Allowing non-interactive run is a must in my case to avoid distractions.
Personal automations are pretty close. I use them to run shortcuts at specific times.
If you wanted to run a Shortcut hourly from, say, 7 am to 7 pm, you could manually create those 13 automations or maybe you could use Jellycuts to create one and then cut-and-paste it to make the other 12.
Edit: I just searched for it and here’s an article that talks about a cron-like Shortcut:
https://www.joshholtz.com/blog/2021/06/23/automating-ios-sho...
I haven’t looked at it so I have no way of knowing if it works or if there’s an easier way to do it now (that article is two years old).
Do you have a Mac? If so maybe you could run your shortcut periodically from cron on the Mac.
I just tested and shortcuts can be run from cron (once you get cron working...[1]). For example I have a bunch of shortcuts that do things to my Denon A/V receiver, and added a crontab entry to run the one that mutes the receiver, and it worked. You can invoke shortcuts with the "shortcuts" command. E.g., here was my mute test's crontab line:
*/1 * * * * shortcuts "Denon Mute On"
[1] Apparently sometime around Catalina, when they were beefing up the permissions system and privacy protection cron stopped working unless you gave it permissions. You apparently now have to go into the "Privacy & Security" settings and /usr/sbin/cron full disk access.
I took a look at implementing it a year or so ago and it's a decent amount of work even to just have the non UI parts working (so that checking a task doesn't delete it if recurrence was set elsewhere)
Will update the article soon with this addition.
That’s odd, I have (just double checked) an automation to run a shortcut every day at 0030 on my iPad. Maybe it’s just disabled on iOS?
That has changed it seems. Thanks for pointing out.
Having a "planned" date, where the task visually appears in my calendar, as well as a separate due date would be great too.
The second most useful thing is the warning notice. If someone could reliably figure out how many hours/days/weeks heads-up people are going to need to execute before suspense, that would be amazing. But it's very hard for someone without visibility on your own workload to estimate that. If you're buried under a global crisis, sure people will cut you some slack on your HR actions, but you can bet they'll hold you accountable for budget. An executive assistant can make their career if they can make accurate warning notices for their boss.
I was on the Calendar team at that time, and some project managers enlisted me to do some data analysis to determine if Tasks should be shut down in favor of reminders. At this time, there was no mobile app for Tasks - the only official mobile site was the basic HTML version for pre-iPhone smartphones. There were, however, several third-party mobile apps using the Tasks API. I was able to demonstrate that those apps, plus the official desktop client (inside Gmail), were responsible for a lot of monthly active users. So I guess Google has finally deduplicated by shutting down Reminders instead, which I think was the right decision. It was probably around 5 years ago or so that Google released an official Tasks mobile app.
I mean, if it meant that they started working on interop features and standards again maybe I'd be excited...
Seems like a 7 year old tech still beats Siri and could have a halo effect for android but most Americans have iPhones and can’t use it.
Am I missing something?
It sounds like basic calendar events will be unchanged by this (though I'm not completely clear on that), but Google Calendar is the only Google service I'm using regularly and I wouldn't mind getting off it.
Such a basic productivity area has languished for a decade ...
I migrated emails to Fastmail a few years ago, but have been holding on to Google Calendar.
> Android does not support CalDAV natively, however you can easily add support by installing a CalDAV sync adapter. We have tested and recommend DAVx⁵, which is approximately $5. Once you have added your account in DAVx⁵, you can use the native Android calendar or any other calendar app and the changes will sync with Fastmail.
https://www.fastmail.help/hc/en-us/articles/1500000279881-Se...
It also covers contacts integration too!
* https://techcommunity.microsoft.com/t5/microsoft-to-do-blog/...