There has to be a reason for subscription, as in continued incremental value, cloud data storage and sharing, some sort of server backend that incurs monthly cost to run and so on.
Just slapping a subscription to a regular desktop application doesn't make sense.
Also, bundling is not a justification for a subscription, especially if one only cares about a single application. This is similar to the music albums which have already undergone unbundling and no one wants to see a bundling where there is no benefit to the consumer.
I'd also recommend Neovim[0], a community-led fork of Vim. One of the focuses of Neovim is exposing an API that lets other apps embed a headless instance in their UIs. This lets you use your Neovim editor along with all its configuration and plugins inside other apps! I've never been satisfied with any of the available vim emulator plugins out there for VSCode, IntelliJ, etc. (there's always something missing!), so this feature has been really valuable to me if I need a better debugging experience but want to use nvim as much as possible.
Someone's already made plugins that embed a Neovim instance right inside VSCode[1] and Sublime Text[2]. There's no emulation here, just straight up neovim running behind those buffers. I've tried VSCode with that plugin (though personally I like to stick to just neovim when I can), and I can say that despite some minor bugs it's been a pretty smooth experience. Even stuff like `gt` to change editor tabs is supported.
How does the Vscode and Neovim thing work though? I've tried to implement something similar for Kakoune and Atom editor, but i noticed that Atom would always manage its own buffer, and Kakoune as well, so i'd have to basically duplicate the buffers - passing tons of data around constantly.
I feel like the keyboard issue is the prime example of why companies shouldn't have monopolies on who runs their OS. Sure on the one hand, they control everything and when things work, they Just Work (TM). On the other hand, they control everything and will inevitably screw something up. Your option? Deal with it, or change laptops. Oh, and change OSes too, since that was your only option for macOS.
I use pass[0], which is essentially just a wrapper on top of Git and GPG. All your secrets are stored in text files that are then encrypted by your GPG key, which is then tracked in a Git repo that you can store anywhere. I use the PassFF extension[1] for Firefox, and Password Store for Android[2]. There are plenty of pass-compatible clients for all platforms and extensions for pass on the first site.
If I need to get my password for eg. GitHub outside of Firefox, I just type `$ pass -c dev/github`, decrypt, and it's in my clipboard for 45 seconds.
I actually do this with the subkeys I put on my Yubikey, and thanks to the NFC capability, I can use it on Android.
I've been wanting to switch to a Yubikey with USB-C, since both my laptop and phone have that port and I don't have to rely on NFC, but this has been working fine so I can't really justify the cost.
Someone has an opinion/solution for the problem of exposing the list of everything you are using a password for? The fact that pass doesn't encrypt that makes me somewhat uncomfortable about hosting the remote git repo on an Internet accessible machine / service. Keeping the data "offline" (if such a thing exists) makes the sync across devices more challenging...
I use gopass a lot, but they direction in which they took the pass API is absolutely horrible. The ammount of irrelevant commands they added made it an UX nightmare.
You make it sound custom, so I wonder if you know that pass's source repository and at least Archlinux's package includes passmenu which lets you access your passwords with dmenu.
> Press a button and the screen would push back and/or buzz
What do you mean by "press"?
"Pressing" a button (ie. pressing down on it) requires a bit more force than just touching it, so this seems like bad UX to me.
Even if you only were to have the device buzz as you touch any interactive control, I feel like that would start to feel real gimmicky real fast. There's probably a reason that technology didn't make it past CES.
Productivity has increased ~250% almost linearly over the past 70 years. Wages increased only ~100% (adjusted) and only until ~50 years ago, basically plateauing since then.
Going to a 4 day work week would bring the weekly productivity to the level of a 5 day week from the early 2000s while also providing a "virtual" wage increase.
This assumes that weekly productivity scales evenly with hours worked. It doesn’t. It will vary per person, profession, etc. but four day work weeks can be surprisingly effective even before considering the effects of potential mental health improvements.
You could also consider 10 hour four day weeks, though I guess at 10 hours many would see fatigue affecting their work.
An episode of Marketplace this week highlighted a recent survey showing a strong majority of people preferring 4x10 rather than 5x8. It's unclear why it isn't more common.
I work 4x10 and keep a timesheet; any time I work over I take back at some point in time. Planning on decreasing the hours slightly soon, drop back to 38 or whatever the norm is here in Australia.
I don't find that 10 hours fatigues me, it feels the same as an eight hour day anyway.
I think you're a big outlier, or at least you would be in the US. I have no idea what work culture is like in Australia, but it stands to reason it would be better than here.
Was this a US specific survey? 10h of work, 1+h of break, and at least 1h of commute means you leave home at 8 in the morning and come back after 8 in the evening. I doubt this would be very popular in Europe with anybody in a non-managerial position or at least very high on the pay grade. But it sort of makes sense because it improves the ratio of productive time (work) to unproductive time (the commute), and results in getting a full extra free day.
On the other hand it's possible that a 10h work day becomes less productive than an 8h one if applied long term. Of course it depends on the actual job but it wouldn't leave much time for anything personal. For a lot of jobs 10h shifts sounds like looking for trouble.
I was pretty great when working a blue-collar industrial job. I'd work 6:30-5 (unpaid half hour lunch) monday-thursday one week, then tuesday-friday the next. So you'd get a four-day weekend every other week, and if there was extra work, you'd get time-and-a-half for it. A ten hour day wasn't bad, because you went until 9, had a half hour break, went to noon, ate lunch, went until 3 and had another break, then closed out the day until 5. It was almost an extended pomodoro.
Even now, programming on salary, I'd rather do 4 tens, 7-5. The overhead of commuting is what kills me, and it'd actually be less traffic if I went in earlier. Besides, those two hours before everybody else gets there could be more productive than the whole rest of the day.
Even a short commute just kills me. This is why WFH is so effective IMO. Up between 5am and 6am. Workout, then work. Those few hours in the morning before anyone else gets in is when I do most of my 'work' for the day. The rest of the day is meetings, answering questions, etc...
Her in Norway most people work 37.5 hours (40 hours present with an unpaid half hour lunch break). And if you are salaried then you can pretty much come and go as you please. I read somewhere (VG perhaps) that about 100k people work a four day week (same total time) here (out of a total 5.5 million population). It's especially attractive for people who have a long or stressful commute.
The calculation was back of the napkin of course, and worst case scenario. But it never implied all people and professions are the same (that's not how statistics work), or that weekly productivity varies with hours worked. Rather with days worked. A day is a fully self-contained unit of work/life so to speak: it contains the commute, the breaks, the work, the recovery time (sleep), etc. This makes it a much more relevant unit of measure.
8 hour days are probably a good balance between how much work you can productively put in during one day and the auxiliary time. Increasing the number of hours per day would probably decrease productivity via fatigue. Decreasing the hours wouldn't bring much of a benefit for the worker since the auxiliary time stays mostly the same (you still commute).
A good chunk of the lost productivity would probably be offset by higher worker satisfaction.
Wages won’t track linearly with productivity because different goods and services have different demands for labour and capital. As we get better at making physical goods prices decrease or quality increases or both. A haircut or a dance recital still takes the same amount of human labour though and it gets more expensive in relative terms. If it didn’t people would leave the service industry for the higher wages available in goods production. And more advanced economies have a higher share of services than less advanced ones. That’s Baumol’s cost disease and it explains huge amounts of the growing relative costs of highly labour intensive services like education or healthcare. See Why Are the Prices So Damned High?, Helland and Tabarrok.
As well as that you have growing land use regulation making certain types of housing illegal, e.g. apartments or houses below a certain area, or flophouses/SROs. Zoning more generally does this everywhere. Most of the residential buildings in Manhattan would be illegal to build today. So land holders capture much of the gains in productivity and that doesn’t show up in the labour share of productivity growth. But land ownership is extremely widespread; this is the lower middle class and up, not just the rich capturing increases in productivity.
Less important but still significant is that due to the US tax code’s treatment of benefits compensation has gone up more than wages have. The cost of health insurance has gone up enormously since the 70’s, A’s has what you get for your money.
I was just pointing out that companies care about productivity and workers about wage. I see no reason for shareholders to be the only ones happy at the end of the day.
No, this does not preserve the history of the repo, just gives a single commit with the existing files. You might also think of this as cloning, then squashing down the entire history (well, everything in master anyways) to a single (root) commit.
I find it funny that Apple themselves mentions that for "Reader" apps users can make purchases outside of the iOS environment, but developers are expressly forbidden from even mentioning this in their app.
What about a physical killswitch goes against Apple's design? The iPhone has always had a ringer/vibrate switch on its side since it was introduced. Obviously it doesn't do anything hardware-wise, but it's still a switch that disables some functionality.
That has a common use case of not wanting your phone ringing in a meeting.
People expect their computers not to listen to them when not actively using the mic. Your switch is the program that records you, a second is redundant and confusing.
Almost all of Apple's security features are invisible and by default and don't get in the way. They're striking the balance between user experience and security.
