Your approach mirrors mine but what is Step Two?
Itās a MFA code generating app by Neil Sardesai. I like it because itās a one-time purchase, built as a passion project by a member of the Mac community.
Step Two: https://steptwo.app
Neil Sardesai: x.com
As one reviewer on the Mac App Store noted, Step Two doesnāt include end-to-end encryption, so it might be a good idea to avoid storing oneās most important TOTPs in it.
There are plenty of excellent free authentication apps from major software companies. I used the Google Authenticator until 1Password added this feature.
Yeah, Google Authenticator is OK unless/until you have to change phones, in which case things get ugly, as noted in this Ars Technica article from 2020: Choosing 2FA authenticator apps can be hard. Ars did it so you donāt have to | Ars Technica
Yep, thatās one reason I switched to 1PW. I couldnāt remember when that feature was added so I asked Google and found this:
You think this TJ Luoma guy knows anything about tech?
Point taken but Iāve seen plenty of similar unfavorable comments elsewhere about Google Authenticator, especially concerning the differences between its Android version and its iOS version. Of course, one would expect Google to favor the Android version and thatās all well and good, but I would guess most of us here are iOS users and, soā¦
[Made this reply before you added the MacStories link, of coiurse.]
The Wirecutter article says "Google Authenticator helped create the standard of two-factor authentication, but it has lagged behind other authenticators in basic ways.ā I mentioned GA as a free app that served me well until, as it turns out, 2015 when 1PW added that function.
I posted the macstories article as a tip of the hat to @tjluoma
If it is elitist to want an app on my platform of choice to follow the UX standards of that platform, then I am totally an elitist and very proud of it. I use Apple devices for a number of specific reasons and one of those reasons is my expectation that the UX will be high quality and consistent across apps. Every Electron app Iāve ever used sacrifices this aspect of UX that to me is very important for cross-platform portability. This harms not only my enjoyment of using these apps but also my productivity because of the way shortcuts, gestures, and other structural elements tend to follow their own logic rather than the logic found among native apps. Sometimes itās even worse, like not following audio capture standards or other critical platform integrations. Iām currently experiencing this right now with a particular app that I canāt get around using and itās extremely frustrating from a user perspective.
This harms not only my enjoyment of using these apps but also my productivity because of the way shortcuts, gestures, and other structural elements tend to follow their own logic rather than the logic found among native apps. Sometimes itās even worse, like not following audio capture standards or other critical platform integrations.
In particular, Electron apps that refuse to use the Mac menu ā looking at you, Bitwarden ā frustrate me to no end.
EDIT: Well, guess they heard me; the new version released today fixes that very issue. The Bitwarden Mac app now uses the Mac menu. I could swear it did once before, but wasnāt doing so recently.
EDIT #2: I see now what was causing this. Hereās the deal: if you click the red ācloseā button and then click the dock icon to re-awaken Bitwarden, it comes back without the Mac menu working. However, if you bring it back by clicking the Bitwarden icon in the menu bar (assuming yours is set that way), that wakes up the Mac menu. Hmm.
And itās one thing if they donāt follow the platform integrations but can therefore provide a user-facing benefit. āWe completely re-wrote (some internal OS thing) ourselves, and it gives you 5x faster performance.ā Thatās cool.
But when itās more like āyou canāt use Shortcuts because we wanted to save some moneyā, thatās not as cool.
Or like Discord: we just donāt respect sound handling nor DnD because, meh. Itās not even something practical like money, itās just a complete lack of consideration for being a good member of the OS community because Mac users are a minority among the gamers that the platform was primarily built to support.
I think Iād still file that under āsave some moneyā as theyāre using Electron and supporting a platform-specific feature like DnD takes development time / effort. But it really does feel like if they wanted to be a good platform citizen, DnD should be pretty low-hanging fruit.
Iām wondering if itās feasible and worthwhile to contribute to the Mac build of Electron. It seems to me that might be more constructive than complaining about the general quality of Electron apps.
But, being a techie, I recognise there might be inherent limitations - such as the lack of native controls to automate.
Is the problem that Electron isnāt well-implemented on Mac? Or is the problem that it basically relies on the underlying browser engine?
Itās good that you are willing to help. Electron and similar apps, such as the one Microsoft is developing, are quickly becoming the preferred framework for cross platform applications.
1Password published their electron-hardener on GitHub and hopefully big companies like Slack will also give back to the project and support it.
The main problem I see with electron is that when you build an app with AppKit you get a ton of built-in functionality āfor freeā, functionality that is replicated across the entire macOS system. The way keyboard shortcuts work, how drag and drop is implemented (mostly), what menus to expect (services?), how the menus look and behave (see 1Passwordās menu thatās constrained to the window, not right), and so on.
All of these things can be implemented with another framework, but AppKitās got 20+ years of development behind it, so itās going to take any competing framework a long time to catch up. Those established behaviors are what make a Mac feel like a Mac. AppKit can be time consuming to work with, but personally I enjoy it, and I also enjoy knowing that any app I write in it will inherit the new shiny Apple releases mostly for free, without me having to do much.
But, like I said, Electron or any other framework can, theoretically, get pretty close to emulating what AppKit does, but since it will take so much work, and since most companies using Electron are doing it for Windows, I donāt see it happening.
Is there an issue with a forked codebase as well?
I.e. letās say we wave our magic wand and somehow we have all the AppKit stuff (either exposed directly or emulated in Electron) available on Mac.
Because the Mac UI and the Windows UI arenāt the same, it seems that being a proper āplatform citizenā on both platforms would (effectively) still require a fair bit of extra work to develop multiple UIs, wouldnāt it?
Thatās correct. The reason choosing something like electron is to build something once which looks and works the same on every platform reducing development and design effort. Doing platform specific UI decision and implementation contradicts this in my opinion.
That is nearly the same as using iOS for a desktop computer or tablets. Microsoft still tries that with windows. Apple knows for different use cases and hardware you need different designs and interaction capabilities, so there is now iPadOS etc.
Electron is good but all the apps do not āfeelā right for me. Same goes for Java applications.
Iām curious and Iām NOT being snarky, those who are unhappy with 1Password going Electron (Iām not concerned about that, only the subscription) are you using Obsidian and if so, why is one ok and the other not?