I recently visited my friend from our programming club who saved these letters and was reminded a couple people wrote additional angry letters years after paying for the shareware. They demanded support in return for their one payment. (Of course, not only had I moved on to other projects, but I had long sold the type of computer the shareware was created for!)

So I learned early on that people unreasonably expect support for no additional cost. Or they believe the amount they paid is for support in the future, not work done in the past. It doesn't work that way economically. A constant flow of additional money has to come in the door to pay a team to do the actual support.

AFAICT, the only feasible models for supported software seem to be subscription, microtransaction or advertising. Any one-time-up-front price means it's abandonware. Which is fine for some types of software, of course, but probably not as often as users expect support.

A better solution honestly is just to accept some percentage of people will be little assholes but hopefully, as you remembered before your friend brought it up, that most people are gracious and kind. Focus on those people.

If I buy a piece of software, it's not unreasonable to expect it to work for some period of time after the purchase. This is especially true on the Windows side, where Microsoft has gone to great lengths to keep old APIs around and support most (but not all) old software within reason. I have engineering software programs that are a decade old that still run fine on my Windows 10 machine.

macOS has been less shy about deprecating old APIs and forcing software updates. I probably spend $500-1000 every year just upgrading a certain few software packages that charge for a new version every time a new macOS comes out, and I hate it. I don't mind paying for new versions of software, but it's becoming saddening to watch all of my macOS software rapidly decay away with each macOS upgrade unless I buy the newest version.

"it works" and "the developer gives me support" are two different things. In this case, I'm sure the shareware he wrote still "worked," but clearly they thought they were entitled to perpetual updates or the ability to chat to the developer any time they like.

As far as I'm concerned, the SLA of $10 shareware I volunteered to pay for is "whatever the developer is willing and able to provide." It's $10. ¯\_(ツ)_/¯

Fwiw, 1Password isn't nuking 1P7 or existing local vaults. Those users are free to keep using v7 for as long as Apple or Microsoft allow the app to be installed on the OS. (And I do agree, macOS and iOS are both quite abrupt about cutting off support and I commend Windows for going to great lengths to avoid it)

One could argue that.... but one would be silly to do so.

Why does a password manager need a subscription? My password should never touch a 3rd party’s server, I don’t need extra features, I don’t need a login, I don’t need long-term or even short-term support.

What ongoing development does a password manager have? Is it that buggy from the get-go to need constant updates?

However, in the case of 1Password it's not just about being a password manager; it's also about syncing passwords between devices, staying on top of (sometimes rapidly!) changing standards between devices and browser extensions, and being aware of the evolving landscape of best security practices.

This worked perfectly fine until they “fixed” it by removing Dropbox support, and now are apparently removing iCloud sync too

Way too often, automatic upgrades silently break my existing software, take away functionality or introduce new bugs.

Anyway thanks for showing me KeePassXC, looks like something I’ll be very interested in

I shutter to think about just trying to keep the status quo across all those platforms, devices, browser extensions. How often do we have new versions of iOS or macOS? What about Windows, Android, Linux, Chrome OS?

I can think of a TON of work without adding a single feature.

And the fact that we need to constantly change stuff just for the sake of change is the problem.

To add to this: I’m perfectly happy to pay a premium price for software when I need a new version. I did this with Photoshop for years, $399 per copy was perfectly fine for me because I spent that money every 3 or 4 years.

But you take all those platforms and all those extensions and the result is maintenance by itself is constant work. I'm not talking about change for the sake of change. I'm talking about the work just to keep the features that you have.

I have personal projects I'm working on where it feels like all I have time for is just keeping up with security updates, Ubuntu versions, DB version upgrades. Work projects are even worse with SOC2 requirements and the endless stream of CVEs.

There's no way a password manager can do nothing for 3-4 years ignoring security vulnerabilities in their dependencies that need to be patched.

It comes down to what value it provides to me for what I need it to do, which is store and retrieve passwords for me, and sync via wlan. That’s it. Why should I continually pay for Android, Ubuntu or Windows development when I don’t use their app on either of these platforms?

And additionally, as a consumer it’s not my responsibility to find a way for a company to fund its product. Saying “development goes on even without you upgrading, so you have to pay a subscription to support that” is kind of a weird argument, isn’t it. Imagine if you had to pay a subscription for using a car because next year a new model will require development and therefore you need to pay for it.

It does not, that is why there is KeePassXC and other alternatives.

I've been using 1Password since 2008 and I'll be doing the same thing. I have tolerated the UI regressions and even subscription with version 7, but Electron is just unacceptable for what was once an amazing Mac app that put Apple's apps to shame.

I got rid of two of them — Evernote for exactly the same reason.

1pwd going the Electron route would have put me close, but not over the edge — but the fact that they seem to have willy-nilly removed local vaults does.

These days it sometimes feels like one has to write all essential software oneself. Or go back to DOS and plain text files.

They have been talking about this for years. It was not done willy-nilly. There are limits to local vaults.

They're a business, and 97% of their users were already on the subscription model. If you were running a business, and 97% of your users had abandoned a feature that was a headache for you to maintain, would you stubbornly keep wasting time and money on it?

Bitwarden is right there waiting for you.

The only wish I had over the years(and I would have gladly paid for an additional license): a linux client, even CLI would have been fine. I'm not sure which year they started, but what they pushed instead was the announcement of online sync and other things I did not care about, at some point it became hard to even find the regular version on their website.

This was when I slowly started moving to alternatives. With the move to electron and this anouncement here I'm happy that I moved away a few years ago and their clear signal that I am not the kind of customer they have any interest in anymore.

The costs of developing and maintaining software are recurring -- especially for security-critical software. Subscription business models align incentives towards ongoing maintenance.

Whom would you rely on to handle something that matters to you:

A. Someone who refuses to eat until they start collapsing from hunger?

B. Someone who eats regular meals?

if an app has ongoing development that you benefit from, it seems entirely fair to pay a subscription. the fair alternative is a one time payment for a lifetime license with few patches priced in. I would probably prefer the latter for something like a word processor, but would you really want to use the same version of a security-critical program like a password manager for the rest of your life? if not, how do you expect them to fund development/maintenance indefinitely?

That last bit I don't believe applies to 1Password, because there are certain things you can't do without some kind of centralization, and the article makes that case.

...but look at something like Adobe CC, what exactly does moving from a purchase to a subscription benefit me? And let's not forget about the more subtle effects, like losing the right of first sale, silent T&C changes, mandatory updates, etc - things that are only to the vendor's benefit.

It keeps the vendor financially healthy, stable and willing to keep developing the stuff you use.

Would you want to dedicate your work into a product for meagre & sporadic standalone payments pressuring you to endlessly churn out marketable feature upgrades with little time for maintenance work just to barely make ends meet? No? Then why on earth do you expect other software vendors to do that?

Even Bitwarden pushes subscriptions.

I'd love to see how many developers who complain on HN about subscriptions actually make a living primarily from selling standalone software to consumers.

This model is simply more lucrative and strips customers of a number of their rights which are inconvenient for corporations.

Trust that the company which provides something you need is far more likely to continue to be around. Suppose you tell your employer that you're taking a 1-week vacation. How does it benefit them for you to do that?

Agreed. Though, this latest development – the move to Electron – is a negative for me, so it leaves a part of me wondering what kind of development I have been paying for. I imagine I am not alone with such a sense of disappointment.

I don't _want_ the ongoing development. Photoshop from 5 years ago is perfectly fine for me. Same with Lightroom, etc. I mean, I've only paid Apple _once_ for Logic Pro and have been getting upgrade after upgrade for no cost -- a nice bonus, but I'd be perfectly happy if logic's code had been set in stone at the moment of purchase, too.

The consumer-friendly option is to let the consumer decide if they want upgrades. Or for a security-focused app like 1password I'd have believed something like "we don't want to be responsible for security problems if you decide not to upgrade, thus you must buy a yearly license" but that wasn't the message at all.

Why did 97% buy the subscription? Because they hid the other version in a locked filing cabinet in a basement with a broken staircase.

The old business model works but you have to keep innovating and diversify your product line. Microsoft was the best example with things like Encarta, Age of Empires and other tools like Project.

This new trend of doing these apps once, with far easier programming languages in a connected environment with plenty of docs, crash report data and things like stack overflow really makes it look that we are talking about cheap people trying to make a quick profit not unlike those free to play games.

My take on the hatred boils down to these things:

Utility type applications built as a service offer an inexpensive purchase of some kind, or are AD driven, etc... Then, features are changed, roadblocks added, user experience degraded to create problems that subscriptions pay to remedy, often poorly.

Noisy subscriptions. It's not enough to send a few bucks a month. ADS, various pitches, in app sales happening, all contribute to what might otherwise be a simple, worthy experience and solution.

It can be hard to cancel. -->if this happens to someone even once, the hate can be visceral afterword. Everyone else, no matter how well they do business is impacted to a degree.

Subscription apps / services going away on short notice. What were people paying for?

This has nothing to do with VC and everything with trying to build a better product for users that is easier to work on for developers.

As a long-time 1password shill I have hit my limit and will slowly start migrating to BitWarden and iCloud Keychain.

Bitwarden's desktop app is built on Electron..

Our decision to built the macOS app in Electron was absolutely not driven by VC money. For the past few years, we've been working on consolidating 1Password's business logic into a single Rust-powered core that could be shared across all our apps. This has many advantages: feature consistency across platforms, faster development cycles, and better security. When building the front-end for the desktop platforms that would take advantage of this new core, Electron suited us perfectly, since we could write our UI code once and make it consistent across Linux, Windows, and Mac. We actually did build a native Mac app initially alongside the cross-platform Electron app, but we eventually decided that having two separate versions of the macOS app (one in Electron, one in SwiftUI) would cause a lot of needless development churn and hassle for both customers and our support team.

I can understand your frustration about Electron, but I hope you find my explanation reasonable. Please stop spreading misinformation.

I can absolutely attest to that with a relatively underpowered computer (4 gb of RAM). I can barely use 2 electron apps after which my computer grinds to a crawl (I’m running VSCode and Slack mostly). I have stopped using the discord desktop app and exclusively use the website now.

That AgileBits has been doing everything it can to force people to the subscription model and that this push to subscriptions very coincidentally lines up with two rounds of VC investment for over $300M over the past couple of years? No, that is not misinformation.

It may have been easier for the dev team to use Electron as their cross-platform toolkit, it is not easier for the users to put up with the attendant bloat and reduced performance.

The ones who should stop spreading misinformation regarding the forced subscription all seem to be working for AgileBits.

There's a reason people prefer one over the others. You can't have one front-end for all these different platforms. Well, you can, but then it's a compromise for at least 2 out of these 3 platforms.

Even Microsoft has "Office 365 for Mac".

That doesn't make the macOS experience better, it makes it worse.

You can create a shortcut in the iOS shortcuts app to open the Passwords area of Settings via an icon on your Home Screen. Just open the following URL in the shortcut:

prefs:root=PASSWORDS

Best tip I have for you around iCloud Keychain right there.

For anyone who doesnt use shortcuts often, what you need to do in Shortcuts is:

1. Make a ‘URL’ action to prefs:root=PASSWORDS 2. Hit the ‘+’ and make a ‘Open URL’ action from safari.

Save, add to Home Screen, and you’re done.

But that doesn’t make VS Code “shit”. It’s pretty much the shining star of Electron done right.

that is a very low bar. VS code is still slow and eats up a ton of resources. not to mention I don't trust anything from microsoft. OP said electron = bad and you should be ashamed of using it because its helping propagate it's usage when its a cancer.

But now with Electron, which I don't like for the same reasons, as a friend once told me, allowed me to have some of my favorite apps running in Windows, Linux and Mac almost flawlessly with a good interface that finally the promise of Java was fully realized.

So while I would like people to follow more the Sublime Text approach, there is value in these Javascript based apps that lower the barrier of entry, provide widespread availability and are definitely easier to debug. Also sometimes I don't get to decide, since my org for instance makes it extremely convenient to stick with JetBrains stuff.

But hey, I'm the type of person that considered a Gentoo machine running Fluxbox far more useful than the very polished MacOS.

Electron is a tool. It can be used well. It can be used poorly. Any tool shares the same issues. I can write a shit native app and I can write a shit Electron app.

All Electron does is lower the barrier of entry to making an app and making it work cross-platform.

VS Code is an example of Electron being "used well". I still find myself using other apps because they're more responsive. That tells me that Electron is inherently making the app experience worse, despite being used well

I understand the 'why' of electron, I am just not happy with the results at all.

Though maybe the world is better if that barrier of entry isn't lowered, I'm not sure what's better- a shitty app available everywhere or a good app available only on a couple of platforms...

Someone wants to build a 5 minute app for themselves and Electron happens to be the easiest way to do it? Go for it, there's nothing stopping them. If that app happens to be useful enough for others to use it, even better, that person just solved what could have potentially been a big deal for that user.

If the people using the app are content with the features/quality and the resources it uses, why does it matter so much, especially to a third party like armchair engineers on HN, if it was built natively, on Electron, or CrappierFrameworkThatWillEventuallyReplaceElectron?

If the app isn't usable by you (and this is a general you, not specifically directed at you), then the answer is simple: don't use it. Nobody is forcing you to use Electron or any app built with it if you don't want to.

If as a user your needs are not met, whether that's due to sluggishness from Electron, incompatibility from having a native-only solution, whathaveyou, then all you really have to do is wait. A competitor will come and take its place eventually; that's what the market is there for.

So far, it seems like VS Code is more than meeting its users' needs, but like I said above, other alternatives exist and will continue to exist, and they're all great if VS Code doesn't work for your particular use.

> If as a user your needs are not met, whether that's due to sluggishness from Electron, incompatibility from having a native-only solution, whathaveyou, then all you really have to do is wait. A competitor will come and take its place eventually; that's what the market is there for.

not the first time we've heard this argument used in different industries. I'm old enough to remember this argument about DLC's, DRMs and games being released before they are finished and get charged DLC packs to finish it. This argument falls apart when everyone starts doing it due to economic factors and complacent users/customers who blindly use what everyone else is using.

Businesses aren't stupid, if a decision were to actively lose them more customers than it'd gain, they wouldn't do it. If every single business and OSS alternative suddenly switched to Electron and you had no other choice, then maybe it's you that is wrong about the value of Electron.

But, of course, that's a ridiculous hypothetical not really grounded in reality. In the real world, plenty of alternatives exist for practically every Electron app out there so if you don't want to use it, you don't have to.

And for every one of you, there's hundreds of people that think their Electron apps are fine and appreciate that they can use the exact same interface regardless of which computer they're using.

i see you're moving the goal post

once you have a monopoly you can do whatever you want, thus why everyone keeps talking about EEE. look at what apple is doing now with photoscanning. What they are doing is inherently bad but they are able ignore everyone and go ahead with it because their "target audience" doesn't care about this kinda stuff. Doesn't mean the rest of us shouldn't actively fight against it. your argument basically boils down to let the market and customers decide for itself. Over here there are some of us trying to tell people to not support electron because of the potential consequences, and the downward trend of the web as we see it.

There's choice with password managers. But look at music. You have Deezer and Spotify. The only two free services that I'm aware of that only have Electron apps on desktop. (YT Music doesn't have an app on desktop). Spotify has exclusive content. You don't have choice when apps are the same as services. Hate the Hulu interface? You can't watch The Handmaid's Tale on Netflix. It's that kind of thing.

Over the past few years, we've been working on consolidating 1Password's business logic into a single Rust-powered core that could be shared across all our apps. This has many advantages: feature consistency across platforms, faster development cycles, and better security. When building the front-end for the desktop platforms that would take advantage of this new core, Electron suited us perfectly, since we could write our UI code once and make it consistent across Linux, Windows, and Mac. We actually did build a native Mac app initially alongside the cross-platform Electron app, but we eventually decided that having two separate versions of the macOS app (one in Electron, one in SwiftUI) would cause a lot of needless development churn and hassle for both customers and our support team.

I can understand your frustrations about Electron and our subscription-based model, but I hope you find my explanation reasonable. Please stop spreading misinformation.

Can you quantify the "needless development churn and hassle for both customers and our support team" in some way? Presumably, 1Password 7 and its ancestors used native macOS APIs, which meant some degree of that given you had to do something different on Windows and/or Linux. I don't know what your support team has had to endure, but as a long-time sample size of 1, I've been incredibly satisfied with the way you've designed and engineered the macOS application (and the iOS app too!) to date; I'd be hopeful that whatever tradeoffs y'all will be making moving to Electron, the "native" feel of the macOS client wouldn't be sacrificed. Is there anything you can speak to there that I should prepare for with 1Password 8?

Sure, happy to elaborate on that! Since we were rebuilding our app from the ground up, it was a significant slow-down on development to create a user interface for both Electron and SwiftUI, requiring two separate teams of platform developers for every feature we needed to implement. There were also concerns by the documentation and support teams that we would need two separate sets of instructions for many common tasks, due to small differences in layout and look between the applications. Eventually, we had to make the tough decision to focus on a single common framework for desktop. This will allow us to ship features across every single platform far quicker than we could before.

> I'd be hopeful that whatever tradeoffs y'all will be making moving to Electron, the "native" feel of the macOS client wouldn't be sacrificed.

We've tried our very best to keep the experience the same so that the transition from 7 to 8 is smooth, and from my point of view 1Password 8 feels right at home on macOS - I especially love our new translucent sidebar. That being said, this is still in an early access stage, so there are bound to be hiccups and UI issues that need to be resolved. Please let us know if you run into any problems or have suggestions on how we can improve. And thank you for being a long-time user!

Absolutely nothing about any decision AgileBits has made in the last 4 years has had anything to do with what customers (that's us, the people that used to give you money) want, and everything to do with nickle and diming the suckers dry.

UI consistency between different operating systems is NOT a user-focussed feature. When I'm on a Mac, I want my apps to behave like a Mac app. When I'm on Linux, I want my apps to behave like a Linux app. If you _actually_ believed that all apps should look and behave the same on any OS, why does the Android version look and behave nothing like the Mac app?

You've removed features with every major release, and this is just smashing the final nail into 1Password's coffin. You've ruined what used to be the best password manager on any platform.

So your margins are more important than your users’ native experiences. Got it.

I'm sorry for not being more clear earlier as to why we couldn't support two separate teams for the same platform. Hopefully this clears up any confusion.

You take money to provide software. But then you become lazy and greedy and want 1 size fits all. End result is your users having clunky, high latency experience.

Unfortunately, it’s normal in software development for multiple platforms to increase development complexity when feature and UX parity is prioritized.

Okay, I have a suggestion: drop Electron and keep supporting the native app you have.

> What gives me pause is how I write regularly asking for separate vaults for trivial passwords and passwords that could lead to financial ruin.

Just to clarify, what solution are you asking for? Do you want a local vault option to store sensitive passwords? Or something else?

Yes, because the implementation decision has implications for both performance and UX. I’ve used 1Password since version 3 (2013!) and gotten friends and family to do the same, but I think I’m done when 7 stops working.

That's worse, not better.

At least being forced to by investors makes sense. The current direction of travel being voluntary means you've just got a bad nose for building security.

Could you elaborate on this?

What's your experience building security systems? Do you have a LinkedIn profile or a CV we could review?

The hassle of doing what your users are paying you to do? Any child can hack a UI together in HTML but there's a reason no one (usually) pays for that.

I think a subscription business model is the only honest way to sell software that will require ongoing support. If you're comfortable with a snapshot w/o updates, then by all means buy once, but I think coming to terms with the demands of ongoing support also means coming to terms with continuing to support the product in some way.

That said, I wish there were more variations in the way to pay - a long term license with a high upfront fee and a low monthly, an immediate access option with a high montly and no up-front fee, etc.

BTW, I think they do have an option that you pay for 3 years upfront. At least, that was one of the options they mentioned when I complained about the lack of the option to buy a license. To me that did not seem an acceptable solution because you pay upfront and still have all the drawbacks of the subscription model such as being dependent on the trustworthiness of a quite obviously untrustworthy company. Add to that the removal of local vault option, and it becomes even less acceptable.

I know that it's very common for subscription models to coincide with forced upgrades (as this one does), but that seems like a choice on the part of the company as opposed to something inherent to the revenue model. I'd be quite happy to pay a developer to continue to maintain an older version of their software.

Like you pointed out, I think a fee structure where you pay for major updates and otherwise pay a maintenance / hosting subscription fee makes the most sense.

I think that you see open source projects that struggle along all of the time because their developers cannot afford to work on them enough. Not every project, but enough of them. I try to support projects like that too.

Done.

It seems like you'd be happier with a community product that has less support, but is available for free - and thankfully you have that option.

edit: apparently 8 will be electron-based. So... no standalone subscriptions AND they've moved from native to electron :/

And let's not forget that you're basically running a completely separate browser that can't re-use any memory from the other 5 separate browsers (=Electron apps) you need to run for work all day.

Can you go more into how non-native apps are a "hallmarks of rot by VCs"?

I hate them too, but my impulse is to blame MBA thinking (build once, less investment, who cares if it sucks) than VCs specifically.

They just like to say that they use Rust for the backend code. Rust for the backend of the client apps, React for the UI, wrapped in Electron.

Electron apps typically don’t work for me because they don’t integrate with the rest of the system cleanly, and so once you stray off the designers’ happy path it becomes clumsy to use the app. This isn’t an esthetic or ideological argument; simply for my usage an electron app can rarely be as convenient as a native app on the Mac. Things like input integration, system service integration, selection, and responsiveness are much harder to do when you are fighting the electron abstraction, so no wonder devs leave those things out.